Introduction - The Coz-E

This is the build log (and other relevant information) for my cozy-derived electric canard airplane.

You can think of this as a Cozy Mark IV with the rear seats replaced with a large battery compartment and an electric motor. Because that’s what it will be.

Unlike my digital garden, which provides more generic knowledge and things I’ve gathered, this site is intended to scope down to the Coz-E.

Using this site

Pages are organized on the sidebar to the left. It’s sorted alphabetically, and most top-level sections have nested subsections. Sections with the chevron next to them can be expanded to show additional subsections. You can also search for things by clicking or tapping on the search icon in the top left.

There is an rss feed to track updated articles here. You can also paste any article’s URL into your rss reader and it should discover the feed.

Recent Work

The pages that I’ve recently been working on.

• Chapter 7 - Fuselage Exterior (2023-11-28)
• Chapter 8 - Shoulder Brace, Safety Belt Mount (2023-11-13)
• Fiberglassing Techniques (2023-08-07)
• Chapter 6 - Fuselage Assembly (2023-04-12)
• Chapter 5 - Fuselage Sides (2022-12-14)
• Chapter 4 - Bulkheads, Instrument Panel, Front Seatback and Temporary Firewall (2022-12-06)
• Mods to Investigate After Flight Testing Phase 1 (2022-11-27)
• Materials Used (2022-09-09)
• Automated Epoxy Dispenser (2022-08-28)
• Introduction - The Coz-E (2022-08-25)

Last updated: 2022-10-30 11:22:17 -0700

Build Log

This section is meant to describe the build log for the aircraft, and will roughly reflect the chapters in the Cozy Mark IV plans.

Last updated: 2022-08-22 17:31:53 -0700

Chapter 3 - Education and Practice Layups

Chapter 1 lays out administrivia of building a plans-built aircraft, and is largely obsolete nowadays anyway. Chapter 2 lists out the materials needed, both in total, and by chapter, which is very helpful. Chapter 3 involves actual work, as it details how to actually build the plane, and includes 3 practice layups.

I only did the flat layup and confidence layup before I proceeded onto chapter 4. I forgot to order the foam for the bookend project, so I skipped it.

Flat Layup

The flat layup is a very simple layup of 6 12.5“ x 18“ pieces of BID. The goal is to internalize how to apply epoxy and fiberglass, and how to apply the correct amounts. This piece is then cut down to a 10“ x 16“ rectangle that should weigh between 10.5 ounces, and 11.5 ounces, with 11 ounces being near ideal.

I ended up doing 2 of the flat layups. The first, I ended up running out of mixed epoxy as I finished it, and as a result, the layup was too light (~9.7 ounces). You can see this in the many light spaces in the fiberglass, indicating both the lack of epoxy and what looks like a bunch of trapped air bubbles?

Because the first layup was obviously bad, I decided to do a second flat layup to get it right. For the second version, I mixed just over 8 ounces of epoxy, which is way more than I needed. I left some in the cup to cure, so that I could verify that it was properly mixed in the first place. This one also turned out to be too light (10.2 ounces), though it had significantly less light spots. Still, there was enough errors that if it were a real part, I’d be either rejecting or spending effort to repair it.

For the third layup, I left an electric space heater on in the area while it cured. I also utilized a sheet of 100% polyester as peel ply. Combined, this worked to produce a pretty good layup with very little noticeable defects. Downside, the polyester I used was a bit annoying to peel off, so I’ll find a different fabric to use for peel ply.

Confidence Layup

The confidence layup is a piece of urethane foam sandwiched between layers of BID and UNI. The goal is to learn how to shape urethane foam, learn how to apply fiberglass to foam, and overall gain confidence that the part, once cured, is incredibly strong. I used scrap foam that Aircraft Spruce included to be packing material.

I did two of these. In the first, I accidentally measured the BID incorrectly. You’re supposed to cut it at a 45° angle, with a width of 4 degrees. I instead measured as if I wasn’t cutting the fiberglass at a 45° angle - as if I was cutting straight into the roll of fiberglass - which resulted in the BID strips being too short, and part of the layer directly on top of the foam delaminated during curing.

I made sure not to repeat this mistake on the second one. I also utilized the polyester peel ply for this (I did this at the same time I did the third try on the flat layup). This, combined with the space heater from earlier to maintain curing temperatures, resulted in another pretty good and strong layup.

Last updated: 2022-03-27 09:15:44 -0700

Chapter 4 - Bulkheads, Instrument Panel, Front Seatback and Temporary Firewall

Chapter 4 starts off fairly simple - you’re working entirely with flat shapes that are all internal to the aircraft. The only curves are the edges.

Current status: Finished! ✅ 2022-12-05

Front Seatback

Status: Finished! ✅ 2021-10-23

The front seatback is very simple - cut down a slab of 0.75“ thick foam, shape two of the edges to be 45°, and glass it. Following the advice of other builders, I used a tablesaw to make those cuts. You can see the result below.

The foam for the front seatback prior to glassing.

Closeup of the 45° edge on the front seatback foam.

A few days later, I did the next step, which is to add micro and UNI fiberglass to one side. I interpreted “micro” in this case to be wet micro, not micro slurry. We’ll see how it turns out.

I cut out 2 lengths (approximately 42-45 inches) of full-width UNI and laid them out. I had to cut and rearrange the glass while laying it out, and there was some excess, but this worked well. Instead of using the polyester sheet as peel ply, like I had previously, I used plastic sheeting. We’ll see how it turns out.

The forward side of the front seatback, while curing after being glassed.

The forward side of the front seatback.

As you can see, it turned out fine. I did go back to using polyester peel ply for the aft side of the front seatback. Which also turned out fine and was relatively easy to pull out.

F-28 Bulkhead

Status: Finished! ✅ 2022-07-14

The F-28 Bulkhead is a relatively small piece, and will be made of fiberglass glassed onto clark foam. I cut the foam out using a utility knife and a Fein tool.

One side turned out better than the other. The first side I glassed, I put down a significantly thicker layer of wet micro than I should have. I determined that while that wasn’t enough of an issue to warrant remaking the part, and made a note to only place a thin layer of wet micro when I did the other side.

The image below is a composite image of both sides of the F-28 bulkhead, which shows the differences in the layup. Each side is labeled for convenience, both sides of F-28 received the same amount of fiberglass.

.

Instrument Panel

Status: Finished! ✅ 2022-09-18

The Instrument Panel is a little bit larger than the F-22 bulkhead. I made this by cutting pieces of it out of clark foam, and them gluing them together with 5-minute epoxy. Afterwards (much later in my case), I laid up both sides with 2 layers of BID everywhere, and then a layer of UNI running top span only.

I chose to attach and glass the stiffeners in 6 parts: First, I attached the horizontal stiffeners, which largely run across the upper span of the instrument panel. Second, I glassed the upper part of those stiffeners. Third, I glassed the lower part of those stiffeners. Which includes adding the “lip” to run wire channels and harnesses through. Fourth, I attached the vertical stiffeners, which run vertically along the center span of the instrument panel. Fifth, I glassed the inner parts of the vertical stiffeners (which don’t have a lip on them). Lastly, I glassed the outer part of the vertical stiffeners. I went with this order so that the fiberglass used for the lower-horizontal stiffener would already be in place when I attached the vertical stiffeners. Additionally, I didn’t want to deal with trying to add fiberglass underneath one of the “lips”. For all of this glassing, I used BID tape with either 1 or 2 plies of BID, as called out in the plans.

Note: I actually ended up using UNI instead of BID for this. Which while not according to spec, I don’t expect this to be an issue. I’m also expecting to make a slightly different instrument panel later on, to account for the glass cockpit I’m intending to build.

F-22 Bulkhead

Status: Finished! ✅ 2022-11-03

Tip: When adding the doubler, be sure to apply wet micro to both sides of the doubler. Micro is supposed to applied to every side of foam that touches glass.

I had built out one F-22 bulkhead, with the aft face looking pretty decent. However, I had made a number of errors glassing the forward side. These errors were such that I’d essentially have to strip most of the fiberglass from the forward side. Additionally, after discussing this with other people, I realized that I had added way too much wet micro to the foam. With both of those, I made the decision to remake the F-22 bulkhead entirely.

For the second go around, I had ordered more clark foam to use. Which meant I could be way more liberal with material usage as this high-density clark foam is not used after this chapter. This meant that I could minimize the amount of joins I needed to make, which made things a little bit easier.

Glassing the aft face went went. It took a while before I got around to trimming the excess fiberglass. Once I did, the next day I glassed the aft face.

Landing Gear Bulkheads

Status:

The landing gear attach bulkheads are the bulkheads used to attach the main landing gear to the rest of the airplane. When installed into the fuselage in Chapter 6, they’ll form a box with an open bottom. This bottom is where the mains will be inserted, and then clamped into. During installation, the forward and upper landing gear bulkheads will be joined together into one bulkhead, while the aft landing gear bulkhead will remain separated.

Note: The plans call the upper landing gear bulkhead and the forward landing gear bulkhead the same part. But they are made separately and then later joined. For that reason, as well as because “upper forward landing gear bulkhead” is a bit much, I’m going to refer to the lower portion of the forward landing gear bulkhead as the forward landing gear bulkhead. Similarly, the upper portion of the forward landing gear will be referred to as the upper landing gear bulkhead.

Tips:

1. When glassing the forward and aft bulkheads, add epoxy onto the hardpoints before adding glass. I found this significantly improves adhesion and also reduces air bubbles and other imperfections.
2. When you glass these bulkheads, it’s a really good idea to apply masking tape to opposite faces. Especially around the hardpoints. I found that epoxy could very easily get underneath the foam where the bulkheads are. Masking off the underside face will likely save you from having to sand off that epoxy.

For all three of these bulkheads, I initially misread the plans and made them out of high-density clark foam like the other bulkheads are. These should be made out of lower-density PVC foam, which I am working on remaking these with. Thankfully, I had not done much other than assemble the foam core of the bulkheads. Additionally, these are still useful as templates for making additional copies of the foam cores.

When I remade them with the correct foam, I was able to cut them in one piece - no epoxying with 5-minute epoxy required.

When joined in the fuselage, the landing gear bulkheads will form a box. The fuselage struts will be bolted on in between the hardpoints on either side.

Making the Hardpoints

I made 12 inch by 8 inch sheets of 22 ply BID fiberglass for the bulkhead hardpoints. Which is enough to make 8 hardpoints. Which has turned out to be a good mistake because I irreparably damaged 2 of them with my trim router while I was trying to trim them to their final shape.

Tip: If you’re using a handheld router, hold the router with both hands. Don’t try to hold it with one hand, and a vacuum hose (for dust collection) with the other. I was lucky and didn’t get hurt when this happened.

My second attempt was much more successful. I didn’t bother with the router at all, and cut the fiberglass with an oscillating tool. I used a belt sander for final shaping, and it came out quite well.

Forward and Upper Landing Gear Bulkheads

At this point, the forward and upper landing gear bulkheads should only be glassed with 2 layers of BID and a layer of UNI (horizontally). Later in Chapter 6, the forward and upper landing gear bulkheads will be joined together with some additional reinforcement layups. In my first version of the upper landing gear bulkhead, I made the mistake of adding the reinforcement layups while building the initial part. That and other mistakes (not applying the horizontal layer of UNI to the vertical sides of the part, the 45° edges delaminated from the foam during cure) led me to conclude that I should redo the upper landing gear bulkhead. Don’t make the same mistake I did, double check the layups you’ll be doing before doing them.

However, my forward landing gear bulkhead was built with the 3 layers as described in the plans. Unfortunately, I rejected this part after curing both sides because I accidentally used the wrong kind of fiberglass.

On 2022-09-11, I discovered that some replacement BID I had switched over to in August was actually UNI. This mistake was due to blindly trusting that the BID I had ordered was actually BID, without verifying that. Once I realized this mistake, I went through my build footage to check which parts were affected, and to my relief, it was really only the forward landing gear bulkhead (the other parts I had used the not-BID for were either rejected, or were the stiffener ribs on the instrument panel, which I don’t think will be an issue). I marked that part as not-for-flight, ordered new foam and fiberglass, and made a note to check that in the future the fiberglass I receive is actually I actually ordered. Once the new fiberglass came, I did verify that it was actually BID, and that the fibers were running as they should be.

The next forward landing gear bulkhead I made had what looked like too many air bubbles over the hardpoints. After trying to sand off enough fiberglass to repair that mistake, I realized it would be faster to just cut off the hardpoints and re-use them with a new foam core. Which is exactly what I did.

This third forward landing gear bulkhead I tried was successful. This is where I had the idea to try adding epoxy to the hardpoints before adding fiberglass, which worked amazingly well. Once both faces were glassed, cured, and trimmed, I sanded off the stray glass then drilled the quarter-inch holes into the forward bulkhead.

(The white spots here are artifacts left over on the very top layer of epoxy from pulling off the peel ply, not indications of air bubbles. Took me a while to realize that too. It means the layup was a little too wet.)

Upper Landing Gear Bulkhead

The upper landing gear bulkhead is made much like the lower landing gear bulkhead. It does not have hardpoints, but it still receives the same 2-BID, 1-UNI layup.

I made a mistakes here, which I didn’t discover until near the very end. The top and bottom 45° edges are not supposed to be glassed. I did glass them. Fortunately, because these edges didn’t turn out evenly, decided to sand them off, with the intention to reglass them. I should’ve stopped once I sanded those layers off, but I also thought the faces should still be glassed, so I also sanded off some nearby fiberglass, per the instructions for doing repairs. I have since re-glassed those areas, and am awaiting curing.

Additionally, after I glassed the upper face, while removing the peel ply, I took off some of the top layer of UNI. Which isn’t a big deal. Sand down the area to create a good surface for a mechanical bond, also sand the surrounding area a bit. And reglass with a layer of BID (to replace the layer of BID I damaged by sanding in to), and a layer of UNI. Just in the affected areas. Not too difficult.

Aft Landing Gear Bulkhead

Unlike the forward bulkheads, the aft landing gear bulkhead does receive side reinforcement layers at this stage. Additionally, the entire bulkhead is covered in 2 layers of BID and 2 layers of horizontal UNI. The reinforcement layers for the aft bulkhead are 8 layers of UNI (vertically skewed, aligned with the inside edges) on each outer edge of the forward side, and 3 layers of UNI (vertically skewed, aligned with the inside slant) on each outer edge of the aft side. Then at the end, you drill in 2 7/8“ holes near the top on the outer edges.

Firewall

Status: Temporary firewall complete. 3 out of 4 pieces of the permanent firewall are also complete.

Temporary firewall, cut from a large 4x8 sheet of 1/8 inch birch ply I got from a big box store. It’s terrible wood, but that doesn’t matter because it’s not going to fly. It’ll be used briefly in chapter 6 as a fit check before assembling the fuselage with the lower piece of the permanent firewall.

The permanent firewall is made out of very high quality baltic birch (with a price to match).

I delayed building the permanent firewall due to worrying about the blind screws. The way you’re supposed to secure the blind screws is essentially by applying a bunch of flox and hoping that the epoxy holds - has been known to cause a ton of issues, because the flox does not handle torque well. There are a couple possible solutions. One is to use hex bolts and chisel out a recess for the bolt heads. Another is to use clickbonds from cozygirrrl. A third is to wrap a metal thread around the screws. After some deliberation, I decided to go with hex bolts. I ordered a bunch of AN3-5A hex bolts from Aircraft Spruce (21/32“ nominal length, 1/4“ grip length, cadmium plated, undrilled), and will be finishing the firewall with them once they arrive.

After the hex bolts arrived, and not a whole lot of other excuses to put this off, I decided to start cut out the aluminum blocks for the engine mount. This was a pain, because I don’t really have metal cutting tools available to me. Thankfully, aluminum can be cut with woodworking tools - it just takes a while. I cut these 4 using a dremel and my oscillating tool. This was my first time using a dremel, and I’m just happy I didn’t explode a cutting wheel. If I were to build another firewall, I would 100% buy already-cut blocks of aluminum. Even if I have to oversize them, it would be worth it to me.

A few days after I had cut the motor mounts, I installed them in the firewall pieces and glassed the aft faces of all 4 pieces of the permanent firewall.

About a week after glassing the aft face, I used a dremel to route out recesses for the hex bolts. I installed the hex bolts, using flox to fill out the remaining space, and then glassed the forward faces of the firewall pieces.

The next day, I pulled off the peel ply. I lined up pieces for a final “yay, I’m done” photo, when I realized I had accidentally made 2 copies of the left-middle piece.

The following day, I cut out a wood core for a right-middle firewall piece. I marked where the bolt holes should go, then I marked which face is which. Then I checked that those markings were correct. Twice. Then I glassed the aft face of that firewall piece. A few days later, I drilled out the recesses for the bolts, installed the bolts, and laid up the forward face. The next day, I pull off the peel ply and cut off the excess fiberglass. Thus completing the firewall pieces and chapter 4.

Log

• 2021-10-09 - Cut front seatback foam panel to size.
• 2021-10-12 - Glassed forward side of front seatback.
• 2021-10-15 - Cut out corners from front seatback. Glassed aft side of front seatback.
• 2021-10-23 - Finished front seatback (cut rest of holes).
• 2021-10-27 - Cut foam shapes for F-28 & the F-22 doubler (initial).
• January 2022 - Cut foam for F-22 (initial) & Instrument Panel. Glued them together. Also cut & glued the wrong foam to make the landing gear bulkheads. Didn’t realize this mistake until June 2022.
• 2022-03-29 - Start cutting temporary firewall out of home depot 1/4 inch birch plywood. Which is fine, because this won’t go on the final aircraft.
• 2022-03-31 - Finished temporary firewall.
• 2022-04-05 - Cut rough outline for permanent firewall.
• 2022-05-14 - Laid up 2 of 4 fiberglass bricks for landing gear bulkhead hardpoints
• 2022-05-15 - Sanded down angles of F-22 doubler to the correct shape
• 2022-05-20 - Laid up 2 more fiberglass bricks for landing gear bulkhead hardpoints. You don’t necessarily need 4 bricks (you can easily get 2 hardpoints per brick. 3 if you’re really careful. At the time, I thought it was 1 brick per hardpoint). But this worked out for me because I made errors cutting the 2 of the hardpoints.
• 2022-05-22 - Laid up aft side of F-22 bulkhead (initial).
• 2022-05-25 - Trimmed excess fiberglass from aft side of F-22 (initial).
• 2022-05-28 - Laid up aft side of Instrument panel and one side of F-28.
• 2022-05-29 - Laid up forward side of F-28 bulkhead
• 2022-05-30 - Cut initial hardpoints for the forward landing gear bulkhead. Router slipped from my hand and both were damaged and in a to-me unrepairable state.
• 2022-05-31 - Trimmed rest of the two larger shapes of the permanent instrument panel.
• 2022-06-03 - With help of my wife’s labmates, re-cut the landing gear bulkheads. This time in the correct foam.
• 2022-07-09 - Trimmed fiberglass from aft side of Instrument Panel.
• 2022-07-10 - Re-cut hardpoints for the forward landing gear bulkhead. These ones came out much better.
• 2022-07-11 - Glassed other side of F-28.
• 2022-07-14 - Trimmed fiberglass off F-28. It’s now finished!
• 2022-07-19 - Glassed forward side of Instrument Panel.
• 2022-07-28 - Cut stiffeners for instrument panel. Also cut new foam for F-22.
• 2022-07-?? - Cut out side panels for permanent firewall.
• 2022-08-03 - Drilled holes for screws in permanent firewall (side panels & bottom).
• 2022-08-04 - Sanded down hardpoints for landing gear bulkheads.
• 2022-08-06 - Attached horizontal instrument panel stiffeners with 5-minute epoxy.
• 2022-08-10 - Glassed forward side of upper landing gear bulkhead (the one without hardpoints).
• 2022-08-23 - Pulled off pointers tape from instrument panel. Should’ve done this while the epoxy was still green. There’s still some small amount tape that’s under some cured epoxy. Because of the location, the best I think I can do is scrape at it with a knife.
• 2022-08-24 - Removed as much painters tape on the instrument panel as I could, time-boxing to 2 hours. Also cut fiberglass for glassing the stiffeners.
• 2022-08-25 - Cut fiberglass in preparation for glassing the aft side of the upper landing gear bulkhead (the one without hardpoints).
• 2022-08-26 - Beveled edges of upper landing gear bulkhead. Was going to glass the aft side of this bulkhead, but ran into an issue with the new revision of the epoxy dispenser board and had to work around it.
• 2022-08-27 - Glassed the aft side of the upper landing gear bulkhead. Also glassed the upper side of the horizontal stiffeners on the instrument panel.
• 2022-08-28 - Pulled off the peel ply from the upper landing gear bulkhead & horizontal stiffeners.
• 2022-08-29 - Epoxied (with 5 minute epoxy) the hardpoints to the forward landing gear bulkhead. (The aft landing gear bulkhead’s hardpoints still need a little bit of material trimmed from them).
• 2022-08-30 - Cut fiberglass for both fore and aft sides of the forward landing gear bulkhead.
• 2022-08-31 - Glassed one side of the forward landing gear bulkhead.
• 2022-09-01 - Glassed lower side of the horizontal stiffeners on the instrument panel. Also pulled off fiberglass from failed layup of the forward landing gear bulkhead.
• 2022-09-02 - Glassed the 45° edge on the upper end of the horizontal stretch on the aft side of the upper landing gear bulkhead.
• 2022-09-04 - Installed the vertical stiffeners on the instrument panel. Glassed the inner corners of these as well.
• 2022-09-05 - Installed hardpoints to the aft landing gear bulkhead. Re-glassed the forward landing gear bulkhead.
• 2022-09-06, 2022-09-08 - Cut fiberglass for the landing gear bulkheads.
• 2022-09-09 - Glassed the other side of the forward landing gear bulkhead. Glassed the exterior of the vertical stiffeners on the instrument panel.
• 2022-09-10 - Cut & shaped foam for a new upper landing gear bulkhead.
• 2022-09-11 - Realized I had accidentally been using UNI as BID, paused work as I figured out which parts were now useless and awaited for more actual BID to arrive. (Thankfully only need to redo the forward landing gear bulkhead)
• 2022-09-18 - Replacement foam for landing gear bulkhead arrived. Cut new foam for this. Also cut new hardpoint from last remaining hardpoint sheet. Also trimmed the last bit of fiberglass for the instrument panel, finishing it.
• 2022-09-22 - Glassed aft face of the forward landing gear bulkhead attempt number 2.
• 2022-09-23 - Glassed F-22 attempt number 2.
• 2022-10-02 - Glassed upper face of the upper landing gear bulkhead attempt number 2.
• 2022-10-08 - Glassed aft face of the aft landing gear bulkhead
• 2022-10-09 - Cut out and sanded foam off hardpoints from forward landing gear bulkhead attempt 2, to be used in attempt 3.
• 2022-10-12 - Rough trim of fiberglass off of the aft landing gear bulkhead. Will probably build a router table for the final trim.
• 2022-10-13 - Cut out new foam core of the forward landing gear bulkhead. Epoxied in place the hardpoints.
• 2022-10-14 - Glassed one side of the forward landing gear bulkhead.
• 2022-10-18 - Trimmed fiberglass off forward landing gear bulkhead. Did same to most of the upper landing gear bulkhead.
• 2022-10-19 - Sanded off epoxy that got on the unglassed sides of all of the landing gear bulkheads.
• 2022-10-22 - Glassed the other sides of the forward landing gear bulkhead and the upper landing gear bulkhead.
• 2022-10-23 - Cut fiberglass in preparation for glassing the forward face of the aft landing gear bulkhead, as well as the forward face of the F-22 bulkhead.
• 2022-10-24 - Glassed the forward face of the aft landing gear bulkhead.
• 2022-10-26 - Trimmed fiberglass on the forward and upper landing gear bulkheads.
• 2022-10-27 - Sanded off fiberglass from lower 45° angled side on the lower face of the upper landing gear bulkhead.
• 2022-10-28 - Trimmed fiberglass off F-22 bulkhead. Sanded excess micro & epoxy off F-22 bulkhead. Drilled 1/4 inch holes in hardpoints of forward landing gear bulkhead, finishing the part. Sanded off fiberglass from the upper 45° angled side pieces on the lower face of the upper landing gear bulkhead.
• 2022-10-29 - Glassed the forward face of the F-22 bulkhead.
• 2022-10-30 - Finished the Aft Landing Gear Attach Bulkhead (trimmed fiberglass & drilled holes). Pulled peel ply of of the F-22 bulkhead.
• 2022-11-01 - Re-glassed parts of the lower face of the Upper Landing Gear Attach Bulkhead.
• 2022-11-02 - Re-glassed parts of the upper face of the Upper Landing Gear Attach Bulkhead.
• 2022-11-03 - Trimmed excess fiberglass from F-22 and Upper Landing Gear Attach Bulkhead.
• 2022-11-16, 2022-11-17 - Cut out the 1“ by 1“ by 0.25“ aluminum blocks, to be used for the motor mounts on the firewall.
• 2022-11-18 - Alodined the aluminum motor mounts.
• 2022-11-21 - Glassed the aft face of the the 4 parts of the permanent firewall.
• 2022-11-28 - Installed hex bolts - a deliberate change from the plans. Glassed the forward faces of the permanent firewall.
• 2022-11-29 - Pulled off peel ply from the forward faces of the permanent firewall. Realized that I had accidentally made duplicates of the left-middle pieces in the firewall.
• 2022-11-30 - Cut new wood, marked where holes for bolts should be drilled, marked which side to glass first, and glassed the aft face of the new right-middle firewall piece.
• 2022-12-03 - Cut holes & drilled recesses for the bolts on the new right-middle firewall piece.
• 2022-12-04 - Fully installed bolts & laid up forward face of right-middle firewall piece.
• 2022-12-05 - Cut off excess fiberglass from right-middle firewall piece, completing chapter 4.

Last updated: 2022-12-05 20:39:19 -0800

Chapter 5 - Fuselage Sides

Chapter 4 was about making a bunch of bulkheads which are all internal to the fuselage structure. Chapter 5 involves making external parts - the fuselage sides.

In this chapter, only the interior faces of the fuselage sides will be glassed. The exterior faces will be glassed in chapter 7, when the fuselage is assembled.

Current Status: Finished! ✅ 2022-12-13

FJ* Jigs

Status: ✅ 2022-10-30

The FJ* jigs are used for shaping the longerons and providing a shape for the foam that will be epoxied. For the most part, these are cut from 1 by 8 (nominal) pine board. For the FJA jig, with the curve over the length of it, I chose to 3d print a template for it, which I’ll use to copy onto pine board using a trim bit in my router. I also 3d printed the FJD jig, but as I was cutting FJB, I realized that it wasn’t that much more effort to also create a pine version of FJD as well. Which works better because it’s easier to drill in nails to wood.

Fuselage Side Forms

Status: ✅ 2022-11-08

The fuselage sides need a form to rest on, otherwise they’ll sag and not be the correct side. Per plans, I got a large sheet of masonite, and used it to create forms for the 2 fuselage sides. I did this by ripping it into 2x 21 inch wide by 8 foot long sheets, keeping the scrap. The scrap was used to extend the width of these sheets to the necessary 102.5 inches. Note: I accidentally cut my strips of scrap too short, by about an inch or so. I don’t think it’ll make that much of a difference (and if it will, I’ll add on that extra inch or so of material).

Once the forms were at their full length, I aligned them on top of each other and drew out the points for the shape. Then connected the points with straight lines, then tried my best to cut it using a jigsaw. This produced two forms of the exact same shape, which I’ll use to help shape the fuselage side foam.

Upper Longerons

Status: ✅ 2022-11-08

Tips:

• Place the forward doubler about 0.5“ back from where the plans indicate. This prevents a canard placement issue that many people report.

The upper longerons (one for each side) are made of 3 long strips of 1“ by 0.25“ by 105“ sitka spruce. Along with a stiffener and 2 doublers, also made of spruce. Because these will go on top of the foam for the fuselage sides, they need to have a slightly tighter curve than the fuselage sides. This is done by laying FJA, FJB, and FJC on their sides, with the ends of FJA and FJC being 0.5“ apart, but angled so that they are adjacent by the time they meet FJB. Once you have the jigs set up, you’re supposed to join the 105“ spruce by painting the interiors, then use clamps & nails to hold the wood to the jigs. I was able to get away using just clamps.

Once the epoxy cured, I added the doublers and stiffeners. Per plans, I pre-drilled holes for nails on the forward doublers and the stiffeners. Then, once the wet flox was mixed up, I painted it on to the forward and aft doublers. For the center stiffeners, I found it easier to paint the flox onto the spruce strips, not the stiffeners. I also added regular flox into the saw holes, to try to counteract the fact that I made way more saw holes than I needed to. Once everything was epoxied and in place, I re-clamped everything, then added weights.

You’re supposed to let this whole thing cure for at least a 24 hours. I let it cure for more than that. This is because when I initially installed the doublers, I mistakenly placed the aft doublers in the wrong orientation (I had it aligned such that the taper was aligned with the curve of the longeron. You’re supposed to have that taper be facing up). I didn’t realize this error until I was preparing to go to bed, ~4 hours later. I got up and re-installed the aft doublers in the correct position (and checked with the plans multiple times while I was doing so - didn’t want to make an error because I was tired). Everything worked out, but it did mean that it would be way more convenient to just let it cure an additional night (and work on something else for once).

Fuselage Sides

Status: ✅ 2022-11-14

The plans state to hot glue the fuselage jigs to your workbench. Which is what I started to do, except I ran into issues both with my workbench not being sufficiently level (one of the main downsides of wanting to create a large workbench from several smaller workbenches - leveling them is nearly impossible), and with the jigs themselves having twist and otherwise not being perfectly straight. So instead, I made a series of spacers that I glued & screwed into place to keep the jigs the correct distance apart. This forms kind of the skeleton of a box, and is why I’m calling it a jig box.

Once you’ve done that, you’re supposed to place down the fuselage forms, and nail them in place. Per Wayne Hicks, you should only nail to the inside corner to ensure that the forms have entirely convex curves.

Ideally, prior to this point, you should have cut & joined your foam. Which I did not do. Instead, I placed the foam on the forms, then I used a sharpie to trace the outside curve (the non-straight edge). After, I cut the foam along traced lines, then sanded it smooth. Which worked, but was annoying. Afterwards, I then joined the foam panels together in-place on the forms. Which, again. I should have done that on a flat surface, but I didn’t really have one available. It worked out, but just took longer and more effort than it should have been.

Once you have the foam put together, you’re supposed to secure it directly to the forms using 5-minute epoxy. People have had issue with the foam tearing upon removal. I thought I’d be clever by applying painters tape to both the form and the foam, then applying 5-minute epoxy between the painters tape. Except the painters tape did not actually adhere to the foam. The next day, I came back and used masking tape, which worked immensely better.

Fuselage Side Spacers & Control Stick Depressions

Status: ✅ 2022-11-20

Once I had the foam secured, I started cutting the side spacers. These spacers are 3/4“ low-density last-a-foam that are supposed to to be micro’d to the top, bottom and aft edges of each side panel. Unlike the high-density last-a-foam used in F-22, F-28 and the instrument panel from chapter 4, this low-density stuff cuts very easily with a utility knife.

I cut out the spacers for the top and bottom, then 3d printed jigs so I could easily cut them on my bandsaw at the angles described in the plans. Once these had been cut, I installed them to the fuselage sides with wet micro (albeit, it was on the dry side of wet micro). As per plans, I used finishing nails to help hold them down + maintain shape.

While the top and bottom spacers cured, I went about figuring out the dimensions for the aft spacers. This was mostly done by eyeballing it, along with what little precise dimensions are given by the plans. It’s good enough. Again, I installed these with wet micro. While they cured, I used a router to add the depressions for the control sticks. For a gas-powered cozy, this is also where you’d create the depressions for the fuel sight gauges. I’m building an electric cozy. I don’t need fuel sight gauges. While I could take off a few grams by creating these depressions, it’s not going to make a difference one way or another.

Fuselage Side Interior Layup

Status: ✅ 2022-11-26

The Interior Fuselage Side Layup is the first large layup you do. This is glassing the interior faces of the entire fuselage sides. It’s a source of a lot of trouble for many people. I came away relatively unscathed.

A few days later after installing the spacers, my wife was available to help glass the fuselage side interiors. You could probably do this layup without any help, but her help made this immensely easier and much faster. Even with her help, this took close to 6 hours to complete. The first task was to coat all of the foam with micro, which took about 2 hours on its own. Then, with my wife’s help, we laid down the UNI fiberglass on the foam. You’re supposed to add 2 layers of UNI, at 30° fiber orientation from horizontal, with one layer oriented mostly forward to aft, and the other oriented in a forward-facing direction. Once the UNI was laid down, we massaged it into place. It’s really important to get the fibers to conform to all of the curves and corners, which was a little bit of a pain, but we managed to do it with minimal wrinkles. Once that was wetted out, we laid down the second layer of UNI and wet it out. Again, we had some minor difficulties with handling the complex curves, but we got it. Once this layer was fully wetted out, we grabbed a flashlight and double checked the layup for any potential issues. We came across a few, but took care of them. Then, I grabbed the upper longerons, and my wife painted the mating faces with wet flox while I held them. I placed them onto the fuselage sides then clamped them down. Then, we put down peel ply everywhere, and let it sit.

After letting it cure for an hour, I pulled off the peel ply to discover the results of my work. I was very pleased with the result. Yes, there were bubbles, but they were all repairable. Most were repairable with a simple injection of epoxy, which was incredibly satisfying to do. However, a few required me to sand off the fiberglass and re-apply it. In some of these cases, I ended up sanding off too much of the foam. For those cases, I applied dry micro as a filler to get to the desired shape. In the end, I’m very happy with the results.

Glassing the Upper Longerons

Status: Awaiting Cure

Once the fuselage side layup was finished and repaired, I moved on to glassing the upper longerons. Frankly, this was my least favorite layup I’ve done so far. It’s a large layup, but the overall section widths were very small, as well as some vertical sections. Which meant that I couldn’t really use a squeegee to spread epoxy. So I had to stipple pretty much the entire layup. I squeegeed as much as I could, to both improve the speed as well as to remove air bubbles and voids. But still, this was not a fun layup.

Because the upper longeron forward of F-28 will be getting cut off in a future chapter, I did not bother glassing forward of the forward doubler.

This layup turned out horribly. Bubbles galore and even some delaminations. Most of the issues were along the wood longerons, and only really the first 2 or 3 layers deep. Still frustrating. After spending a few days dreading this, I sanded off most of the fiberglass. This took 2 days. The day after that, I re-glassed the upper longerons. This second layup turned out significantly better - I only identified 3 issues, and then 2 more that actually looked like non-issues when I re-examined while repairing the other 3. I decided to let the epoxy finish curing for another day before repairing.

Two days after that, I performed the second repair of the upper longeron.

The Lower Longerons

Measuring & Marking the Lower Longerons

The plans are slightly unclear on how you’re supposed to make sure the lower longerons are correctly installed. The important thing is that the lower longerons need to be installed such that each side exactly matches - each lower longeron needs to be exactly the same distance from the corresponding upper longeron. They should be close to the numbers specified in figure 5, but since it’s impossible to actually get that correct (considering you’re now measuring on a curved surface, when figure 5 is for a flat surface), the distances just need to be close to what figure 5 specifies. I measured the distance with a carpenter’s square, using wood spacers to avoid any error from measuring a vertical distance. I recorded the distances at 10 inch intervals, and then computed their differences. After, I marked the insets on the relevant sides, where relevant. Then I prepared for glassing. Here are the values I measured, along with what figure 5 specifies:

So, they’re fairly close to each other, and relatively close to the values specified in figure 5.

Installing the lower longerons

The plans state to add saw cuts on the inside every 4 inches, then every 2 inches where the curvature is sharper. A bunch of other builders have reported better success with doubling that - making cuts every 2 inches and every inch in sharper curves. Which I also did. Additionally, I made these cuts using a thin blade on my oscillating tool, so I feel that the more-frequent cuts will better help with this curve.

To aid with clamping the lower longerons to the fuselage sides, most builders make wood blocks with a convex inner 45° corner to match the triangular lower longeron piece. I decided that the best route for me was to instead model these jigs in CAD and 3d print them. I modeled 1 up, printed a test print to verify it works. After checking that I had the dimensions correct, I printed a bunch more of them.

In actual usage, these ended up being a bit too long, and I ended up cutting off some of the tabs with an oscillating tool. Afterwards I redesigned the jig to remove that extra tab. For those interested, there’s an STL with this updated design.

• STL of the above rendering

With these clamp jigs, I installed the lower longerons. I didn’t want to bother with using nails to hold the wood in place, so instead I used clamps with the clamp jigs. I didn’t have enough clamps to install both triangular lower longerons at once, so I installed them on two separate days.

After both port and starboard lower longerons were installed, I added the longeron doublers - a 20“ long piece of triangular spruce which joins both lower longerons at the aft end along the hypotenuse side. I also measured and installed the LWX and LWY stringers. These are used to provide additional stability and better anchor the landing gear to the rest of the airplane.

Completing the Fuselage Sides

With the lower longerons and LW* stringers installed, the next step is to fill the area between the LW* stringers and the lower longerons with foam, and leaving room for the electrical channel (which will be covered with 1 ply of BID). I made the channel covers using 2 molds made out of 0.75 inch low-density clark foam. I built these molds and channel covers much earlier - before I had done the first glassing of the fuselage sides. These foam pieces serve as a mold. I wrapped them in packing tape and then glassed them with 1 play of BID. Once the stringers were in place, I installed the covers with 5-minute epoxy. Then I cut out foam to fill the area out of the same 0.75 inch low-density clark foam that I had used for the mold. I even re-used parts from the molds to fill in the area above the sloped parts of the covers. I then installed this foam with micro, and filled the voids with dry micro.

A day later, I came back and glassed that area with 6 plies of BID. Which turned out great.

With the glasswork finished, there are 2 things left for chapter 5: Cut the fuselage sides to length, and cut out the area for the main spar.

When I measured my fuselage sides, I found that they were both too short. One measured 101.5“, the other 101.???“. The plans state these must be exactly 101.75”, and, after panicking a bit, I checked the mailing list. According to discussion in the 1998 archives page, it’s fine that they’re short.

Nat has in person told me twice about builders with the 101.75 a little too short and said it does not matter. Just make sure that everything is square and it will be fine.

If you put the upper longerons where they are supposed to be and nail them as instructed all you have to do is spread the bottom a little. The jigs for the sides on the table aren’t that exact. Nat told me that the long EZ sides were built flat and bent to fit so that also isn’t an exact science. He just has us build them prebent to make assembly easier.

I talked to Nat a long time about this stuff one day and came to the conclusion that I was trying to build to tolerances that are not in the plans and don’t matter. If you have already built these things I guess that it is obvious. It wasn’t obvious to me, I am new to this. That is what we pay Nat the big bucks for though. He was very gracious and helpful despite all my questions that must have seemed silly to him.

After calming down from that, I felt confident continuing. I squared off the forward end of each side, and then cut out the spaces for the main spar. Thus completing chapter 5.

Log

• 2022-06-05, 2022-06-06 - 3D printed FJD jigs & FJA template jig.
• 2022-09-03 - Made FJB, FJC, and FJD jigs. Made most of FJE, but need to trim a little bit more wood to get it to the final shape. Made rough cuts for what will eventually be the FJA jigs.
• 2022-09-04 - A little further progress on one of the FJA jigs.
• 2022-10-15 - Cut one of the FJA jigs to final & correct size. 3 left!
• 2022-10-29 - Ripped a 4x8 sheet of masonite into 2x 21 inch by 8 foot sheets.
• 2022-10-30 - Finished the FJ* jigs.
• 2022-11-04 - Assembled & leveled tables. Installed jigs for eventually creating the upper longerons.
• 2022-11-05 - Epoxied the 105“ spruce strips together and clamped them to the jigs.
• 2022-11-06 - Shaped doublers & stiffeners. Installed them into place. Upper longerons nearly complete!
• 2022-11-08 - Unclamped upper longerons. Removed screws from stiffener. Upper longerons complete! Shaped to both fuselage side forms to the correct shape, finishing them!
• 2022-11-09 through 2022-11-13 - Constructed jig box for fuselage sides, nailed fuselage forms to the jig box, and cut and joined foam for the fuselage sides.
• 2022-11-14 - Started on the fuselage side spacers.
• 2022-11-15 - Cut out the foam that will be the upper and lower side spacers.
• 2022-11-17 - Cut to shape the upper and lower side spacers.
• 2022-11-18 - Installed the upper and lower side spacers.
• 2022-11-19 - Cut and shaped the aft side spacers.
• 2022-11-20 - Installed the aft side spacers, contoured the fuselage sides (added depressions for the control sticks)
• 2022-11-21 - Glassed the electrical channel plugs.
• 2022-11-23 - Glassed the interiors of the fuselage sides, installed the upper longerons.
• 2022-11-24 - Pulled off peel ply from the fuselage sides, inspecting the results.
• 2022-11-25 - Repaired interior layups of the fuselage sides.
• 2022-11-26 - Glassed the upper longerons - which ultimately failed.
• 2022-12-02 - Sanded off glass from failed upper longeron layup.
• 2022-12-03 - Sanded off glass from failed upper longeron layup.
• 2022-12-04 - Redid the upper longeron layup.
• 2022-12-06 - Repaired upper longeron layup (repair part 2). Added initial saw cuts through entire triangular pieces of lower longerons.
• 2022-12-07 - Some additional, final repairs of the interior fuselage side layup that I hadn’t noticed until now. Additionally, measured and marked locations for lower longerons.
• 2022-12-08 - Installed the lower longeron on the port fuselage side.
• 2022-12-09 - Installed the lower longeron on the starboard fuselage side.
• 2022-12-10 - Installed the longeron doublers as well as the LWX and LWY stringers on both fuselage sides.
• 2022-12-11 - Installed electrical cover & filled area between the stringers on the lower longerons with foam.
• 2022-12-12 - Glassed the area between the LWX and LWY stringers on both fuselage sides with 6 plies of BID.
• 2022-12-13 - Trimmed fiberglass from area between LW* stringers and the lower longerons, squared off forward edges, cut out space for main spar.

Last updated: 2022-12-14 12:11:19 -0800

Chapter 6 - Fuselage Assembly

Rejoice! At the end of this, you’ll actually have something to show to people instead of just mostly flat panels.

Now that the fuselage bulkheads and sides have been built, it’s time to put them together.

Current status (Late February, 2023): Fuselage and bulkheads mostly assembled. Center keel is assembled and waiting for install. Preparing to install the floor.

Fuselage Side Assembly

I decided to assemble my fuselage upside-down, as suggested and made popular by Wayne Hicks.

I struggled a lot with building the platform to do the assembly on. It’s incredibly important that this platform be level and square and I had extreme difficulty figuring out a wood structure that would suffice. My existing workbenches, which worked fine for holding the fuselage form jigs, would not work for this. I decided to purchase some T-slot aluminum extrusion and construct a work surface with that. The extrusion is machined to much higher tolerances than what I can do for wood. Additionally, for the fuselage assembly, I want to have the surface be low to the ground - my existing tables are not height-adjustable. The T-slot aluminum, while difficult to adjust the height (I can’t do it with any load or surface on it), is at least possible to lower.

Once the table frame was put together & leveled, I installed the table tops. I re-used the tops from my existing tables. They’re flat enough for this work - my real issue is that leveling 5 tables is incredibly hard to do. With the tops installed, including a space for the instrument panel, I used a laser level to draw a centerline across the entire table. I then drew lines marking where each bulkhead would be installed. I screwed in a piece of wood at the location where the top of the front seatback will be. Finally, I clamped a piece of wood to mark where the instrument panel should be installed.

Trial Fit

With the table put together and set to the desired height, I worked on attaching the temporary firewall. First, though, I added some wood beams to stiffen it and help it keep square. I attached it to the tabletop frame using 2 m5 bolts into T-slot nuts, then screwed in a brace to keep it square with the tabletop. Once it was sufficiently attached, I tried slotting in the fuselage sides. As expected, these didn’t fit. I found that a utility knife worked well to help carve out extra space from the thin and light wood I used for the temporary firewall.

Once I got both fuselage slides fully slotted in, I worked on the other end. I marked the centerline of F-22, lined that up with the centerline on the table. This allowed me to line up F-22 with the end of the fuselage sides. I found I needed to trim off material from forward end of the fuselage sides, which I did now. With that done, I drilled holes for screws to aid with re-assembling F-22, and also drew lines to mark where the fuselage sides end up when F-22 will be installed.

With the forward ends of the fuselage sides taken care of, I worked on getting the front seatback in place. This required me to trim some additional material from the front seatback to get it to fit. Which was disappointing, but not surprising. I tried my best to not trim more foam than necessary, but that proved impossible. In assembly, I filled the gaps with flox.

Once the front seatback was fitted, I worked on getting the instrument panel in. Which also required some trimming, but it was more around making the edges flat, and removing material from the horizontal stiffeners/cable channels. In other words, it went much quicker.

With the instrument panel fitted, I again placed F-22 on the forward end. This time, I drilled holes in the bottom longerons for screws to hold F-22 in. This provided full assurance that the bulkheads were in place. I marked on the fuselage sides where all these bulkheads would be installed, then I disassembled this and prepared for the full assembly (the next day).

Actual Assembly

With the trial fit complete, now was time to start on actual assembly. I mixed up a bunch of flox (2 batches ended up being necessary - first batch used 4 oz of epoxy, the second used 2 oz of epoxy), and applied flox to the mating surfaces of the fuselage sides and the bulkheads.

I started with the front seatback: the widest bulkhead, and in some ways, the most difficult. I used some drywall anchors to hold it in. Which worked well enough, but weren’t great. In hindsight, I probably shouldn’t have done this. Once everything was cured, I ended up stripping one of the drywall anchors upon removal. So that’s in there permanently. Thankfully, it’s lightweight plastic, so I didn’t take that much of a hit. I also used lengths of twine twisted taut between two pieces of wood to also clamp these together. Which worked pretty great. Probably why it’s recommended by the plans.

Next up was the instrument panel. This was much easier to install, because it’s just vertical. No issues. Definitely too thin to use drywall anchors, so I didn’t bother.

The last of the initial bulkheads to install is F-22. F-22 fits in front of the fuselage sides, with the doubler fitting between them. I used screws to hold them in, and also more of the twine-based clamps.

I let this cure for about 36 hours, to ensure it was solid.

Once it was cured, I sanded down the flox and prepared for the reinforcement layups. These are 2-inch wide strips of 2-plies of BID. Make these using the BID taping technique. These get applied along the entire length of the seams between each bulkhead and the fuselage sides. F-22 does not get tapes to the forward side. However, it also receives 4-inch wide strips of 4-ply BID where the doubler meets the fuselage sides.

Once I taped as much as I could of these panels, I installed F-28. Similar to F-22, it only receives reinforcement on the aft face. That’s because in the next chapter, I’ll be removing foam and upper longerons immediately in front of F-28, so there’s no point adding reinforcement layups there.

Landing Gear Box

After securing as much as needed in the forward half of the fuselage, I moved aft, to the landing gear box. First, I installed the aft landing gear bulkhead. At 5 inches forward of the firewall. I had to trim it down slightly to fit, but I got it in there. After curing and adding reinforcement layups, I installed the lower part of the forward landing gear bulkhead. I hot-glued a box 8-inches long to the bulkhead, which provided exact spacing, as well as helped guide the drill so I could match-drill a 0.25 inch hole from the forward landing gear bulkhead to the aft landing gear bulkhead. This hole was drilled once the flox cured. After drilling the hole, I used a rubber mallet to knock out the wood. I then added reinforcement layups

Center Keel and Seatback Brace

The center keel adds strength to the fuselage, and contains the hot air duct (likely unused for me, but that duct is still a structural component, so I’m keeping it). It’s the base of the center console and is part of the mounting point for the safety harnesses. Additionally, the fuel selector valve nominally in the center keel, but I don’t need to install that, so I’m not going to work on that.

You start the center keel by first microing and glassing foam, then cutting it out. I marked the shapes in black sharpie on the foam, then I micro’d and glassed it. The triangular pieces are received 2 layers of BID, any orientation. I chose diagonal with respect to the base. The rectangular pieces received 2 layers of UNI, oriented along the length of the piece.

The Fuselage Bottom

With the sides assembled, now it’s time to install a floor.

Log

• 2022-12-14: Started work.
• 2022-12-15 through 2022-12-16: Prepared shop for fuselage assembly - tore down the fuselage forms jig - building platform to assemble fuselage on.
• 2022-12-17: Ordered parts for constructing aluminum workbench, started work on the center keel.
• 2022-12-18: Glassed inside fiberglass parts of the center keel.
• 2023-01-08: Attached temporary firewall to assembly table. Trimmed temporary firewall to allow fuselage sides to slot in. Trimmed forward ends of fuselage sides to allow correct positioning of F-22. Installed screw holes for F-22. Started work fitting the front seatback in to the trial assembly.
• 2023-01-18: Finished trial fit of front seatback, instrument panel, and F-22 with the fuselage sides.
• 2023-01-19: Installed the front seatback, instrument panel, and F-22.
• 2023-01-21: Sanded off flox from installed bulkheads, prepared for reinforcement layups.
• 2023-01-23: Added reinforcement layups to joints between front seatback (aft joint only), instrument panel (aft side, plus below the electrical channel on the forward side), and F-22 (aft side + the 4-ply reinforcements to the doubler).
• 2023-01-24: Trimmed reinforcement layups. Prepared to install aft landing gear bulkhead.
• 2023-01-25: Installed aft landing gear bulkhead. Prepared lower part of forward landing gear bulkhead.
• 2023-01-26: Installed forward landing gear bulkhead.
• 2023-01-27: Drilled matching holes from forward landing gear bulkhead into aft landing gear bulkhead. Installed the lower piece of the permanent firewall.
• 2023-01-29: Added reinforcement layups to the landing gear attach bulkheads and F-28.
• 2023-02-01: Rotated the fuselage to be right-side-up.
• 2023-02-03: Installed the upper piece of the forward landing gear bulkhead. Added reinforcement layups to the forward side of the front seatback.
• 2023-02-04: Added layups to join the 2 parts of the forward landing gear bulkhead on the forward side. There were issues in this layup discovered after cure.
• 2023-02-07: Prepared forward landing gear bulkhead for repair.
• 2023-02-08: Moved fuselage onto sawhorses, disassembled table used for fuselage assembly.
• 2023-02-09: Added repair layups to forward landing gear bulkhead.
• 2023-02-10: Rotated fuselage back to upside-down.
• 2023-02-11: Added reinforcement/joining layups to the aft side of the forward landing gear bulkhead. Joined & glassed internal pieces of the seatback brace. Floxed them to one side of the seatback brace.
• 2023-02-12: Joined pieces for the conduit duct.
• 2023-02-13: Floxed other side of seatback brace in place. Floxed safety harness in place in the conduit duct.
• 2023-02-16: Laid up 7-ply UNI layup over the safety harness mounting point in the conduit duct.
• 2023-02-17: Glassed the outside edges of the front seatback.
• 2023-02-18: Glassed the outside of the conduit duct.
• 2023-02-20: Removed the firewall & prepared for reinstall. Joined the heat duct and the front seatback.
• 2023-02-21: Test fit center keel, determined need for an extension of the heat duct.
• 2023-02-22, 2023-02-23: Added foam as spacers for the firewall install. Sanded down to the exact length needed.
• 2023-02-24: Reinstalled firewall. It’s now square. Created inside layup for the heat duct extension.
• 2023-02-25: Joined heat duct extension pieces with flox, glassed exterior.
• 2023-02-26: Installed center keel, along with heat duct extension.
• 2023-02-27: Started work on floor. Cut panels to shape, joined with 5-minute epoxy.
• 2023-03-01: Built frame to support floor during contour and layup. Cut out landing brake. Marked where contour pieces will go.
• 2023-03-02: Cut out contouring foam for floor, fit in place.
• 2023-03-04: Epoxied contouring foam for floor into place.
• 2023-03-05: Glassed ~3/4 of the floor. Stopped after running out of fiberglass.
• 2023-03-09 through 11: Sanded floor in prep for finishing glassing.
• 2023-03-12: Glassed remainder of floor.
• 2023-03-18: Installed floor onto fuselage.
• 2023-03-27: Installed BID strips on starboard side (part 1 - the strips over the lower longerons were too short)
• 2023-03-29: Installed BID strips on port side
• 2023-04-01: Installed BID strips on starboard side, finishing up the fuselage assembly.

Last updated: 2023-08-12 07:19:09 -0700

Chapter 7 - Fuselage Exterior

In chapter 7, we take the assembled fuselage built in chapter 6, and do most of the exterior work. This is adding the air intake scoop at the rear, contouring it to have those nice, sleek curves, and glassing everything. At the end of this, the fuselage structure should be at its full strength.

Air Intake Scoop

Contouring Floor

One of the last things to do is to create these 1/16th inch depressions around F22, and around the landing brake. To make these, I set my calipers to 1/16th of an inch, and pressed the depth finder gauge into the foam, creating depressions that were close enough to correct. Then I sanded into the foam until the depressions were gone. These depressions only exist to enable overlap with future parts without creating a bump, so as long as they’re close enough to correct, it’s good enough. Highly recommend this.

Also during this, you have to tape a skirt 1 inch around the landing brake, which will be used later when we pull off the landing brake and and finish making it.

Antennae

Because fiberglass is radio transparent, I can embed the antennae underneath the skin. Giving a slight aerodynamic improvement and a massive aesthetic improvement. With some exception, all of the antennae I’ll be installing are simple dipoles, which can be made out of copper tape. The ones here, definitely so.

I spent some time on the antennae. Initially, I was thinking of placing 3 separate antennae here. A glideslope antenna, a nav antenna, and a marker beacon antenna. After discussing with people, I decided against the marker beacon, and also to combine the nav and glideslope into one. Some people also place a transponder antenna in the floor, which I decided against. I think I’ll place them in the far edge of the strakes, to minimize potential interference with the other antennae.

By the way, when testing antennae layout, I highly recommend using strips of tape to do that. Much cheaper if you decide to move things around.

Once I had the antenna layout figured out, I routed a depression for the copper tape, as well as a path for the coax cable. Then I installed the copper tape, soldered up the antenna, slid some ferrites on, and potted everything.

I did not do any actual testing to check the SWR of this antenna, so the length is based off theoretical numbers - speed of light in a vacuum, with no interference. Which means this is almost certainly too long. However, even if I had used an antenna tester, the test wouldn’t be valid until I finish installing the batteries and other sources of interference.

Glassing Floor

With the antenna installed, I hade no more excuses. Time to glass the floor. The floor is intended to be 2 layers of UNI fiberglass, 60 degrees rotated from each other. One layer rotated 30 degrees one way from the centerline, the other rotated 30 degrees the other way from centerline.

I unfortunately had to glass the floor in multiple parts. After applying the first layer of glass, we ran out of epoxy hardener. We only had enough hardener to glass about a third of the second layer of glass. We peel plied everything and ordered more epoxy.

After inspecting this layup, I noticed a number of errors, and I spent quite some time sanding off fiberglass to repair this. Then I finished the glasswork… let’s say more carefully. I glassed the section between the aft landing gear attach bulkhead and the firewall. Including 3 reinforcement layers over the landing gear attach point and the motor mounts. Then my wife and I did some repair work on the forward third of the fuselage. Then, my wife and I basically redid the middle half or so of the fuselage. Except we still had to do repair work on that, and then I had to repair the repairs. Then I was finished.

fuselage sides.

With the floor glassed, the next step is to contour the fuselage sides.

Canard Cutout

Part one of contouring the fuselage sides is making the canard cut-out. The cut-out line vertically is at W.L. 18.9, with the top of the longerons at W.L. 23.0, that means flush with the forward face of F-28, we’re supposed to cut out 2 4.1“ by 6.25“ squares from the longerons, foam, and the top “ears” of F-22. One on either side. I drew lines for this on the foam with pencil, then used the oscillating multitool to cut them out.

Making the fuselage contour templates

Next up is making the templates for the contours. These are 4 templates that represent what the fuselage sides should look like at F-28, F.S. 33, F.S. 38, and at the firewall, with the actual contour of the fuselage smoothly transitioning between them. I printed the 4, glued sandpaper to them as sanding sticks. For the sections where I’m supposed to smoothly transition to the next, I just eye-balled it. It turned out pretty good.

Anyway, I created out the A, B, C, D templates in CAD and created the gcode for them. The next day, I actually printed them. This is because the job would take 9 hours and I’m not at all comfortable running my 3d printer overnight. I’ve never had an issue in the 6 years I’ve owned that printer, but it just feels wrong.

Contouring the Fuselage Sides

With these printed, and the sandpaper attached, I spent a few hours sanding down the upper fuselage sides/upper longerons to the correct contour. Most of the work was done with a belt sander, but I frequently paused to check with the appropriate contour checker. Once it was close enough, I used the contour checker to finalize that sanding.

Glassing the Port Side

Once the fuselage sides were contoured, I rotated the fuselage so that the port side was facing up, and I glassed that. It took me about 6 hours to do so, and thankfully I managed to get it all in one shot.

Glassing the Starboard Side

While the port side was still curing, I prepped the fiberglass for the starboard side. I rotated the fuselage so that the starboard side was facing up, and glassed it. It took slightly less time to glass than the port side, but still close to 6 hours. This layup also came out great, thankfully.

Finishing

While the starboard side cured, I trimmed the glass for the port side. Once the starboard side cured, I trimmed the glass on the starboard side. Once both sides were trimmed, chapter 7 was finished!

Log

• 2023-04-06: Made paper template for air intake scoop.
• 2023-04-08: Traced air intake scoop template onto fuselage structure
• 2023-04-09: Installed 1“ and 2“ blocks of urethane foam to form basis for air intake scoop forward of the forward landing gear attach bulkhead.
• 2023-04-10: Built birch reinforcement triangles for landing gear box. Started on birch reinforcement blocks for landing gear box -> firewall.
• 2023-04-11: Attempted install of birch reinforcement blocks, but installed support blocks in incorrect locations… somehow? Would have resulted in a skewed install of reinforcement block. Spent ~2 hours removing hot glue after pulling off support blocks.
• 2023-04-14: Finished removing hot glue from aft landing gear bulkhead and lower firewall. Re-shaped support blocks such that the D parts would be level.
• 2023-04-15: Added layer of fiberglass to repair the sanded-through layers on the aft landing gear bulkhead and the lower firewall. Unfortunately, this glass was not properly wet out, and I discovered that it didn’t adhere correctly and needed to be removed.
• 2023-04-16: Removed incorrectly applied fiberglass on aft landing gear bulkhead and lower firewall. Replaced with fiberglass that was pre-wetted out prior to applying to the correct surface.
• 2023-04-19(?): Installed support parts A, B, C, and D. Glassed the D parts (horizontal supports) with 2-ply BID, overlapping onto the bulkheads 1-inch. Installed urethane foam on all support parts.
• 2023-04-20: Noticed that the port-side vertical support (Part C) between aft landing gear bulkhead and lower firewall was pushed in and was not flush with the lower longerons. Removed support & urethane foam. Remade support (which I determined was quicker than sanding the support down to the wood) and re-attached it. Also re-attached new urethane foam on top of the support. Verified that everything was in the correct spot before closing up.
• 2023-04-21: Measured, cut, and fit foam panels to fill out the air intake scoop from the aft landing gear bulkhead and lower firewall.
• 2023-04-22: Installed foam panels for air intake scoop between aft landing gear bulkhead and lower firewall. Installed additional urethane foam to fill space between start of air intake scoop (as calculated) and what was previously installed.
• 2023-04-23: Built sanding block for sanding naca scoop (glued a sanding belt to a 3ft long 2x4). Applied packing tape to area forward of air intake scoop to prevent accidentally sanding through that foam. No actual sanding, though.
• 2023-04-25: Sanded urethane foam from naca scoop down to the correct shape.
• 2023-04-29: Glassed inner part of naca scoop.
• 2023-05-01: Started repair of naca scoop layup.
• 2023-05-05: Finished repair of the naca scoop layup.
• 2023-05-08: Made scans of the large M-size drawings, 3d-printed contour template for sanding the forward half of the fuselage floor to contour.
• May and June 2023: Sanding fuselage floor to contour.
• 2023-06-18, 2023-06-19: Sanded 1/16th inch depression around landing brake. Placed landing gear in place (supergluing between masking tape).
• 2023-06-21: Built up 1-inch wide “skirt” around landing brake of tape. Sanded 1/16th inch depression in forward half of fuselage floor.
• 2023-06-22: Sanded 1/16th inch depression in forward half of fuselage sides.
• 2023-06-27: Started work figuring out antennae layout. First pass: Marker beacon, 1 nav, and 1 glideslope.
• 2023-06-28: Committed to removing the marker beacon. Will also combine the nav and glideslope into a single antenna.
• 2023-07-08: Routed slots for combined nav1 and glideslope antenna. Routed slots for coax for that antenna. Cut length of coax for nav/glideslope antenna.
• 2023-07-09: Cut copper tape for nav1 and glideslope antenna. Soldered copper tapes to coax to make nav/glideslope antenna. Installed and potted antenna/glideslope antenna.
• 2023-07-10 through 2023-07-22: Prepped fuselage floor for glassing. Sanding, sanding and more sanding.
• 2023-07-17: Glassed approximately 2/3rds of fuselage floor before we ran out of epoxy hardener. Covered floor in peel ply and ordered more epoxy.
• 2023-07-20 through 2023-08-05: Sanded and prepped fiberglass to finish glassing fuselage floor.
• 2023-08-05: Glassed the aft section of the fuselage floor (between the aft landing gear attach bulkhead and the firewall).
• 2023-08-06 through 2023-08-13: Sanded and prepped fiberglass to finish glassing fuselage floor.
• 2023-08-13: Finished glassing fuselage floor.
• 2023-08-15: Repairwork for the fuselage floor.
• 2023-08-16: Finished repairing the fuselage floor.
• 2023-08-22: Released video about process of glassing the fuselage floor exterior.
• 2023-08-22: Rotate fuselage right-side up. Created canard cut-out.
• 2023-08-23: Created 3d models and gcode of the upper longeron contour checkers.
• 2023-08-24: Printed upper longeron contour checkers.
• 2023-08-25: Glued 36-grit sandpaper to upper longeron contour checkers.
• 2023-08-26: Contoured the upper longerons/fuselage sides.
• 2023-08-27 through 2023-08-29: Prepped for glassing the port side.
• 2023-08-30: Glassed the port side.
• 2023-08-31: Prep for glassing the starboard side.
• 2023-09-02: Glassed starboard side.
• 2023-09-03: Trimmed glass from port side.
• 2023-11-20: Came back later and sanded smooth the seams where the fuselage sides transition to the fuselage floor.

Last updated: 2023-11-27 19:14:49 -0800

Chapter 8 - Shoulder Brace, Safety Belt Mount

Chapter 8, for me, is largely just installing the front safety harness attach points.

Changes

This is simple for me. I’m only going to install the 2 front seats, so no rear harnesses nor their attach points. I’m also not installing the front step - I’ll install a retractable one near the end of the build.

The only “common” plans change I made is to install the nutplates in the shoulder brace early.

Additionally, I might not be installing the rear half of the center duct?

Shoulder Brace

The shoulder brace mounts to the top of the front seatback and contains the top mounting points for the safety harnesses. It’s also where the headrests will be installed.

First, I cut out the foam cores for the shoulder brace. This is comprised of 2 pieces of 3/8th inch PVC foam that will be butted together into a corner piece. I cut these out using a box cutter knife, then I used a bandsaw to cut angles in to them so that they’d fit correctly on the front seatback. The top piece gets 4 hardpoints made of birch installed, which will anchor the shoulder straps of the safety harnesses. I installed these with 5-minute epoxy. While I was test-fitting the birch hardpoints, I broke the foam, but I repaired it with that same batch of 5-minute epoxy. One of the best things about composites is that everything is repairable. Sure, sometimes it’s faster or easier to remake a part, but everything can be repaired.

Next, I joined the two pieces with finishing nails and made test fits to get it right. It took a bit of back and forth to match the contour. Later, I picked up a contour gauge which would have made this go by much quicker. But I got it to fit.

After fitting it, I pulled out the nails and joined them properly with 5-minute epoxy. In another example of how everything is repairable, after this cured, I realized that I had joined them in the wrong orientation. So I cut them apart and rejoined them in the correct orientation.

Once these were joined, I reinforced the birch with BID. I could’ve done this prior to joining the foam, but I didn’t think to.

Next up is making some nut plates to secure the bolts that will hold the safety harnesses. These are plates of 1/16th inch thick 2024-T3 aluminum, with holes drilled to hold these nut anchors. I made some trial plates, learning how to use the tools and how to drill the holes correctly before I made the ones I’ll actually be using. Once I had the aluminum drilled, I riveted the nut anchors by using a table vice to squeeze the rivets into place - the rivet squeezer I ordered hadn’t arrived at that time. They’re not pretty, and they wouldn’t be accepted if they were, like, the wings on an rv-whatever. But, given the use case - preventing the nut anchor from rotating so that the safety harnesses can be secured - these are just fine.

With the nutplates machined, I prepped them for glassing with the recommended 220 grit sandpaper, then alumiprep and Alodine. Remember to wear a respirator when dealing with Alodine, it’s pretty nasty stuff.

I drilled holes through the center of the birch+BID hardpoints, I tested the fit of the bolt, then I glassed the interior.

I used a single ply of BID and let it sit for a few hours until it was in the green stage. I floxed the nut plates in, using some blocks and quarter-inch bolts to clamp them in place, and then attached the shoulder brace in place with fresh epoxy and flox, clamping it in place with the help of 3d-printed blocks.

For easier to follow steps on installing the nutplates during glassing, here’s what I did:

• Once the initial batch of epoxy (the one used to apply the fiberglass) is in the green stage, use a drill to open up the holes you previously drilled in the shoulder brace hardpoints.
• Apply flox (from a fresh batch of epoxy) to the nutplates where they’ll meet the fiberglass.
• Place the nutplates in place, threading in a quarter-inch bolt (AN4-6) from the exterior. I also used a quarter-inch piece of wood to help spread the clamping force on the exterior foam.
• Then finish installing the shoulder brace per plans.

Once it’s installed and the exterior is glassed, use a flashlight to help identify where the holes that you glassed over are. This’ll make opening them up significantly easier.

For the exterior, I used 2 plies of BID, extending an inch over all edges. Around the upper longerons, I wasn’t able to get the BID to wrap - too much complex corners - so I went back later and added some short strips of BID around there.

Additionally, all 4 shoulder harness attach points get 3 plies of UNI extending 2 inches forward of the shoulder brace, and about 3 inches around the back of the front seatback. This reverse 45 degree angle presented obvious problems with getting the fiberglass to adhere, so I printed these blocks to help clamp the fiberglass in place. Unfortunately, when I did clamped them in, I disturbed the fiberglass enough that the top layer of UNI was pulled out of place on this inner angle - which I didn’t notice until after the epoxy had cured. That, with some other errors, gave me a few hours worth of sanding, plus about 45 minutes of glassing to repair this.

After repair, I opened up the holes for the shoulder brace attach bolts. First I drilled a 1/8th inch pilot hole to check I got the location right, then I widened it with a 3/16th inch bit, then finally a quarter inch bit. I got the pilot hole correct first try on 3 of the 4 attach points. For the fourth, I got it on the third try, and I filled the other 2 holes with epoxy. Finishing the shoulder brace.

Lower Harness Attach Points

I started by cutting out 2 3 by 4 inch pieces of Finnish Birch Plywood. These are used in the outer harness attach points, to provide additional bracing against the lower longeron. By this time, I had a contour gauge, which I used to get the exact contour of the area, so I could sand the birch to fit in place. With that, I was able to get a very good fit on the first try. After that, I chamfered the front and rear edges so that the glass would smoothly transition off the birch and onto the fuselage side.

For the center harness attach area, no birth piece is needed, and instead I’ll be adding fiberglass on top of the center harness attach point. This is a piece of aluminum tubing which already had 7 layers of UNI laid up on top of it. I made a paper template to figure out the dimensions needed, and cut fiberglass for the area.

Upon install, I added wet flox to fill in gaps between the birch and fuselage. Per plans, it’s only necessary that the birth be flush with the lower longerons, and that it smoothly transition at the edges. Other gaps are acceptable. Then I brushed pure epoxy on the birch hardpoints and put them in place.

For laying this up, I chose to lay up all the fiberglass on plastic then transfer the layup to the install points. This is both easier and faster than trying to directly lay up on the fuselage. For the outer attach points, I made 2 rectangles with 7 layers of BID, which extend past the birch pieces by an inch. For the center piece, which had significantly more curves to deal with, I created 3 rectangles: 1 with 3 plies BID, UNI, BID, and the other 2 with 2 plies of UNI and BID. I cut these to size and individually installed them. This allowed me to better ensure that the fiberglass would conform to the curves. I didn’t do this for the outer attach points, because they’re significantly flatter.

After cure, I used a drill to open up the hole for the aluminum tubing in the center harness.

Center Duct Extension

I’m not certain if I’ll be using this yet. But I’ve started work on this, so I’ll include this.

This uses a mix of old and new school techniques. First, I started work on a template for the rear electrical duct extension. Because of the curves of the fuselage floor here, the Cozy plans advise creating a template for the duct out of cardboard or scrap foam. I elected to use scrap foam for this. Next, for a transition piece from the duct through the landing gear hellhole, the cozy plans have you create a plug that transitions from square on one end to being a circle on the other end. I elected to instead 3d-print that plug, mocking it up in CAD first.

[STL of duct transition plug](src/assets/cad/chapter_8/Duct Transition Plug.stl)

Once printed, I traced out the outline to figure out how much fiberglass to cut, and then I wrapped it in packaging tape as a cheap form of mold release.

Back to the duct itself, I used an extra piece of side wall from the front duct - I had accidentally made 2 right sides instead of a left side and a right side - to create the top, trimming it to size using my band saw.

Will I even finish the center duct?

There’s not a consensus on whether (the center duct) is structural or not. If it’s not structural, then I don’t see much of a point in me installing it. Except maybe as a form of cable management. The center duct is meant to be a heat duct, but the electric motor won’t generate enough heat for it to be worth it for me to route that heat forward. The gas builders install it because the primary output of a gas motor is heat, with only incidental amounts of usable output. So even with the (I assume) large amounts of waste in the heat duct, it’s worth it to install. My plan for heating has always been more direct heat - heating pads on the seats.

Videos

The first video I made on Chapter 8 covers the harness attach points. I had to stop because I didn’t have the hardware necessary to finish them. Additionally, as I covered earlier, I’m not entirely certain if I’ll even install the rear half of the center duct

Regardless, the first video covers only the harness attach points:

Log

• 2023-09-09: Made shoulder brace pieces, cut wood hardpoints, installed them in top shoulder brace piece using 5 minute epoxy.
• 2023-09-10: Joined shoulder brace pieces together, sanded front seatback in preparation for attaching shoulder brace to the front seatback.
• 2023-09-16: Contoured birch piece for outer seatbelt attach points.
• 2023-09-17: Installed birch & BID layup for outer seatbelt attach points. Installed layup for center seatbelt attach points.
• 2023-09-19: Started work on template for rear electrical duct extension. Created CAD duct transition piece plug.
• 2023-09-20: Printed transition piece plug. Wrapped it in packaging tape as a mold release. Finished template for rear electrical duct sides. Used an extra piece from the front duct to make the top for the rear duct.
• 2023-09-21: Cut aluminum for practice nutplates.
• 2023-09-23: Machined & Assembled practice nutplates.
• 2023-09-26: Machined & Assembled actual nutplates.
• 2023-09-30: Add first set of BID reinforcements to shoulder brace hardpoints.
• 2023-10-01: Added second set of BID reinforcements to shoulder brace hardpoints (first set didn’t end up flush with the foam).
• 2023-10-04: Alumiprep’d and Alodine’d nutplates (first try). I didn’t like how this came out, so I tried again later.
• 2023-10-05: Drilled center holes in shoulder brace hardpoints. Alumiprep’d and Alodine’d nutplates (second try)
• 2023-10-07: Glassed interior of shoulder brace. Installed nutplates. Installed shoulder brace onto front seatback.
• 2023-10-08: Removed clamping blocks from front seatback.
• 2023-10-09 through 2023-10-19: Sick with what was likely COVID. Be sure to take care of yourself!
• 2023-10-20: Glassed exterior of shoulder brace.
• 2023-10-21: Removed peel ply from shoulder brace. Marked spots to repair.
• 2023-10-23: Sanded fiberglass around repair spots.
• 2023-10-25: More sanding. Re-glassing the spots that needed repair.
• 2023-10-26: Opened up holes in shoulder brace hardpoints.
• 2023-11-13: Published video about Harness Attach Points

Last updated: 2023-11-13 07:50:16 -0800

Chapter 10 - Canard Construction

Log

• 2022-06-20: Took delivery of wings & Canard cores

Last updated: 2022-08-22 14:40:19 -0700

Chapter 11 - Elevators

Log

• 2022-06-20: Took delivery of wings & Canard cores

Last updated: 2022-08-22 11:12:01 -0700

Chapter 19 - Wings and Ailerons

Normally, you’re supposed to spend most of chapter 19 making jigs for the wings, hot-wire-cutting foam, epoxy’ing the foam together into the wing shapes, and then ultimately fiberglassing the wings. I got incredibly lucky and was able to skip all of that. In early May 2022, I reached out to Steve at Eureka CNC (Steve has since sold Eureka CNC) intending to purchase some of his cnc-cut wing cores. While he did have some wing cores available, he also offered to sell me these wings for only a few grand more. Obviously, I took the wings. A month and a half later, I took delivery of them. And they looked amazing.

This allows me to skip approximately 3/4 of chapter 19. I’m still taking stock of exactly where I need to pick up from. I also expect to not touch this for quite some time.

Log

• 2022-06-20: Took delivery of wings & Canard cores
• 2022-07-23 & 2022-07-24: Built cart for to store wings on until I need them. Link to tweet

Last updated: 2022-08-22 11:12:01 -0700

Mods to Investigate After Flight Testing Phase 1

The goal of the initial build is to get a safe, flying airplane as soon as possible. These are ideas for modifications that I want to do which are antithetical to that goal - they’re either experimental even for a homebuilt (and thus will take time to investigate and get right), or are ridiculous/a level of extra that I don’t need yet.

I also want to get these out of my head so that I can stop thinking about them when I should be thinking about/working on/doing research for the main build.

Carbon Fiber Canard, Wings, Strakes and Winglets

Just shy of making an entirely new airplane is replacing most of the exterior with carbon fiber parts. This would almost certainly result in a ton of weight savings, but would also require a lot of testing of sample pieces to get right. Especially if I want to try to get a similar level of strength as a fiberglass part, but with less layups.

Ducted Propeller

This will likely be the first mod I try.

This would be adding an airfoil in the shape of a ring around the propeller to turn it into a ducted fan. Ducted fans are more efficient than standard propellers. The only question is how much more efficient this is, and is it worth the extra weight?

Heated Canard, Wings, Strakes and Winglets

Slightly related to the “Make new lifting surfaces out of carbon fiber” is to embed heating material (I think I can use Carbon Fiber?) in the wings and run a current through them. This would serve as an anti-icing feature, making IFR flight much less scary.

There are a number of problems with this. Most obvious is I don’t want to risk a runaway heater. If the heater gets too warm, then it can soften the epoxy and cause structural failure. I don’t want to deal with temperature sensors that can potentially fail, so the best solution is to pass through a current/voltage amount that will never be enough power to heat the epoxy past the glass transition temperature. Depending on what material I use as the resistor, that depends mostly on the voltage.

Obviously, this would require a ton of time to get right, but this is one of the first things I want to investigate once I have a flying airplane.

Larger Lifting Surfaces

Increasing the length of the lifting surfaces (main wings + canard) should result in both increased efficiency and an increased carrying capacity. Meaning increased battery capacity. I would have to do research to determine how much longer each should be. If you make the main wings larger without adjusting the canard, all you’ve done is make it easier to enter a deep stall. I’m less sure about the effects of a larger canard relative to the main wings, but I’m sure it’s also not as desirable.

This would require a bunch of research to get right. Which means I’ll put it off for quite some time.

Motorized Landing Gear

This is an idea that’s maybe on the “don’t do” side, but there are a number of upsides to it and it might be worth the weight penalty. But electrically driven wheels would be great for ground operations. Using the propeller to taxi is very inefficient compared to directly driving the wheels. Additionally, this could be used to assist takeoff and landing by using the wheels for some small extra acceleration, as well as regenerative braking on landing.

However, this would add at least 30 lbs of weight to the system, which is not available. For an airplane that’s being optimized for cruise flight, it seems silly to try to add some optimizations for ground operations.

V2 Electric Airplane

This would be an entirely new airplane. One of the things I want to do with this is make a larger, 4+ person airplane. This would likely be made most of carbon fiber, with the battery compartment being lined with fiberglass (because fiberglass is non-conductive, unlike carbon fiber). Other thoughts include changing the fuselage side structure to make ingress & egress easier. As well as every other mod described here.

Last updated: 2022-11-27 11:15:05 -0800

Materials Used

Fiberglass aircraft are made primarily of foam, fiberglass, and epoxy.

Epoxy

There’s a number of epoxy systems you can use with your aircraft. This is a link to presentation on the Dos and Don’ts of epoxy

I elected to go with the MGS 335 system, which seems to be fairly popular amongst the different builders.

The MGS 335 system should be mixed at a 100:38 ratio of resin:hardener by weight.

Microballoons

Microballoons (or microspheres) are used with already-mixed epoxy to create slurry, wet micro, and dry micro. Microballoons should be stored in a covered container with as low humidity as possible. I use a small dehumidifier to help with this. To dry out microballoons, bake them at 250° Fahrenheit, then sift them with a flour sifter to remove lumps.

As noted, you mix microballoons with epoxy in the following mixtures, by volume:

Though, really, you’re supposed to add thee microballoons until the desired consistency is achieved.

Slurry/Micro Slurry

Slurry is mostly used to paint or squeegee over foams immediately before glass cloth is applied over them. Do not let the slurry dry before you apply the glass cloth. With urethane foam, use a full thick coat of slurry.

Wet Micro

Wet Micro is used to join foam blocks.

Dry Micro

Dry Micro is used to fill low spots and voids.

Flox

Flox is a mixture of epoxy and flocked cotton. There are two types of flox: standard and “wet”. Standard is epoxy mixed with just enough flocked cotton to make the mixture stand up, whereas wet flox uses less flocked cotton, and is mixed so that it’ll sag or run, similar to wet micro.

When using flox to bond a metal part, be sure to sand the metal dull with 220-grit sandpaper. Also paint pure epoxy (no flox) on the metal part prior to bonding with flox.

Peel Ply

Peel Ply is a term for a layer of fabric used to absorb extra epoxy, smooth out the transition between extra layers of fiberglass, and to improve adhesion when applying additional layups of fiberglass to an already-cured layer. Peel ply is always added as the last layer in a layup, and is removed in the “green stage” - when the epoxy becomes more viscous and changes to being a very tacky kind of goop.

The FAA has studied this, producing a document The Effect of Peel-Ply Surface Preparation Variables on Bond Quality. Which concludes that a thick-weave, uncoated polyester fabric is the best fabric to use as peel ply. I use a white, presumably uncoated, thick-weave polyester sourced from a local fabric store.

Last updated: 2022-09-09 14:09:51 -0700

Cozy Newsletter

Marc Zeitlin has all of the Cozy Newsletters available at cozybuilders.org. Here’s some of the more relevant information to the Coz-E build.

Note for builders of gas-powered Cozys: I have not included any of the tips related to the fuel and engine systems. This will be a fully-electric airplane, and so I have no reason to include those for my reference.

Builder Tips

In addition to some of the dedicated articles going in more depth on specific advice, here’s a sorted list of most of the advice given in the Cozy Newsletter.

General

• Keep joints tight with “no joggles” to improve performance. (Cozy Newsletter #4)
• Do not cut up the large-size drawings. Instead, trace templates onto another sheet of paper and save the original drawings intact. (Cozy Newsletter #12) (Personally, I take photos of the drawings with a ruler for scale, but that’s for archival purposes).
• “NEVER allow yourself to be distracted while working on your airplane, such that you might not complete a task, and then forget to finish it later. We all have had the experience of visitors dropping in while we are working. Even if you have to be rude, finish the job before laying down your tools.” (Cozy Newsletter #24, The Canard Pusher #57)
• Joints between all permanently installed parts should be taped. (Cozy Newsletter #24)
• This one comes up in more than a few newsletters: Check your attach bolts. If more than 2 threads extend through the nut, add another washer. Don’t let the threads bottom out before the bolts are tight. (Cozy Newsletter #34)
• Holes through bulkheads or instrument panel do not need to be glassed. (Cozy Newsletter #36)
• Wherever possible, orient bolts either heads up or heads forward. (Cozy Newsletter #38)
• “Extruded piano hinges MS20001-P6, P5, P4, etc. are all identical except for the width of the flange. It is easier just to order and stock the P6, and if the flange is wider than necessary, trim it on your band saw.”“ (Cozy Newsletter #39)
• Try using a light inside to identify holes that have been glassed over. (Eric Westland, Cozy Newsletter #39)
• For bending aluminum, rule of thumb is that the bend radius should be no less than the thickness of the material. E.G. 1/16 inch thick aluminum can bend 90°, provided the radius is 1/16 inch or greater. Always bend aluminum across the grain, not with it. (Ray Goldsmith, Cozy Newsletter #39)
• Always order a little extra material, especially in the beginning, so that you don’t run out of epoxy, cloth, micro, or flox before finishing a job. (Cozy Newsletter #69)
• “all aluminum parts [should] be protected from corrosion by cleaning first with Alumiprep 33 or metal prep #79, and then soaking in Alodine 1201, which is a visible (golden brown) moisture barrier, greatly increasing resistance to corrosion. This also acts as an excellent surface to bond epoxy or paint. Even if you do not live near the coast, the airplane you are building could some day end up there.” (Cozy Newsletter #81)

Safety

• Use nitrile gloves for your hand, with water-soluble barrier cream on the arms. Switch out the barrier cream early and often. (Cozy Newsletter #89)
  • This has evolved significantly, from the original advice, starting way back in #7: To protect your skin from epoxy, Nat heavily recommends water-soluble barrier creams. If you must use gloves, use nitrile gloves. Under no circumstances, use latex gloves.

Tools & Shop

• Store and dispense glass cloth on a rack in a wall cabinet. Hinge the front of the cabinet so that it opens to form a table to lay out the cloth for cutting. Keep a clean pair of scissors there just for cutting clean cloth. (Cozy Newsletter #14)
• Put a 2 inch overhang all around your work table. This makes clamping parts to the table much easier. (Cozy Newsletter #17)

Neil Clayton says the tool (s) he uses the most are sanding sticks he made himself. He says Home Depot paint dept gives away 2 sizes of paint stirrers; a 1“wide stick and a 2” wide version. Both have a nicely formed handle. I always grab a handful as I pass towards the check out. Then at Lowe’s I pick up some 12” x 18” sanding sheets (the kind used on those big floor sanders). Several grades are available. I use left over epoxy to glue the stirrers to the sheets and then split them into individual sanding tools with a razor blade when the epoxy has dried. One sheet yields about 10-15 sanding sticks of various sizes so they are cheap and disposable. Try it!

(Cozy Newsletter #78)

• Get a Fein tool for trimming cured layups and electric scissors for cutting cloth and peel ply (and trimming wet layups) (Cozy Newsletter #82)

Antennae

• Make sure the nav antenna misses the area in the bottom of the fuselage that will be cut out later for the nosewheel well. (Eric Westland, Cozy Newsletter #39)

Location

From Cozy Newsletter #34, with a discussion on where to place the antennas (For a Cozy 3).

• Transponder: In the nose ahead of the rudder pedals with the probe extending through a 1/4“ hole in the bottom of the fuselage.
• Nav: Two good locations:
  1. Underneath the skin on the bottom of the canard, legs extending forward and offset to the radio side.
  2. Under the skin on the bottom of the fuselage, legs extending forward on either side of wheel well.
• ADF: Use King electronic antenna and locate inside cockpit under right front seat.
• Com: Located under skin of winglets, per plans.
• FM and/or Marker Beacon: Under skin mid-wing.

GPS was not specifically called out, presumably because it wasn’t available back then.

• Be careful in laying out VOR antennas on the bottom of the fuselage to avoid future cut outs for the landing light and nose wheel. (Thomas Kennedy, Cozy Newsletter #55)

Here are the dimensions (each arm) Jim Weir has published for the following di-pole antennas:

• Marker Beacon - 34.3 in.
• FM Music Radio - 26.2 in.
• VOR/LOC - 22.8 in.
• VHF/COM - 20.3 in.
• Amateur 2m - 17.7 in.
• Glideslope - 7.5 in.

The COM is the only one that needs to be vertically oriented. All the rest should be horizontal. The VOR should be a V shape. Some say the marker beacon should be along the airplane centerline. I installed mine in the wing and it worked just fine. You should keep the tips of di-pole antennas away from metal, and away from each other, particularly transmitting antennas. I have used Jim’s measurements and they work well-better than most factory antennas. My GPS antenna is mounted in the nose, just under the inspection cover (pointed upward toward the satellites) and works well there. I like this location because the co-ax is very short. My transponder antenna is in the bottom of the nose, with the probe just sticking through the skin (pointed down toward the ground stations). You can make your own transponder antenna. It should be 2.65 inches long from the tip of the probe to the ground plane, and insulated from the ground plane. The ground plane can be made from .062 aluminum. It is about 8 inches in diameter (I lost my drawing). I have made my own from one of Jim’s kits, and also from scratch, and they work well. You attach the center conductor to the radiating rod, and the co-ax ground to the ground plane. You get a lot of satisfaction when you make your own antennas, they are all hidden, and they work better than the ones that cost mucho bucks!

(Cozy Newsletter #73)

Canard

Several Mark IV builders have advised us that their canards are slightly longer in cross section (chord) then shown on M-11, and when the lift tabs are in contact with the forward face of F-22, the trailing edge of the canard is 1/8 to 1/4“ aft of the forward face of F-28, and that the alignment tabs cannot be built as shown. Should this also be the case with your canard, you may lay up additional plies of BID locally on the forward face of F-22 up to 1/8“ thick, or remove up to 1/8“ of the trailing edge of the canard locally at the tabs, or a combination of both.

(Cozy Newsletter #44)

Canard installation. Make sure you allow for at least 3/4“ of horizontal travel for the canard so the incidence pins will not bind during installation or removal.

(Cozy Newsletter #50)

Some builders are still having trouble with the trailing edge of their canard being too low, such that the airfoil dips down and they have to mount the elevators lower than desired to get the required 15 degrees of trailing edge up travel. We suggest that you remove 1/16th inch (no more) from underneath the fishtail on the templates before cutting the foam cores, or from the foam after cutting to avoid this problem. We have made this suggestion to Feather Lite as well to do on their precut cores.

(Cozy Newsletter #53)

Canopy

On getting a smooth edge where the fiberglass contacts the canopy:

Builder Wayne Hicks said he made a paper pattern of the canopy deck, then laid up the 2 UND on plastic (and later the 2 BID), used the pattern and razor pizza cutter to make a nice, clean cut along the pattern lines. Then he laid the piece nice and neat up against the glass. There were absolutely no problems. No excess flox, no excess resin, no excess glass. Never had to use a dremel! If anything, all he had to do was cut one or two stray strands with a razor blade. He did this for both the outside and inside canopy deck layups, and highly recommends this technique.

(Wayne Hicks, Cozy Newsletter #70)

to avoid corrosion of the canopy hinge attach bolts. Saturate the bare wood in the holes through the longeron with epoxy to seal the wood fibers. Do this by dipping a cotton swab into epoxy and then inserting it into the hole. Alternatively, you can drill the hole slightly oversize, fill it with flox, and after cure, redrill it to the correct size.

(Nick Parkyn, Cozy Newsletter #83)

Contouring

• Fill low spots before sanding surrounding areas down to contour. This keeps your low spots from turning into high spots. (Cozy Newsletter #22)

I found a method to make the final canard contour perfect per plans. I used the checking contour templates for the canard to cut out (hot wire) a block of styrofoam. Then I stuck a large piece of sandpaper to it and used it to give the canard a perfect contour over its entire length. The same thing worked fine for the elevators.

(Jean-Jacques Claus, Cozy Newsletter #85)

Cowlings

Cowlings should have extra layers of BID along all the edges to make them more rigid when they are removed from the airplane, and to provide more beef for the fasteners. This is shown in Figures 12 & 15, Chapter 23, but builders were not instructed to add them because the supplier (Featherlite) added them while the cowlings were still in the mold. In February 1999, when we inspected AeroCad cowlings, we noticed that they were not doing this. We asked them to add the extra layers while the cowlings were still in the mold. They agreed, and we approved their cowlings contingent upon their doing this.

(Cozy Newsletter #68)

Elevators

Your elevators should balance with the weights shown in the plans. If they do not, do not try to balance them by adding weight inboard. Instead, discard them and make new ones more carefully to make them lighter (you can probably re-use the torque tubes and other hardware). The mass balance called out for the elevator and the specification for balancing them applies only to an elevator fabricated with the same weight and stiffness as that which has successfully passed all the flutter testing. It is extremely important, and life-critical that the manufacturer or owner of each Cozy, or any plane for that matter, assure himself without a doubt, that the control surfaces are conformal to those which have passed flight tests and been shown to be flutter-free.

(Cozy Newsletter #24)

Elevator torque tube offsets (CZNC-12A for the Roncz canard). The drawing we submitted to Brock Mfg. stated that these offsets should be a slip-fit in the l“ OD x .035“ 2024 T3 torque tube. Brock Mfg. cannot guarantee a slip-fit because the tolerance on wall thickness for .035“ 2024 T3 drawn tubing is +/-10 %. To prevent too loose a fit if the wall thickness is on the thin side, the specification for the offsets has been tightened to .930’ OD. If the wall is on the thick side, you will have to sand the offsets to obtain it slip-fit.

(Cozy Newsletter #50)

In the first edition Mark IV plans Chapter 11, p.7, we instructed builders to mount their elevators at zero degrees with jig L so that there was a .2 inch gap between the canard and the elevator. When some builders did this, they were not able to get 15 degrees trailing edge up of the elevators. So we changed the L jig (see p. M-18) to mount the elevators at 15 degrees trailing edge up and zero gap. Please note this on p. 7, Chapter 11, 1st paragraph and Figs. 17 & 18.

(Cozy Newsletter #68)

Epoxy

Starting with Cozy Newsletter #61, the MGS system is now recommended. Either the L335 or L285 system.

• Save the foam scraps. Many parts can be made by gluing small pieces together with 5-minute epoxy. (Cozy Newsletter #5)
• To conserve epoxy, mix microballoons with any excess into a very thick paste, and trowel that over your finished fiberglass parts. This will save you time when finishing. Rough the surface with 36 grit sandpaper first, though. (Cozy Newsletter #5)
• Lava soap works great to remove epoxy from your hands. (Cozy Newsletter #35)
• Use the bottom of 8 ounce mixing cups to make 5 minute epoxy (Cozy Newsletter #17)
• Store near-empty bottles of 5 minute epoxy on their side. (Cozy Newsletter #28)
• Mixing some flox with 5 minute epoxy will make a bit stronger joint and prevent it from running. (Cozy Newsletter #35)
• Epoxy Resin should be clear and colorless (or a slight yellow). Hazy indicates crystallization, and the epoxy should not be used until it is heated and becomes clear again. (Cozy Newsletter #85)

Cozy builder Phillip Johnson, in the Pacific Northwest, protects his resin and hardener from exposure to air by putting a plastic bag over the top of his plastic containers in his dispenser after filling them, with the bag resting on top of the liquid surface and with enough slack so the bag will follow the surface down as material is used, and then putting the cover on the container.

Cozy Newsletter #54

Fiberglassing

• Micro joints (which hold urethane blocks together) cause bumps when glass is laid over them. Undercut all joints with a dremel before shaping and before layup. Fill the undercuts with fresh micro just before the layup. (Cozy Newsletter #5)
• To prevent air bubbles from coming back in after being worked out, cover the cure area with saran wrap (or any plastic wrap) and squeegee the air out. (Cozy Newsletter #5)
• Cut the glass cloth ahead of the layup to the approximate size. Do this after the foam is vacuumed, but before applying any epoxy. Drape the cloth over the part, cut it to size, label it with a felt-tipped pen, roll it up and set it aside. (Cozy Newsletter #14)
• To prevent UNI from unraveling when cutting it lengthwise, lay down a strip of masking tape and cut down the tape’s center. Scissor trim the masking tape off after the cloth has been wet in place. (Cozy Newsletter #14)
• Use a sabre saw, not a band saw, to cut fiberglass. (Cozy Newsletter #17)
• Use a rubber squeegee to spread epoxy on the glass. Take a 6 inch squeegee, and cut it in 2. Remove the epoxy after use. You can use a belt sander to remove any cured epoxy, but it’s less work to just clean it. (Cozy Newsletter #31)
  • It takes a couple of minutes for epoxy to soak in and wet out one or two layers of glass cloth. Rather than waiting for this to happen in a small area, use this time to spread epoxy over as wide an area as possible. Pour a ribbon of epoxy on the cloth and follow behind with the squeegee spreading it. Keep the pressure very light, so as to not disturb the cloth. After the cloth is wet out, you can increase the pressure of the squeegee to distribute it. You can pick up excess epoxy with your squeegee, wiping it off on the edge of your mixing cup. Conversely, you can apply a small amount of epoxy by dipping your squeegee into the mixing cup. If you practice these techniques, you will find that you can make lay ups much more quickly.
  • Do not waste time trying to get just the right amount of epoxy on the lower layers of a multi-ply layup. Actually, cloth wets out better from underneath and not as much air is trapped between layers. It is only on the last layer that you need to spend the time working on getting the air out an applying the optimum amount of epoxy.
  • Do not leave bristles from the throw-away brushes behind in your layups–they weaken it. You can pick them up with the tip of a brush.
• Double check that you remove the peel ply. If you leave it in and glass over it, the part must be discarded and rebuilt. (Cozy Newsletter #37)
• Instead of peel ply tape, get the 38 inch wide peel ply. Then, when you need tape, cut the peel ply to whatever width you need. (Eric Westland, Cozy Newsletter #39)
• “Buy cheap 100% polyester (or Nylon) cloth (peel ply) by the bolt and cover all layups with it. First test a sample piece to make sure no additives are present that may cause removal to be difficult. I find that I do not need to add extra resin to wet out the polyester, just stipple and use a hair dryer. I make it a habit to strip the polyester the next day, so none is left by mistake. This eliminates the need for a lot of sanding, and dust breathing. All layups look and feel smoother.” (Steve Blank, Cozy Newsletter #46)
• “We use peel ply extensively on all of our layups. Glass fibers act somewhat like a sponge, and can soak up extra resin. If a layup is resin-rich, it is not only too heavy, but it is weaker than a layup with just the right amount of resin. The peel ply seems to hold the fibers down more compactly, so they don’t soak up too much resin. Sort of a poor man’s vacuum bagging. If you use peel ply, you can make the layup a little wet, then lay down the peel ply, and with a squeegee and hair dryer, remove the excess epoxy (if the layup was really wet) or else free up almost enough epoxy to wet out the peel ply, and add epoxy as necessary to wet it out completely. The only thing you need to be careful of is that the layup underneath the peel ply isn’t so dry that it is porous, because then the compressive strength would suffer (buckling in compression). Synthetic fabrics with a straight weave (not knit), like nylon, polyester, or dacron, as long as they aren’t fuzzy, and don’t contain cotton or rayon, seem to work well. Sometimes you can find real bargains at fabric stores. But test a swatch first. Alexander Aeroplane lists peel ply at $2.25/yd. in 45” widths, which is quite reasonable.“ (Cozy Newsletter #47)
• “BID is always cut on the bias and overlapped l’ in every direction, but UND is overlapped only in the direction of the major fibers.” (Cozy Newsletter #50)
• Whenever possible, use peel ply. As an extra precaution, sand the surface that the peel ply was on before applying the bond. (Cozy Newsletter #55)
• A wet-on-wet layup usually results in the strongest bond between layers. To do this, wet out the first layer of glass, squeegee the air out, but leave an excess of epoxy and then lay down the next layer of cloth on top to soak up the excess. (Cozy Newsletter #55)
• Do not glass over a cured, microed surface, if structural strength is required. (Cozy Newsletter #55)
• Put flox in a bag and dispense it like cake frosting. (Dave West, Cozy Newsletter #62)
• In tight corners where it’s hard to roughen up the surface with conventional sandpaper, try a carbide studded grout removal tool that has carbide deposited on the edge of a steel blade approx. 1“ long attached to a handle with a slight “S” bent arm. (John Fritz, Cozy Newsletter #66)

Early in the game we learned that various fabrics sold in fabric stores work well as peel ply, and can sometimes be purchased at quite a low price. We learned that hard-weave (not fuzzy or stretch weaves) nylon, polyester and dacron work well. Watch out for cotton and rayon. Often there are roll ends tables where things are reduced for clearance. Once I was able to purchase nylon like is used in ski jackets for $.45/yd, and later wished I had bought the entire table. It is best to take a sample home first, use it on a wet layup, and make sure it peels off, before buying a whole bunch. Make sure you iron any wrinkles out of peel ply before using it. Otherwise they transfer to the layup.

(Cozy Newsletter #68)

• A general rule is that whenever joining two or more intersecting surfaces together, they should always be floxed and taped, whether instructed to do so or not. (Cozy Newsletter #71)

• • A rule of thumb for estimating how much mixed epoxy you will need for any given layup is 1 oz of epoxy for every oz of cloth needed. This includes the micro layer. “For example, the bottom side of one Cozy MKIV wing will need 12 yards UNI (7 oz/yd. 38” width) or 84 oz. total. Epoxy weighs 9 lbs. (144 oz) per mixed gallon, so you will need 84 oz. mixed epoxy to wet out 12 yards of UNI.” (Cozy Newsletter #80)

• Leave a little mixture of epoxy in the cup and check it for cure the following day. If it has cured correctly, a scratch on the surface will leave a white mark (Cozy Newsletter #81)

• “[If] glass cloth wets out and becomes transparent, it is safe to use. If it stays white, throw it away.” (Gary Hunter, Cozy Newsletter #83)

• “DO NOT USE glass that has become wet.” (Gary Hunter, Cozy Newsletter #83)

• Take your time. Do not apply epoxy until you are 100% satisfied with the shape and integrity of the part. (Neal Johnson, Cozy Newsletter #84)

Tom McNeilly’s Vacuum-bagging technique

From the Cozy Newsletter #4, this describes a way to get very light layups using vacuum bagging. Later newsletters state that vacuum-bagging is generally not worth the effort, though.

Place the layup down as quickly as you can, cover it with perforated poly. Cover that with soft carpet underlayment foam, and cover that with regular poly. Tape down the edges, and pull a vacuum with a vacuum cleaner house between the layers. The vacuum should squeeze the excess epoxy out of the layup and into the soft foam. After cure, strip the coverings off, including epoxy. Tom claims this saved 6 lbs per wing.

Dealing with BID

• You can create 2 inch wide BID cloth tape by rolling the BID cloth into 1 inch diameter rolls at 45° to the selvage edge. (Cozy Newsletter #4)
• For installing BID into the corner tapes: Layup the BID you’ll need onto the foil. Squeegee it out to a good layup. Cut your tapes out of this layup, cutting through the glass and aluminum foil. Sand and paint a coat of epoxy onto the area to be laid up over, then use your fingers to bend the aluminum foil to form the tape into an angle to fit into the corner. Position it, then lightly squeegee or stipple it into place. Peel the aluminum foil off, stipple any remaining air bubbles out, peel ply the edges. (Cozy Newsletter #4)
• Sand a slight recess (1/32 inch) into the foam on flat or curved surfaces where you’ll place BID overlap joints so overlaps won’t cause a bump in the finished surface. (Cozy Newsletter #5)

Finishing

• Don’t use featherfill. It does not bond well with epoxy and will cause the paint job to delaminate in a year or two. (Cozy Newsletter #10)
• Make sure to fill and finish paint the inside of the cowlings. (Cozy Newsletter #22)
• Most silicones give off corrosive fumes, Dow Corning Silastic RTV 738 is a non-corrosive sealant, use it to secure the cover plates over the wing attach bolts. (Bruce Ibbotson, Cozy Newsletter #22)
• Use Polyurethanes, not Enamels or Lacquers. Enamels and lacquers chip in heavy rain. (Vance Atkinson, Cozy Newsletter #23)
• “It is not necessary to use a spray booth to paint an airplane and get a good finish. You can spray the finish coat in a garage or outside. After cure, wet sand with 1000 and then 1500 grit, and then buff with 3M Finess-it II. This will remove orange peel, overspray, dust, and other imperfections and result in a very flat surface with high sheen. This works well with Ditzler Deltron (acrylic urethane) which is the finish we recommend.” (Cozy Newsletter #41)
• “We recommend use of a high-build epoxy primer. Cozy builder Nick Parkyn in Western Australia says that he has used the new System Three water reducible Epoxy Primer and Polyurethane paints and is impressed with the performance. We are not familiar with these products and have asked him for more information. We would welcome other builders input.” (Cozy Newsletter #41)
• “Have you ever tried to pour paint from a full 1 gallon can and had it run down the side? Try this. Wrap a strip of masking tape (2- wide works the best) around the top of the can, and punch a hole in the groove around the rim. You should be able to pour without spilling, and the excess paint should drain through the hole back into the can. The hole will be sealed after the cover is reinstalled.” (Cozy Newsletter #41)

Mark IV builder Chuck Larson writes, “I’ve discovered a tip that works great for wet sanding the Sterling primer. I use a grouting trowel which is used for grouting ceramic tile. I bought mine years ago at Color Tile. It is 9” x 4“. It has a wooden handle attached to an aluminum plate which has dense rubber foam bonded to it. Wet or dry sandpaper will adhere to the foam rubber well enough to allow the sanding. It keeps the surface flat, and it is easy to remove and wash out the sandpaper“.

(Cozy Newsletter #44)

If you need to pour paint from a can without spilling and wasting a lot (paint is expensive!), first punch a hole in the trough around the rim to let paint drain back into the can. Then wrap some 2“ tape around the rim of the can, extending up above the rim about 1-1/2“. Now you can pour the paint out of the can without it running down the side of the can, and any excess will drain back into the can.

(Cozy Newsletter #45)

• If you use masking tape when painting, don’t leave it on a long time, and when you remove it, pull it back at 180° to itself, not 90°. (Cozy Newsletter #63)

I just completed painting the bottom of my plane. I’ve used the new poly fiber system. All water based. Very impressed. Pin holes are a total non-issue. Particularly when you use the Superfil instead of normal microballoons. Superfil is substantially better than micro. Trapped voids are rare. Sands easy, mixes easy. Areas where I used micro have more pinholes, but using the Smoothprime primer is highly effective at filling pinholes. It really does flow into holes. Must have low surface tension or something. I also elected to use their water based “topgloss” top coat (polyurethane). So far, I really like it, but I have yet to color sand and buff the paint, so the verdict is still out. I haven’t done adhesion tests yet. On my second coat (of three) of topgloss, I got too aggressive and had quite a few runs on corner radius. I elected to sand the entire coat before applying my last coat. Final coat looks good, although areas have a matte appearance (too lean). No runs. The top coat appearance is quite sensitive to spray quantity. More is better unless you develop runs. I had no problems with vertical surfaces. I learned that I don’t need to use finer than 120 grit when sanding micro or superfil. The primer fills all those scratches fully. I used a sanding stick only to achieve straightness, then I used an electric random orbit sander with 220 grit. I was willing to sand the primer completely off to achieve uniform appearance. I mixed paint and primers 1 cup at a time to avoid splatter. I’ve heard people describe sanding and painting as the worst part of the process, but I enjoy it. It’s great to see objects finally transformed into airplane like appearance. Just wanted you to know more about these new products. My superficial conclusions: Superfil micro replacement is way better. Smoothprime primer is way better. Topgloss top coat may not be better (lotta work)….we’ll see when I’m further along.

(Al Wick, Cozy Newsletter #67)

I put the drip lip on the front fuselage cover last night. Here is a hint for those who have yet to do this. The plans say to build up an 1/8 inch ledge with balsa or foam. I used a piece of 1“ wide webbing (either climbing rope type webbing or load binding strap type webbing) with duct tape on top of it. The webbing will bend around the curve just right, and it is about the right size.

(Norm Muzzy, Cozy Newsletter #72)

• Do not use polyurethane paint. Nat recommends a acrylic-urethane (PPG “Deltron” or “Concept”) (Cozy Newsletter #83)

Fuselage

• If you have difficulty stickling the copper foil tape to the bottom of the fuselage, try heating it with a hair dryer and pressing down on it with a hard rubber roller. (Eric Westland, Cozy Newsletter #39)

Fuselage Sides

• Put a 6“ wide strip of wax paper between the masonite and the foam pieces where the edges butt together. This will make the eventual removal from the jigs MUCH easier. (Thomas Kennedy, Cozy Newsletter #49)
• In assembling the bulkheads to the sides, forget the ropes to tighten the whole assembly. Use threaded rods and wing much more precise control. (Thomas Kennedy, Cozy Newsletter #49)
• Buy a kiln-dried board (not construction grade) for your WL 0.0, and use bolts rather than nails to secure. This will allow precise placement and later shimming if required. (Thomas Kennedy, Cozy Newsletter #49)
• I had very good results mounting my WL 0.0 board on my work table which was mounted to 4x4s glued to my garage floor. There is no such thing as a level garage floor. The work table becomes your level surface. (Thomas Kennedy, Cozy Newsletter #49)

Foam

• Before cutting wing sections, clamp the styrofoam blocks and glue them using pour foam. This produces very thin joints that keep the styrofoam joined together while still allowing the hot wire to cut them without catching. (Tom McNeilly, Cozy Newsletter #4)
• You can use a dremel tool with a router attachment to make the small depressions in foam as called out in the plans. (Cozy Newsletter #15)
• Covering the edges of hot wire templates with copper foil tape makes them easier to use (Cozy Newsletter #38)
• “The 2 lb./cu.ft. large cell blue styrofoam called for in the plans is called Styrofoam in the Aircraft Spruce catalog, Polystyrene in the Wicks catalog, and extruded Polystyrene in the Alexander catalog. Urethane foam may also be called Polyurethane foam.” (Cozy Newsletter #39)
• Use a router to carve 1/8 inch and 1/16 inch relief cuts in the foam (e.g. in chapter 9, on the landing brake). Attach the router base to a 36 inch long board to make quick, smooth work of the cuts. (Eric Westland, Cozy Newsletter #39)
• Try using double-sided tape on the forms to hold the foam down. Recommended the kind used on golf club grips or carpet tape. Should peel right off with little or no effect on the foam. (Mark Loy, Cozy Newsletter #64)

If you “screw-up” when cutting airfoils, and have a bad cut on one side (top or bottom), no need to throw it away and start over. Just slice the airfoil apart, and splice in a piece of styrofoam (like making a sandwich), and re-cut it. Don’t know how to make a sheet of styrofoam? Just clamp your hotwire saw 1⁄2 inch or so above your work bench, and slide your scrap chunks of styrofoam underneath, sorta like cutting boards from logs. The beauty of composite construction is its hard to spoil something so bad you can’t fix it.

(Cozy Newsletter #77)

Joints

For Structural reasons, all joints should be taped with 2 plies of BID 2’ wide (1“ on each surface) cut on the bias (45 degrees) unless otherwise specified. After application, the tape should be peel-plied. One recommended way to accomplish this is to wet out 2 plies of BID on wax paper or Saran wrap, cut into strips, apply over joints, remove the wax paper or Saran wrap, cover with peel ply and wet out same. Wax paper should only be used with discretion. It should not be left in place while the epoxy cures, because wax will be left behind, and it should only be used on those joints which will not be later covered with additional layups.

(Cozy Newsletter #50)

Landing Gear

• “If you wrap the gear leg first with 1/8” fiberfrax, and then with shiny aluminum, you will greatly reduce the heating by radiation.“ (Cozy Newsletter #24)
• “Provide a vent for hot air to escape from your wheel pants. Hot air rises and will be trapped inside the wheel pants unless you provide a vent, on the top, on the disc side, for it to escape.” (Cozy Newsletter #24)

In Chapter 9, p.2, it says that the [landing gear] strut should be cut at 8 degrees, but on p.3 it says 13 degrees. There has been some disagreement as to what is the correct angle, and it also depends on whether it is measured at the leading edge or trailing edge. The angle is not critical, so shoot for somewhere in between.

(Cozy Newsletter #68)

I thought it was in the plans, but maybe not. An easy way to make the fairing where the main gear strut enters the fuselage is to shape it with molding clay (Playdough), glass over with at least 2 layers of BID, attaching to both the strut and the fuselage. Then after cure, make a saw cut at least 1/8“ wide halfway between the fuselage and strut, remove the strut, and dig out the clay. If the cut is wide enough and the inside of the fairing is empty, the strut can flex without cracking the fairing. This should be done when the fuselage is upside down with no weight on the gear. Then when you are flying, both halves of the fairing will be aligned for minimum drag.

(Cozy Newsletter #72)

Panel

• Don’t cut holes for your panel until everything else is done, and you’ve put together the instrumentation. Build a mock panel first (Nat suggested 1/4 inch plywood. You can also use cardboard) to test it out. (Cozy Newsletter #5)

Spars

The 3 inch spar cap material presently being supplied is only 2-3/4 inches wide, but not to worry. If you pull out the cross threads as instructed for the canard, centerspar, and wings, and then squeegee after applying epoxy, the strands spread out to 3 inches, and you will be able to fill up the troughs with the maximum amount of glass they can hold.

(Cozy Newsletter #65)

Strakes

Bending 3/8 in. PVC is a universal problem. When I re-built my right strake, I made a scribing tool from a block of wood and a hacksaw blade. I made a saw cut in the wood and 5-minuted in the hacksaw blade so it protruded 1/4 in. I carefully scribed a number of cuts in the PVC where I wanted it to bend, then bent it into the frame and fastened it in the bent shape, and then microed and glassed the scribed surface. That worked better than anything else I have tried to date.

(Cozy Newsletter #10)

• Postpone installing the center section spar and building the strakes until you have completed as much work inside the fuselage as possible. It is much easier to work inside the fuselage with it sitting on its side on saw horses. (Cozy Newsletter #35)

It is important for appearance that the strakes are straight (no bumps), and properly aligned with the wing. The wing should be installed on the center section spar when the strake is built, and a straight edge should be used to make sure that the leading edges of the ribs are aligned with the leading edge of the wing, and that all points on the rib airfoils align with the wing airfoil. Of course, you will need to make allowance (using spacers) for the 3/8“ foam and several layers of glass skin which will be subsequently be laid over the ribs.

(Cozy Newsletter #44)

Strakes, Chapter 21, p.2. Please note that the fuel cell ends at Rib R-57. The empty space outboard is necessary as a buffer to prevent any fuel, liquid or vapors, from reaching the wing. An open drain at the low point would indicate any fuel leakage through R-57.

(Cozy Newsletter #64)

Keith Scull (England) didn’t want to pay the cost of shipping Featherlite’s leading edges all the way from California, and he didn’t like the idea of carving foam blocks, so he came up with an innovative alternative. He purchased some thin card stock from a local art supply store, wrapped it around the leading edge of the strake using the ribs as guides, trimmed it so it just fit into the recess provided for the leading edge layup, and taped it in place with 20 inch long strips of masking tape. Then he cut openings in the top so he could pour pour-foam into the empty space. The openings were 3 in. x 1/2 in. every 6 to 9 inches parallel to the strake leading edge. He poured the pour foam thru the slots, let it expand and cure, and then removed the masking tape and card stock. He said the result was a perfectly formed leading edge which required very little sanding to prepare if for the glass layup (see pictures). He said it sounds like a lot of work, but only took a couple of hours, and did a better job with less mess.

Keith said when he leak tested his strakes using an altimeter, the right strake had a very small leak (20 ft. in 3 hrs.). He couldn’t find it with soapy water. So he called in a local refrigeration man who filled the strake with R22 and used an electronic sniffer to locate the escaping gas. He found the leak. It was traveling along the score lines he had cut in the foam to bend it around the ribs, and escaping where the foam joined the fuselage. Apparently he hadn’t completely filled up these score lines with micro. The cure was quite simple. He drilled a very small hole thru the top skin into the score line at the tank rib and injected epoxy using the vacuum method described in the plans. To be doubly sure, he did this with every one of the score marks, and found three to be leaking. After filling with epoxy, the altimeter indicated no more leaks.

(Cozy Newsletter #68)

Wings

There is a discontinuity (dip) in the top surface of the wing at B.L. 67.5 (see Chap. 19, p. 10). This is intentional. You have not made a mistake!

(Cozy Newsletter #53)

when you attach the wings many times in the shop, don’t use the AN365 nuts over and over again, but instead get some fine thread nuts at the hardware store which are much easier to install and you don’t have to worry about them wearing out.

(John Epplin, Cozy Newsletter #73)

Unsorted

Jim White’s advice

From Cozy Newsletter #67, copied verbatim:

1. Attach the wing before working on the stake and bondo a 1/4“ or so thick template to the end of the main spar that matches the contour of the wing. This is needed so that the strake matches up to the leading edge.
2. When cutting out the top and bottom PVC foam pieces, finish one (top or bottom). The top is the same as the bottom except it is 0.2“ longer. All the angles are the same. Use the top or bottom strake as a template by simply shifting it up or down 0.2“ accordingly. You might as well make all four top and bottom skins at once. GET THE FIRST ON RIGHT, OR YOU WILL HAVE 4 WRONG ONES.
3. I placed blocks under the ends of the landing gear struts to level the airplane perfectly and give it a solid foundation. I supported the jig table with two saw horses, and four pieces of 2x4’s individually clamped to the saw horses. The jig table simply rests on the ends of the 2x4’s that are adjusted up or down.
4. INSTALL THE RELEASE TAPE ON THE BACK OF THE JIG TABLE BEFORE ATTACHING IT TO THE MAIN SPAR.
5. Use long clamps from the top of the spar to the jig table beneath to firmly attach the plane to the jig table, inboard and outboard on the spar. I did not use any bondo and the jig table never moved relative to the plane, even when getting in and out of it repeatedly to cut out the openings and tape the inside joints.
6. I believe the strainer inspection hole in B33 is marked wrong in the plans. The inspection hole should start 6“ from the spar, not 3“.
7. For B33 bulkhead reference I clamped a straight edge to the canard bulkhead and strung a string back to a mark on the top of the spar. For the B57 bulkhead I simply used a plumb bob hung over the edge of the jig table at BL 57 and attached the other end to a mark at 57“ from centerline on the spar.
8. The B.L. dimensions are to the INBOARD face of the bulkhead. Carve the outboard to get the contours of the bulkheads to match the sweep of the strake. DO NOT CARVE THE INSIDE EDGES.
9. String a line from the leading edge of the template you made in step 1 above to the fuselage using the leading edge of the bulkheads as a guide. This will ensure the leading edges of the wing and strake come together.
10. 1 inch spacing is fine for the scoring of the bottom foam to bend around the ribs.
11. On the top foam, 1/2 inch cuts minimize the flat spots which occur on the top of the curved part of the strake between the two bulkheads. Make the cuts just after the 1 ply layup has cured. If you can make the cuts while the fiberglass is hard but still pliable, you can bend it over the ribs and let it finally cure in its final shape.
12. I fully trimmed the strake openings into the fuselage before putting the top on permanently. If you jig the top foam sufficiently, it will match up repeatedly to the same place on the fuselage.
13. I wanted to make sure my upper skin was firmly attached to the ribs so I made a flange to go on the bulkheads instead of simply stacking flox on the bulheads. The flox method probably works just fine, but I have heard of some Cozys developing fuel leaks after time and thought this might minimize the possibility of this occurring. This modification added several days work to the process! When the top foam is all ready to flox to the bulkheads, I carefully marked the location of the bulkheads on the bottom of the top strake skin. I then turned the top foam over. You only need to do this for the interior fuel tank edges. Do not extend the flanges into the storage area. You can tape those areas like normal. Do the leading edge, but not the trailing edge along the main spar. On top of each bulkhead marking I put a wide layer of box sealing tape, followed by a layer of peel ply, one layer of bid tape 2“ wide, and peel ply on the outer edges of the 2“ bid tape. On this second layer of peel ply, DO NOT PUT THE PEEL PLY WHERE THE BULKHEAD OR RIB WILL SIT. Err on the cautious side when putting on this peel ply. You can always sand the area not peel plied, but it is difficult to remove peel ply between the flange and bulkhead. Place a thin layer of flox on the bulkheads and place the top on to cure. Let it fully cure for at least a day or the one layer of bid will sag after you pull the top back off. Pop the top off and remove all the peel ply. Glass the interior corner of the new flange with one layer of bid tape, let cure, trim with dremel, and “voila”, flanges that perfectly match your strake skin.
14. When you place the top skin to the bulkheads and ribs, realize that it is a ton of work to tape all the inside edges. But doing the top skin placement and inside tapes in one sitting will save lots of sanding while upside down in the back of your airplane.

Owners Manual

Updates from Newsletter #52

From Cozy Newsletter #52:

Do not fly until you add the following information to the Takeoff performance section of the Owners Manual, and comply:

High Density Altitude Takeoffs

The combination of high aircraft gross weight and high density altitude represent significant dangers for takeoff obstacle clearance. Special care is required to avoid premature rotation, i.e., if liftoff is too slow, the aircraft will be on the back side of the power curve and may not climb. When operating heavy and high (say, within 100 lbs. of gross weight and above 5,000’ density altitude) do not fully rotate to liftoff attitude until your airspeed is within 5 kts. of the best rate of climb speed, for your specific weight and altitude (see climb charts). This will require more runway than a slower liftoff, but will assure the best capability to clear obstacles and continue a safe climb. Never attempt takeoff under conditions in which you cannot achieve best rate of climb speed while still on the available runway. If this ability is not clear at any point during takeoff_abort. Off-load weight or wait for a cooler time of day. Lift-off is possible as slow as the minimum lift-off speed, and can be successfully used at light weights and/or low altitudes to achieve a short ground roll. However, that technique will usually result in inadequate initial climb if used when heavy or high. Runway slope effects are minor when light or at low altitudes, but they become very significant when heavy/high. For example, a 1% uphill runway slope may add well over 1,000 feet to the distance required to clear an obstacle. Never take off uphill when your takeoff toll performance is marginal. Never continue a takeoff if crosswinds require you to brake so much that a safe liftoff is in doubt. Always use best power mixture for high altitude takeoff conditions. An over-gross weight takeoff that seems like an acceptable operation near sea level can be a real killer when hot and high. Never attempt a takeoff when over approved gross weight!

There may be considerable variance in takeoff capabilities from one homebuilt aircraft to another of the same type. Engine installed power and propeller efficiency at low speeds may be less than that for the prototype that provided the basis for the takeoff distance charts. Find a long runway and measure your takeoff capability at the weights you intend to fly. If your actual performance is less than the charts, correct the charts or improve your prop and/or engine.

Updates from Newsletter #56

From Cozy Newsletter #56.

Page 46, ENVELOPE EXPANSION: Add this:

Caution:

Before expanding your flight envelope, you should make sure your airplane has positive pitch stability; i.e., if you trim it for level flight and then either pull back on the stick or push it forward, upon release of the stick, it should return to level flight in no more than 3 oscillations.

Also, before expanding your flight envelope, you should double check that your canard is set at the right angle of incidence and/or your c.g. calculations are correct by comparing elevator position with the flight test data for our plans model shown below. If your elevator position differs by more than 1 degree, land and re-check your canard incidence using template “G”, and re-check weight & balance and c.g. calculation. Do not fly until you have determined the problem and corrected it.

• Page 19, Stall Characteristics, 2nd para.: After “shortening the canard span 6”, add “to a resultant tip to tip span of 151”.
• Page 19, Stall Characteristics, 3rd para.: Instead of “plans”, substitute, “1st edition plans”. Page 19, Stall Characteristics, end of 3rd para.: Add, “This applies to builders with 1st edition plans. The canard span has already been reduced on 2nd edition plans to 151” tip to tip.
• Page 25, under recommended entry speed for Stalls, after “Slow deceleration” add “from level flight only”.
• Page 25: Delete “Accelerated Stalls” as an approved maneuver.
• Page 33, Canard incidence: Delete “templates B & C” (a holdover from the 3-place Owners Manual), and change to “template G”.
• Page 33, Canard incidence: Change tolerance from +_ .3 deg. to + .6 deg and - .0 deg.

Updates from Newsletter #64

• Page 35, Main landing gear: Check wheel camber. When aircraft is empty, wheel camber should be between 2.5 to 3.0 degrees negative (bottom of wheel tilted in). Measure after pushing aircraft forward at least 5’, then aft 5’ and take average. Record in aircraft log book.
• When checking for proper toe-in, wheel camber should increase slightly when pushing forward and decrease slightly when pushing aft.
• Page 50, Annual inspection: Check wheel camber and toe in as prescribed on page 35 and record in aircraft log book.

Updates from Newsletter #65

• Page 35, Main landing gear: Check wheel camber. When aircraft is empty, wheel camber should be between 2.5 to 3.0 degrees negative (bottom of wheel tilted in). Measure after pushing aircraft forward at least 5’, then aft 5’ and take average. Record in aircraft log book.
• When checking for proper toe-in, wheel camber should increase slightly when pushing forward and decrease slightly when pushing aft.
• Page 50, Annual inspection: Check wheel camber and toe in as prescribed on page 35 and record in aircraft log book.

Mark IV Part Weights

David Domeier (unfinished)

As of Cozy Newsletter #50, this wasn’t all of the weights.

Paul Kuntz (unfinished)

Flying advice

• When filing for IFR, use the designator CZ-10. (Cozy Newsletter #36).
• Keep a fire extinguisher in the plane.
• When filing flight plans, the correct designator for the Coze 3 and 4 place aircraft is HKB (Cozy Newsletter #43)
• As of Cozy Newsletter #51, use COZY as the designator when filing flight plans.

Flight Testing

• One of the tips in Cozy Newsletter #51 suggests AC-90-89A, Amateur-Built Aircraft and Ultralight Flight Testing Handbook. The current version is AC-90-89B

Last updated: 2023-09-22 22:19:28 -0700

Canard Incidence vs. Pitch Stability

Copied verbatim from Cozy Newsletter #72. Make sure this is done early in flight testing:

The Cozy Mark IV is designed to have positive pitch stability and to be resistant to a main wing stall throughout its approved c.g. range. But for this to be true, the canard must be set at the correct angle of incidence. During your initial flight tests, you should determine whether your canard is set at the correct angle of incidence, and there is an easy way to do this safely in flight. If you notice on large drawing M- 18, there is a protractor which shows the full travel of the elevator from minus 15 degrees at full forward stick to plus 30 degrees at full aft stick, and there is a notation that at cruise, the elevator should be in trail, i.e. at zero degrees. In newsletter #56, page 5, there is a plot of elevator position as a function of c. g. and speed for our plans-built model, and on this plot, the zero elevator position occurs at a c.g. of 101 and 150 knots IAS. For several years now, we have been including a copy of this page with each Owner’s Manual. It also lists several changes for you to make to the Owner’s Manual. So, before you expand your flight envelope, and before you fly at slow speed and aft c.g., check your elevator position. It is perfectly safe to do this at a c.g. of 101 and 150 kts IAS, or even better, at a c.g. of 100 and 150 kts IAS, your elevator position should be at less than minus 1 degree, i.e. about 1/16th inch trailing edge down. If your canard is set at too low an angle of incidence, it is a dangerous condition. The indication will be that the trailing edge of the elevator will be too low in cruise. The symptoms will be difficulty in rotating, pitch instability, and a tendency to react too fast to back stick. And the result could well be a main wing stall.

Very recently two builders complained of pitch instability, and difficulty rotating. I went for a demo ride with one of the builders in his Mark IV. With both of us in the front seat (and estimated c.g. of 100), the elevator trailing edge was down almost 3/8 inch or 5 degrees. For comparison, we both went up in our plans-built Cozy and the trailing edge of the elevator was zero to 1/16th trailing edge down. We concluded that his canard was set at too low an angle of incidence. After a telephone discussion with the other builder, he sent me this e-mail: “Hi Nat, Thanks for your input the other day. I adjusted the canard incidence by 2 degrees. Did so by relocating the guide bushings that engage the alignment tabs, as we discussed. Had to slightly modify the canard cover, but it was really much less of a task than I’d envisioned. Bottom line is that the aircraft now flies lovely all the way out to the forward c.g. limit. The c.g. and weight envelope is fully expanded as well as all flutter testing and low speed/min speed tests. Cooling is pretty good though I’m still working on it. Will provide more soon. Regards, Rob”

This is an important matter. Make sure you check elevator position early in your flight test program.

Last updated: 2021-08-02 18:00:57 -0700

Cleaning Up Epoxy

From Cozy Newsletter #83. Very good information. The main takeaways are:

• Use barrier cream, replace it regularly.
• Minimize the use of organic solvents as much as possible.
• Use cheap paint brushes you can throw away.
• Wipe squeegees with a paper towel. Let remaining epoxy cure, then sand it off with a belt sander.
• Scissors: Wipe with a paper towel, then scrape away epoxy with a straight edge razor.

In the handling of chemicals, we urge caution. Epoxy resin, by itself, is inert and nonreactive. But the hardener has active ingredients, which you can become allergic to if you are not careful. So we recommend the use of barrier creams on your hands which are water soluble (hydrophilic) which you can wash off with soap and water, but are not penetrated by organic materials (oleophobic). Now when it comes to cleaning mixed epoxy from hands and tools, a word of caution. It is best to try to hold the use of organic solvents to an absolute minimum. MEK (methyl ethyl ketone) is a very powerful organic solvent. It will certainly remove epoxy from your skin and tools, but it can be deadly! It can penetrate your skin, and inhaling the vapors can also cause damage to internal organs. Acetone is a cousin of MEK. It isn’t quite as bad, but can still cause damage. So it is best to avoid the use of solvents as much as possible. How do you do this? Use barrier cream on your hands, and try not to soak your hands in epoxy, and replace the barrier cream quite regularly. Wash it off with soap and water. You can use gloves, but not latex gloves, because they can also cause allergies.

When it comes to tools, for paint brushes, use a cheap variety that you can throw away. If you insist on using them more than once, washing with hot water (really hot) and soap works very well. You can store them in a large bottle with a little solvent for later use, or you can wrap them in plastic and put them in your freezer. With squeegees, wipe them with a paper towel. Rather than using solvents, let any remaining epoxy cure and then sand it off with your belt sander. With scissors, you can wipe them off, and periodically scrape any cured epoxy off with a straight edge razor. Builders report that apple cider vinegar cuts epoxy very well on hands or tools, but we have never tried this. Vinegar is acid, so it should be neutralized or flushed later.

Last updated: 2021-08-02 18:00:57 -0700

Main Landing Gear Strut

From the Cozy Newsletter #78:

The FeatherLite main landing gear strut for the Mark IV is halfway between that of the Long EZ and Defiant in span, chord, thickness, weight, cost, and load carrying ability. It is made of S-glass and very dense, with no voids. It has an excellent spring-constant, and will not deform or take a permanent spread after repeated loading to many gs, nor should it be necessary to “set” the gear after parking. In newsletter #64-6 we explained how to measure wheel camber and suggested checking it annually to determine that no permanent spreading of the strut has occurred. We have yet to hear of a FeatherLite Mark IV strut taking a permanent set.

The strut was purposely designed with a return curvature at both ends to reduce local bending loads. This also results in the wheels having a slight camber (bottoms closer together) of about 2.5 degrees when the Mark IV is empty or slightly loaded. As the loading is increased, the strut spreads and the wheels move outboard and become closer to vertical. As the loading is further increased, like in a high-g landing, the strut has been known to spread enough for the inside of the wheel pants to scrape the runway, and in very extreme cases, spread enough for the brake calipers to scrape the runway. But always the strut should return to its original shape after the load is relieved.

Tire wear is always the greatest at the instant the tire touches the runway, because wheel rotation is instantly accelerated from zero to about 70 knots. Because of the camber of the wheels, which is even greater when the airplane is in the air just before touchdown, the outboard tread receives the greatest wear. So when the outboard tread is nearly worn away, it is the accepted and recommended practice to reverse the tires (outboard to inboard) to even out the wear, similar to what you do with your automobiles when you rotate the tires. Before installation of the strut, you are instructed to wrap it with 8 crossing plies of UNI, to increase the torsional strength. This should eliminate or at least minimize any possibility of wheel shimmy. You should not substitute BID, Bi-Ax, or Tri- Ax cloth.

You are instructed to install the axles with a slight toe-in, about 1⁄4 degrees. This will provide the least rolling resistance in takeoff and landing, and least tire wear. When you think about it, the reason is simple. During takeoff, as the load is relieved from the wheels, the strut tries to come together, and toe-in helps it do that. When you land, the sudden load (unless you “grease it on”) spreads the strut and then as the load is relieved, the toe-in helps it come together.

Last updated: 2021-08-02 18:00:57 -0700

Peel Strength

From Cozy Newsletter #55.

The peel strength of a cured fiberglass matrix is the amount of force required to delaminate a layer of fiberglass from the layer below or the substrate below. It is the measure of the internal strength of the cured epoxy or its adhesion or mechanical bond to the substrate. It is a very important property, because the peel strength of the epoxy is literally what is holding your airplane together.

It should be obvious that when you are making a layup over a dry, previously cured surface, the peel strength will be either the strength of the mechanical bond, or else the internal strength of the epoxy, which ever is weaker. It is important, therefore, in making a wet layup over a dry surface to properly prepare the dry surface so the mechanical bond to the surface will not be the weakest link. There are two methods of preparing a fiberglass surface for the maximum strength mechanical bond. The first is to sand a previously cured fiberglass surface dull with 36 grit sandpaper. The disadvantage of this method is that if the surface is not absolutely flat, it is nearly impossible to remove all of the shine without sanding through some of the glass filaments, weakening the substrate . The second method is to squeegee peel ply over a fiberglass substrate before it cures, and strip it off after cure. The advantage of using peel ply is that the shiny surface (which might also be waxy) is stripped off with the peel ply, and the surface remaining is flat but fractured and rough. Sanding this surface is a wise, extra precaution to maximize the mechanical bond.

It is normally assumed that maximum peel strength is obtained by making a wet-on-wet layup, because then there is no mechanical bond involved. There may be an exception, or at least a reservation to this rule. We recently received a call from a builder in the state of Washington who said that he was using Epolite 2427 resin, and he noticed after glassing the inside of his fuselage sides, that the peel strength of the second layer was poor, and he could peel off the second layer rather easily. After discussing his technique with him, we think we know why this happened. Very often, first-time builders are too meticulous, and take a long time to squeegee each layer of glass to remove all of the excess epoxy before laying down the next layer of cloth. The fuselage sides have a lot of area so this could take quite a long time. If, as has been alleged, 2427 is more susceptible to contamination by humidity (and/or carbon dioxide) in the atmosphere, and it is exposed to the atmosphere for a long period of time before the next layer of glass is applied, there could be a contaminated surface between the two layers of glass which would reduce the internal strength of the epoxy at the worst possible location – between the glass layers.

Experienced builders have learned that it is much faster, and results in better layups, to wet out the first layer of glass, squeegee the air out, but leave an excess of epoxy and then lay down the next layer of cloth on top to soak up the excess. This wets out the second layer faster, because the excess epoxy from underneath pushes the air out ahead of it. This saves much time, results in less air in the layup, requires less squeegeeing, and, if the epoxy is susceptible to contamination from the atmosphere, there is much less exposure to the atmosphere, and any contaminated surface epoxy does not end up being between two layers of glass. The same considerations apply also and argue for the use of peel ply over the top layer.

We have tested the peel strength of Epolite 2427, and haven’t found it to be any different from the several other epoxy systems we have used. It is true that we have low humidities in Arizona most of the year (not during the monsoon season, however), but it is also true we follow the procedure for faster and better layups recommended above.

Last updated: 2021-08-02 18:00:57 -0700

Post Curing of Epoxy Layups

From the Cozy Newsletter #61

All epoxy systems have a minimum and a maximum curing temperature. At the minimum curing temperature, for example, room temperature, an epoxy system achieves a good degree of curing, that is, over 80%, and produces a hard, shiny, tack free surface, with good properties, but the properties, including glass transition temperature, are not maximized. For both L335 and L285, the properties achieved from room temperature cures are quite acceptable for aircraft use. The glass transition temperature of L335 after a room temperature cure, would be about 130 deg. F (the glass transition temperature is the temperature at which the epoxy begins to soften and become rubbery). To obtain 100% of the available properties and maximum glass transition temperature requires either post curing or curing at the maximum curing temperature. For L335, the maximum curing temperature or post curing temperature would be about 160 deg. F and results in a glass transition temperature of about 185 deg. F. But to achieve maximum cure and glass transition temperature, the layup would have to be held at the maximum curing temperature for about 16 hours. For L285 the maximum curing or post curing temperature would be about 170 deg. F and result in a maximum glass transition temperature of about 240 deg. F. Again, it would have to be held at the high temperature for about 16 hours.

So is it necessary to post cure your airplane? Actually, no. As either parts of your airplane or the entire airplane sit at temperatures above 70 deg. F, some post curing occurs. Even if your airplane sits outside in the sun, and the ambient temperature reaches 100 deg. F, it will do so gradually, and the post curing will proceed gradually so that glass transition temperature will always be considerably higher than ambient. The wings will not sag. Remember, you are going to paint your airplane white, and the surface temperature (only the top surface in the sun) will only get about 10 degrees higher than ambient. So on a 100 degree day, some post curing will already have occurred, so the glass transition temperature will already be higher than 130 deg. F and approaching the maximum of 185 deg. Now there is a possible exception to this rule. If you are going to paint your airplane black, or some other dark color, a fully post-cured L335 might not have a high enough glass transition temperature (185 deg. F might not be high enough). So, if you intend to paint your airplane black or some other dark color, you probably better use an epoxy like L285, and do some post curing before you park it out in the sun on a hot summer day. Remember, with an epoxy like L335, the glass transition temperature will always be higher than ambient and the epoxy will not soften unless or until the surface temperature reaches the maximum glass transition temperature, which in the case of L335 is about 185 deg. F. If you paint your airplane white, at least on the top, the surface should never reach this temperature even outside in Saudi Arabia in the summer. Even if you are not going to paint your airplane a dark color, but are willing to spend a little extra money on the best epoxy available, consider using L285. That would be our choice. If you do not understand, or have further questions, please contact us.

Last updated: 2022-03-26 21:49:07 -0700

Unapproved Design Changes

From Cozy Newsletter #59, copied verbatim.

We referred earlier to design changes which we haven’t approved, which we think are ill-advised, and which in some cases may even be dangerous. Here are design changes to avoid:

1. Do not change the way the elevators are constructed or installed. The top surface of the elevators must be reflexed, as we show in the plans, as for example in the cross sections we show in Chapter 11, pages 5 & 6. If yours have not been built this way, scrap them and start over. A false rumor has been circulated that there is a “dead spot” in elevator response. This is not true if built according to plans, and we have proved it and can demonstrate it with our “plans-built” Cozy Mark IV. It has been reported that the same person circulating this rumor has a problem with pitch trim. This could be a symptom that there is something more seriously wrong with his airplane, if not the shape of his elevators, it could be that his canard is not set at the correct angle of incidence, and his airplane is unstable in pitch. Our flight tests apply only to a Mark IV built according to our plans. You should be aware that the fuselage is a lifting body, and the canard span and incidence needs to be adjusted for the amount of lift contributed by the fuselage. If you change the shape of the fuselage, either by making it wider, higher, or using a different canopy, or you change the angle of incidence on the main wing, there is only one way to determine whether you have a safe airplane, and that is to install a traveling 135 lb. weight, like we did, and go up to 11,000 ft. with a parachute, and determine whether the point of neutral stability is well aft of the aft c.g. limit, and whether the c.g. range is safe at the aft c.g. limit and at least 1.2 inches beyond.
2. Do not limit elevator trailing edge up (nose down) travel to 11 degrees rather than the 15 degrees we specify. If you should ever remove your vortilons, or miscalculate your c.g., or set the canard at too low an angle of incidence, or change the fuselage, and do stall tests and find the main wing start to sink at the same time the airspeed starts to fall, you will want as much nose down authority as you can muster to keep from becoming a statistic. The plans say 15 degrees trailing edge up, and that is what you should have.
3. Do not use “hard shelling” (Note: applying a mixture of micro to the foam, letting that cure, then applying glass). This results in a poorer bond between the fiberglass and the core. On airfoils, the top surface is usually in compression. A poorer bond means that the airfoil (canard or wing) skin can delaminate, buckle and fail sooner, that is, fail at a lower g-loading, than an airfoil built according to plans.
4. Do not use “tri-ax” cloth. We are not aware of any static load testing that has been done with Cozy or Long EZ wings to prove that it has as much torsional, tensile, and compressive strength as the individual layups specified by Burt Rutan, and which we use. Furthermore, it is much harder to work with because it does not conform as well to curved surfaces and can cause air bubbles (poor lamination), and it has a tendency for one or more layers to wrinkle when squeegeed, which it is not possible to correct. Also, our plans call for one less layer of glass where the ailerons will be cut out. You can’t do that if you are using tri-ax. Ailerons covered with tri-ax cloth will not balance leading edge down, with the very likely possibility of “flutter” which can cause catastrophic failure of an airframe, unless you sand off at least one layer of glass, which is more work than if you had used the correct glass cloth layups in the first place.
5. Do not increase the chord length of the ailerons. There is no possible justification for doing this. It not only makes it much more difficult to balance the ailerons, but it reduces the wing cross section at the point where the bending load is the greatest. In other words it weakens your wings both in drag and in bending loads.
6. Do not increase the span of the ailerons. Our 3-place Cozy has excellent roll response. The Mark IV has even greater roll response because the 2 ft. greater wingspan puts the ailerons farther outboard which results in a greater roll moment, and also farther outboard of the downwash from the canard. There is no possible justification fo this change.
7. Do not buy “fast build” wing kits where the foam cores have been cut for wider chord and greater span ailerons, and the “kink” in the trailing edge of the wing has been eliminated. The latter changes the angle of incidence of the airfoil, which makes it different from the airplane we have tested, and causes problems in fitting the cowlings.
8. Do not eliminate the extra layups on the top of the wing which we show in Chapter 19, page 6, Fig. 32. These provide added strength to the wing where the bending and drag loads are the greatest.
9. Do not increase the span of the rudders. They are even more effective on the Mark IV than on our 3-place because of the increased wingspan. Very little rudder input is required in flying a Mark IV. Furthermore, it was discovered on the Varieze that too much rudder authority could cause a Varieze to lose control. You wouldn’t want this to happen on base or on final.
10. Do not leave the lower winglets off or make them smaller. We found in our very thorough aft c. g. flight test program that the lower winglets shown in our plans give you a 0.5“ c.g. safety margin protection against a main wing stall. They also provide more roll stability at slow speeds and high angles of attack. Again, something that is important on base or on final.
11. Do not eliminate the NACA scoop in the fuselage bottom and use arm pit scoops instead. The NACA scoop gives better cooling of the engine when parked nose down, and probably also in flight. Better cooling of the carburetor or throttle body means fewer problems (if any) with hot starts. Also, if you notice, the Cozy Mark IV is quite clean underneath. We believe this (and better cowling contour) contributes to the extra 20 mph we get as compared to another aircraft with arm pit scoops, fixed gear and the same horsepower.
12. Do not cover the top of the canopy. The compliment we receive most often on the Cozy and Cozy Mark IV is its wonderful visibility. If we had gull-wing doors, we couldn’t have a full bubble, but with a side-hinged canopy it is a wonderful feature. Install the full bubble first. Then, if sun bothers you (it helps to keep you warm at high altitudes in the winter), you can always wear a baseball cap. Then, if you are still bothered by the sun, you could use cling-type plastic shades, which are moveable, over your head. If you really like reduced visibility, you can always paint part of the bubble. The latest pictures we have seen of the Velocity show that they removed most of the cover over the top of their canopy. We consider this to be a safety consideration.
13. Do not hook your fuel tanks together. Having two separate tanks is a safety feature. If you lose one fuel cap with both tanks hooked together, all of your fuel will be siphoned out, and you will become a statistic.
14. Do not eliminate the sumps under each tank. They make all of your fuel usable, and allow you to fly for a little while after the gauge shows empty. Although not recommended, you can run one tank completely dry, switch, and keep on trucking. We have done this (accidentally) twice in the last 12 years.
15. Do not locate your fuel selector valve remotely. Experience with the Varieze was that the remote location of the fuel valve was the cause for more than one emergency landing.
16. Do not buy components from custom shops without calling us first. Custom shops are in business to make money. They may use unskilled labor and take shortcuts, and sometimes their parts don’t fit or are unairworthy. They will probably claim their parts are better, but then ask you to sign a waiver relieving them of any product responsibility or liability, and will not guarantee the airworthiness of parts they supply. We have heard of some very bad experiences some builders have had with custom shops. If you are not willing to build the parts yourself per plans, you would be better advised to build a Velocity. The Swings have a pretty good record.
17. Do not install retractable main gear. Our airplane was not designed for it. Most people agree the benefit (if any) isn’t worth the cost (and we aren’t just talking about dollars).
18. Do not change the size or shape of the engine cowlings (in other words, use the cowlings we approved from Feather Lite). Our cowlings were contoured to have nice clean airflow back to the prop, to avoid a vacuum or reverse airflow at the prop. It was interesting to hear that one 4-place canard aircraft (we won’t mention the name), which has been highly advertised, and which has a very blunt rear end, has not yet been able to meet its performance goals of useful load and speed because it cannot generate enough thrust after becoming airborne. We suspect that it is because the prop is spinning in reverse airflow or a vacuum (you know, like behind a semi).

[A bunch of text to backup that unapproved design changes tend to have bad results]

Lastly, if you do not build your airplane according to plans, the Cozy Mark IV Owner’s Manual does not apply, and we ask that you not register or insure it as a Cozy Mark IV!

Which is why I’m not building a Cozy Mark IV, I’m building a Coz-E, which happens to be largely similar to the Mark IV, but is not the same airplane.

Last updated: 2021-08-07 08:24:43 -0700

How do you make Vortilons

From Cozy Newsletter #77:

A way that works well is to make a flat 3-ply 8 x 8 layup on a piece of plastic, and then cut out a pair of each of the 3 vortilons (see M-18). Then tape saran wrap on the leading edge of the wings at each of the 3 locations. Use a generous amount of saran. Then put down a 2-ply layup over each of the saran wraps on the leading edge of the wing. Make it generous, and use peel ply over the top. Sand the vortilons. After the leading edge layups have cured, remove the peel ply and 5- minute the vortilons to the leading edge layups, using the 67 degree template on M-18 to make sure they are at the right angle. After all are in place with the right alignment, micro the inside corners and tape both sides of the vortilons to the flanges with 2 plies of BID, and peel ply. After cure, remove the flange-vortilons from the leading edge, and trim the flanges to about 3⁄4” width each side of the vortilon. Paint them. After the wings are painted, stick them on with clear silicone. They will adhere without coming loose even after many head bumps. Put them on before you fly and never take them off!

Last updated: 2021-08-02 18:00:57 -0700

Resources and Inspiration

Links on building a gas-powered Cozy Mark 4

• The unofficial cozy builders web site, run by Marc Zeitlin.
• Cozy Girrrls makes and sells parts for the Cozy and other canards.
• John Slade’s page on building his Cozy.
• Phillip Johnson’s Kinda Kozy has a subaru engine, fully retractable gear (which makes the cozy look so much better, even with the retracts deployed, a front hinge canopy w/ gull wing passenger entrance.
• Cozy Serenity has a really detailed, modern build log. Definitely standing on the shoulders of giants type deal.
• Eureka CNC - this person built a CNC hot wire machine, and sells jigs that are CNC’d, or fully CNC’d out foam cores.

Finishing

• Wayne Hicks’ Finishing Page
• From Cured Composites:
  • Finishing a Composite Airplane

Last updated: 2021-06-23 22:19:06 -0700

Shop Setup

Here’s how I set up my shop for this:

You’ll need a lot of workbench space. I designed my own 2’x4’ workbench, and plan to build 6 of them.

Additionally, you need a bunch of storage space. I built a cabinet to store fiberglass in, which has space underneath it. I store store epoxy supplies underneath it - extra resin/hardener, flocked cotton for flox, microspheres for micro, etc. I also have a 3d printer under there, which is currently unused in this project. Though, I’ll probably 3d print some jigs.

I’ve also built a nice box where I have a container each of epoxy resin and hardener, wrapped in a heating pad. When I’ll be doing work with epoxy, I turn on the heating pad in the morning, and by lunch the epoxy has heated up enough to be usable. The box is made by cutting half-inch plywood and joining together with glue and screws. Nothing special. The back of it has a cut out for the heating pad + wire to go through, and there’s a door on the other side, which friction-fits closed (there’s no latch to keep it closed). It’s not great construction, but it’s fine for the purpose.

Future known work includes:

• More storage. You will always need more storage.

Additionally, I also have some printouts of information good to have visible in the shop.

Last updated: 2022-05-30 19:12:12 -0700

Automated Epoxy Dispenser

When you use a traditional epoxy pump, there’s no way to tell if it actually dispensed in the correct ratio until the epoxy has cured. If there’s some kind of blockage in one of the pumps, you won’t find out until the epoxy has cured, and you find out it didn’t cure correctly. Because of this, you need to calibrate the pumps regularly.

Still, fear of this is why I elected to use a scale and manually pour epoxy using a scale to measure it. However, manually pouring epoxy is a bit of a skill. Not a hard skill to learn - I had basically mastered it by the fourth or fifth time I mixed epoxy. But it’s a bit annoying and time consuming. Certainly, pulling levers for epoxy pumps is significantly easier to do, and it takes significantly less time to do so. I wanted something even easier, even if it’s still not as fast. As a wannabe electrical engineer, I decided to build an automated epoxy dispenser.

So, I bought some peristaltic pumps and some other components, and I made the simplest “press button, receive epoxy” program I could think of.

Current State

The epoxy dispenser I built works very well. With the press of a button (the green button on the box on the right), it starts dispensing. Currently, it’s set to dispense 120 grams (about 4 oz) total of epoxy and hardener. While it does take quite some time to dispense (approximately 10 minutes), it’s worth it to not have to manually pour epoxy. This would be an issue if I used epoxy with a faster working time, but the MGS 335 system I use has a working time measured in hours.

Hardware

This uses 2 peristaltic pumps to dispense epoxy. Wikipedia has a pretty good article about them, but peristaltic pumps work by squeezing a plastic tube to force fluid through it. This makes them great for a ton of applications where you want to pump something, but you don’t want to expose the pump itself.

These pumps are controlled by an arduino clone, using an L293D motor driver chip. This is a common driver chip and it lets me to control a motor from a low-amperage-output control pin.

On the sensing side, I have a basic load cell hooked up to an HX711 amplifier chip.

The arduino clone reads the load cell (via the HX711 chip), and then uses that to determine which pump to run, as well as what speed to run it at.

Program

The control program runs at a rate of 10 hz. Each loop, it samples the scale to get the current total weight, and runs that through a PID loop to determine how quickly to run the chosen pump. Additionally, it also updates the character display with information about the current state of the system, as well as how much of resin or hardener it has dispensed.

On a higher level, the program is a simple state machine: It boots up into the IDLE state, where does nothing but take measurements from the load cell and display them on the character display. This makes the program useful for manually pouring epoxy, on the occasions I need to do that. When you press the green button, it changes to the RESIN state, where it runs the resin pump. Once the controller has detected that enough resin has been dispensed, then it switches over to the HARDENER state, where it turns off the resin pump and runs the hardener pump. Once the controller detects that we’ve dispensed enough hardener, it switches to the SHUTDOWN state, where it turns off all pumps and resets back to the IDLE state.

Issues

As much as I like this system, it’s not without issues. On the software side, I don’t have a way to pause dispensing without stopping and resetting. So if I run out of either resin or hardener mid-pour, then I need to either throw away what was already dispensed, or manually finish the pour. On a future revision, I may add a switch to allow me to pause without resetting. Another feature I’d like is the ability to specify an exact amount of either resin or hardener to be dispensed. Both of these are useful for solving the issue of running out of either material.

More worryingly than issues replacing the contains, the epoxy hardener I use appears to be able to permeate through the plastic of the pumps. Which makes me worry that it might lead to gumming up the pump, potentially breaking it or causing some other issue. This appears to be the same issue that @cyanoacry on twitter ran into when he tried this same idea, though he’s using a slightly different epoxy system. I’ve attempted to resolve this issue by using different tubing materials. At the moment, the I’ve gone through every tubing material I could find on McMaster Carr, and none of them completely solve the issue. The best I’ve come across is black rubber tubing, which lasts about 3 weeks before enough permeates through that I need to replace it. Thankfully, the epoxy resin does not appear to have this issue at all, and has been perfectly fine with the silicone tubing that came with the pump.

I’m willing to accept this and resign myself to replacing the tubing twice a month, that’s still plenty of time.

Future Work

I’d love to improve this and make it more versatile, some ideas for future work include:

• Redesign the system to have all components on a single board, instead of connecting different breakout boards.
• Add a UI for dynamically configuring different mixtures of epoxy.
• Add support for another motor, so that I can mix fast and slow hardeners for a configurable pot life.
• Add buttons for shortcuts for different predefined mixtures.

Last updated: 2022-08-28 10:04:36 -0700

Fiberglass Cabinet

Unlike the workbench, this is joined mostly with butt-joints. This doesn’t need to support a ton of weight, so butt joints held together with glue and screws is perfectly fine for this.

Bill of Materials

First, the actual materials and lengths needed to construct this. This is made primarily with 2x4 studs (which are actually 1.5 x 3.5 inches, because of course they are).

2x4 studs:

• 2x 60“ (for the backplate)

• 2x 33.5“ (for the front support)

• 2x 30“ (Giving extra depth for the fiberglass container)

• 6x 48“ (4 for the “rails” supporting the front and back of both the top surface and the bottom shelf, then 2 for the top part, covering the fiberglass container)

• 6x 24“ (the “ribs” of the top surface and the bottom shelf)

Plywood. half-inch or even quarter-inch is fine.

• 2x 48“ x 27“ sheets (The shelves)
• 1x 48“ x 31“ sheet

Extra:

• Screws. I used 3 inch #8 screws for most of the frame. For securing the fiberglass container, you’ll want longer screws, 5 inches or so should work.
• a single 48“ x 24“ (or so) sheet of whiteboard.
• Bunch of wood glue.

Raw Materials

Because you won’t be able to find individual pieces cut to the exact size, these are examples of how you can obtain the required materials in the necessary lengths.

The “remaining material” assumes the cuts are made with a 1/8 inch blade saw blade.

Studs

• 2x studs, cut to 60“ and 33.5“ (2.25“ remainder)
• 2x studs, cut to 48“ and 30“ (17.75“ remainder)
• 2x studs, cut to 48“ and 48“ (0 remainder - and it’s ok that one is going to be slightly short)
• 2x studs, cut to 48“, 24“, and 24“ (0 remainder. It’s ok for the short side to be on one of the 24“ cuts)

In total, 8x 8’ studs.

Plywood

From a single 4’x8’ sheet, you can cut it length-wise into the sizes specified.

Plans

1. Cut studs to lengths specified
2. Create the rear-frame by joining via butt-joints the 60“ studs, and 3 48“ studs.
3. Attach the studs for the fiberglass container
4. Attach via butt-joints the “ribs” of the shelves to the rear-frame.
5. Attach the front “legs” as well as the rib to the frame
6. Attach the plywood to the back and the shelves.

Last updated: 2021-07-26 20:57:50 -0700

Printouts

Notes to print out and place in a visible place in the shop.

Here we have:

• Materials is good info on making epoxy, with QR codes for the epoxy and glass bubbles MSDS.
• Basic Layup Procedure is more-or-less a transcription of Chapter 3, page 19. I didn’t want to rip it out of the plans and hang it out on the wall.

Last updated: 2021-09-03 19:22:16 -0700

Basic Layup Procedure

1. Preparation: Ply #9 or Gloves (Nitrile preferred) on hands, shop temperature 75° F ± 10 °. Don’t do the layup unless you know the shop will stay within this range during the entire curing time.
2. Cloth Cutting: Use the electric scissors.

• Unless otherwise specified, cut at 45° to the fibers.

3. Surface preparation:

• Foam: Hot-wire-cut surface needs no preparation. Sand ledges or bumps even. Fill holes or gouges with dry micro immediately prior to layup. Vacuum up the dust.
• Glass: Always sand to completely dull any cured glass surface with 36 or 60 grit sandpaper. Re-sand if it has been touched with greasy fingers.
• Metal: Dull with 220 grit sandpaper.

4. Mix Epoxy:

• Mix for 2 minutes.
• 80% stirring, 20% scraping sides and bottom.
• Hot cup: Throw it away and mix more. A hot cup indicates exotherm.
• Don’t use a brush to stir.
• Micro Slurry: Approximately equal volume of mixed epoxy & glass bubbles.
• Wet Micro: Add enough bubbles for a “thick honey” mix.
• Dry Micro: Enough bubbles so it won’t run.
• Wet Flox: Thick, but pourable mixture of mixed epoxy & flocked cotton.
• Applying to surface:
  • Over foam: brush or squeegee on a very thin micro slurry layer. (urethane foam: Use a thick micro slurry layer).
  • Over glass: Brush on a coat of epoxy.

5. Lay on cloth:

• Pull edges to straighten wrinkles.
• If working alone: Roll the cloth, then unroll it onto the surface.

6. Wet Out:

• Don’t slop on excess resin; Bring epoxy up with a vertical stab of the brush (This is stippling)
• Start in center and work out to sides.
• Most of the time in a layup is spent stippling. Stipple resin up from below, or, if required, down from above.
• “Not wet, not white”.

7. Squeegee:

• If you have excess epoxy, squeegee it off to the side. Use squeegee with many light passes to move epoxy from wet areas to dry areas.

8. Preliminary Contour Fill

• Save sanding by troweling dry micro over low areas while the glass layup is “green”.
• This is done at trailing edges, spar caps, or over any low areas.
  • The low places are overfilled with micro, then sanded smooth after full cure.

9. Knife Trim:

• Razor-trim the edges at the “green” stage, about 3-4 hours after the layup.
• Alternatively: Wait until a full cure, then use the variable oscillating tool to knife trim.

10. General Inspection:

• Look for dry glass, excess epoxy, bubbles, and delamination before walking away from wet layup.

11. Cleanup:

• Remove gloves, or rinse ply #9 off with soap and water.
• Epocleanse can also remove epoxy off unprotected skin areas.
• Brushes: Rinse twice with MEK, and wash with soap and water. Throw away after 2-4 uses.

Last updated: 2022-08-26 09:02:41 -0700

Materials

Safety

Last updated: 2021-10-12 21:29:46 -0700

Workbench

These are heavily inspired by the EAA Chapter 1000 Standardized Work Tables, only designed to to be 2 feet by 4 feet instead of 2 feet by 5 feet. A bunch of people in the cozy community prefer joining 6 tables of these size to ultimately create the plans-recommended 12 foot by 4 foot table for the wings and other large pieces. This is my take on that.

Version 1.4, April 2022

Starting on the fifth table, the primary goal is to simplify construction by removing the use of half-laps entirely. Here’s what I came up with:

• Extend the lip of the top frame to line up with the tabletop.
• Do away with resting the top frame on the legs. Screw the top frame into the legs.
• Increase the size of the bottom frame to match the
• Build the bottom frame into the legs.
• Use height-adjustable casters. Screw them into the bench using t-nuts.

Here’s image of what I came up with:

• USDZ (AR Preview)
• Fusion 360 file.

None of these previews include the wheels, because my CAD-fu isn’t good enough.

Materials

• 2 by 4 studs
  • 2x 48“ studs (the top frame rails)
  • 3x 21“ studs (the top frame spans)
  • 2x 38“ studs (the bottom frame rails)
  • 3x 18“ studs (the bottom frame spans)
  • 4x 28“ studs (the legs)
• Plywood
  • 1x 48“ by 24“ by 3/4“ plywood (the top surface)
  • 1x 45“ by 21“ by 1/4“ plywood (the bottom shelf)
• Extra hardware
  • 4x castering wheels with center screw
  • 4x t-nuts which fit those center screws
  • Many screws
  • Optional: Pocket hole jig + pocket hole screws (for securing things to the legs).

Instructions

1. Top Frame:

• Screw the top frame pieces together. Remember that the spans go inside of the rails.

2. Bottom Frame:

• Screw the bottom frame pieces together. The rails screw into legs, same with each edge span.

3. Placing the bottom assembly on the floor, place the top frame in place (the tops of the legs should meet at the 4 inner corners of the top frame). Make sure that the top of the top frame to be flush with the top of the legs. It’ll help to use scrap wood to hold the frame on the legs. Screw the top frame into the legs from both the rails and the edge spans.
4. Re-orient the frame to be upside down. Drill a hole in each leg, then put in a t-nut. Add a nut to the caster, then screw that in as well. The nut makes it a lot easier to lock in a height once you have the desired height.
5. Re-orient the frame back to normal. Screw the benchtop into the top frame.
6. Cut 1.5“ by 3.5“ tabs into the bottom shelf top, then slide that in to rest on the bottom frame. No need to screw this in, the legs of the table will keep it from moving around too much.

Version 1.3, March 2022

On the fourth table I built, I had what’s really the second meaningful change to the design of my tables This results in a larger bottom shelf, and a much easier build process. There’s still work to improve the build process and make that easier, which will be part of the version 1.4 design.

In this version, the top is still made using half-laps, but the bottom-shelf frame is joined using butt-joints and mostly screws.

Materials

• 2“ x 4“ studs
  • 2x 48“ studs
  • 3x 24“ studs
  • 4x 28“ studs
  • 2x 40“ studs
  • 3x 16“ studs
• Plywood
  • 1x 48“ x 24“ x 3/4“ plywood (you could also use 1/2“ plywood and it wouldn’t matter)
  • 1x 40“ x 19“ x 1/4“ plywood
• 4x 3.5“ tall castering wheels.
  • Get t-nuts and wheels with screws in them. This way, you have height-adjustable wheels that make aligning each table in height significantly easier.

As before, get large studs and cut them to size. You can make this with 6x 8’ studs. You can also use 10’ or 12’ long studs and cut them to size. Whatever will fit in your vehicle. Similarly, it’s much more economical to get 4’ by 8’ sheets of plywood, instead of project boards.

Instructions

1. Top Frame

• Cut half-laps into the 48“ studs. These will have a depth of 1.75“ (half the height of the studs). They should be 1.5“ long (the width of the studs). Cut 2 of them 4“ from the edges, then a third in the exact center of the 48“ stud (starts 23.25“ from either edge). Make these cuts all on the same side.
• Cut rabbets into 3 of the 24“ studs. These will be at the ends of the studs. As before, they’ll be 1.75“ deep, and 1.5“ long. Make these cuts all on the same side.
• Join the 48“ studs with the 24“ studs you just cut into. The correct orientation for this is that the 24“ will slot down into the 48“ (so that, when weight is placed on the 24“ studs, they’ll transfer that weight into the 48“ studs)

2. Bottom Shelf

• Butt together the 40“ studs and the 16“ studs such that the 16“ studs are in between the 40“ studs. Screw them together. Optionally, you can glue them, but honestly, that’s not necessary.

3. Legs

• The legs should be oriented so that the height of them (the nominally 4“/actually 3.5“ side) is parallel to the 40“ studs/the length of the table.
• With the bottom legs flush to the bottom of the bottom shelf, and extending straight up, glue and screw them in to the bottom shelf.

4. Wheels

• With the table inverted (resting on the top of the table), install the castering wheels + any shim material onto the bottom of the bottom shelf.
  • Make sure the caster wheels are installed so that most, if not all, of the weight from the top surface is transferred directly to the wheels. (If you get casters that have a single screw going up through them - which you should, because that makes adjusting the height easier - then predrill & install a t-nut going through any shim material and into the legs.)

5. Installing the Top Frame

• Revert the table so that it’s on its wheels. Lock them at this point (and also brace the table so that it still doesn’t move unnecessarily)
• You should be able to place the top frame directly on the legs at this point. It should just line up. Obviously, this isn’t a long-term solution.
• Using scrap pieces, add brackets to the outside of the legs. These will actually hold the top frame in place and keep it from sliding around. I used 2x4s, but if you have any scrap 3/4“ plywood, you could cut it to size and use that.
• Place the top frame back onto the legs, and screw the brackets into the top frame.

6. Surfaces

• Cut out rectangles in the corners of the bottom 40“x19“ sheet so that it’ll actually fit into the bottom shelf.
• Screw the 48“x34“ sheet into top of the frame. I added screws at where the 24“ studs intersected with the 48“ studs.
• Slide the bottom 40“x19“ into the bottom shelf.

Future Thoughts

Some thoughts I have on improving this one:

• I should entirely remove the need for the half-lap joineries. It would allow me to create a more precise table, as getting the depths of these cuts correct is difficult.
• I should use height-adjustable locking casters. This requires some additional hardware, but makes finely aligning individual tables possible. I’m going to go back and add these at a later date to my tables, where possible.

Because of both, expect to see a version 2.0 sometime in the future, as I create additional workbenches.

Initial Version, July 2021

Initially, I designed these to basically join using half-laps. Essentially, each wood stud would have either dados or rabbets cut into them, then I’d join them together using glue and screws. This makes for a very strong table, but it’s more than I need. Archived here is the initial table version.

• USDZ (AR Preview)
• Fusion 360 file.

Materials

This is made using 2x4 studs and plywood, held together using simple woodworking joints, with wood glue and screws to supplement.

• 2“ x 4“ studs (which should actually measure 1.5“ x 3.5“)
  • 2x 48“ studs
  • 3x 24“ studs
  • 4x 32“ studs
  • 2x 37“ studs
  • 3x 16“ studs
• Plywood
  • 1x 48“ x 24“ x 3/4“ plywood “project board”.
  • 1x 34“ x 16“ x 1/2“ plywood “project board”.

To minimize material needed to buy, you can buy 6x 8’ studs, and have the lumber yard cut them to size. Additionally, for the plywood, it’s significantly cheaper to buy 4’ by 8’ sheets (oftentimes, these sheets only cost about twice as much as a project board, but you get 4 times as much material), and either cut them down yourself, or have the lumber yard do so on site. Most places won’t charge you to get the wood cut.

After the initial table, in the next 2 versions, I added 4x 3“ castering wheels to each table. This required cutting the 32“ studs to only 28“, and some additional spacers between the bottom of the studs holding the bottom shelf and the wheels.

Instructions

First, we’re going to build the frame, then screw in the table and shelf tops.

1. Top Frame

• Cut half-laps into the 48“ studs. These will have a depth of 1.75“ (half the height of the studs). They should be 1.5“ long (the width of the studs). Cut 2 of them 4“ from the edges, then a third in the exact center of the 48“ stud (starts 23.25“ from either edge). Make these cuts all on the same side.
• Cut rabbets into 3 of the 24“ studs. These will be at the ends of the studs. As before, they’ll be 1.75“ deep, and 1.5“ long. Make these cuts all on the same side.
• Join the 48“ studs with the 24“ studs you just cut into. The correct orientation for this is that the 24“ will slot down into the 48“ (so that, when weight is placed on the 24“ studs, they’ll transfer that weight into the 48“ studs)

2. Legs

• Cut rabbets and half-laps into the 32“ (or 28“) studs. These will have a depth of 1.5“, and a length of 3.5“. The half-lap should be made on the other end (but same side) of the stud. The half-lap cut should be inset 4“ from the end it’s cut on.
  • If you instead intend to install castering wheels, both ends of the 28“ studs will have these rabbets cut at their ends.
• Join the rabbet from these studs into the inside corners of the 48“ studs and the 24“ studs of the top-frame.

3. Bottom Shelf

• Cut 3 half-laps into the 37“ studs. All these should be 1.75“ deep, and 1.5“ long. 2 of them should be inset 1.5“ from the ends, and the third should be in the dead center (starting 17.75“ from the ends).
• Cut 2 rabbets into the 16“ studs. Similar to above, these should be 1.75“ deep and 1.5“ long.
• Join the 37“ studs into the legs from earlier, into the unused half-lap cuts near the bottom.
• Join the 16“ studs into the 27“ studs.

4. (Optional) Installing castering wheels:

• With the table inverted (top on the ground):
• Add the shim material (enough to cover the distance between the 4“ from the bottom of the shelf-studs to the floor, minus the height of your casters. In my case, the caster wheels, including hardware, were 3.5“ tall, so I added 0.5“ of shim material).
• Predrill wholes for the casters, and screw them in.

5. Surfaces

• Cut out rectangles in the corners of the bottom 34“x16“ sheet so that it’ll actually fit into the bottom shelf.
• Screw the surfaces to their respective places (the 48“x24“ plywood sheet should screw into the top, the 34“x16“ sheet should screw into the bottom).
  • It’s actually entirely unnecessary to screw the bottom sheet into place. It’ll be held in place by the legs on the bottom.

Last updated: 2022-04-25 21:05:02 -0700

Workdesk

A simple workdesk on wheels with pegboard backing.

• USDZ (AR Preview)
• Fusion 360 file

This project happened because, while I had quite a few workbenches made already, I wanted something that would also be more of a designated soldering station. As well as a place to store tools and other things so that they’re not just sitting on one of the workbenches.

Materials

As with the workbench, this is constructed primarily with 2x4 studs, and uses a 2’ by 4’ sheet of 3/4“ plywood as the tabletop.

• 2x4 studs.
  • 2x 60“ studs (the rear legs). (1)
  • 2x 28“ studs (the front legs)
  • 3x 45“ studs (the cross supports)
  • 3x 21“ studs (the top depth supports)
  • 2x 22.5“ studs (the bottom depth supports)
• Plywood/planes of wood
  • 1x 48“ by 24“ by 3/4“ sheet of plywood (top surface)
  • 1x 48“ by at least 8“ by any thickness sheet of plywood (to catch things from rolling off the back) (1)
  • 1x 48“ by 24“ sheet of pegboard (1)
• Extra hardware
  • 4x castering wheels, to make it easier to move around.
  • Many 3“ screws.

(1): You can change the height of the rear legs to whatever you want. If you don’t need a backing (i.e. basically making another workbench), you can leave it at 28“ to match the front legs. It should be at least as high as the front legs height (28“) + the depth of the tabletop (3/4“) + however high your pegboard is (in my case, 24“) + any amount of “gap” between the bottom of the pegboard and the tabletop. Having a gap is desirable as it lets you better utilize the pegboard. The plans here specify 60“ because it’s easy to remember.

Instructions

This is intended to be as simple as possible.

1. Use woodglue (you can dilute it 1:1 with water to make it easier, but I generally choose not to do this) on the endgrains of the cross and depth supports. This makes drilling into the endgrain result in a much stronger joint.
2. Top Frame

• Line up the cross supports (the 45“ pieces) and depth supports (the 21“ pieces), with the depth supports inside of the cross supports, and 2 of the depth supports on either end, and the third in the center. Join them together with woodglue, then use screws to secure that connection. I simply screwed into the end-grain. Which isn’t that great of a bond. You could use pocket screws from the other end and it would be a much better joint. Either way works just fine for this.

3. Bottom frame

• Line up the cross support (45“ piece) and bottom depth supports (the 22.5“ pieces), similar to the top frame. Only this time, because there’s no center depth support, and only one cross support. That lack of extra cross support is why these depth supports are an extra 1.5 inches longer. Join them with woodglue and screws.

4. Rotate the frames so that their fronts are facing down, and line up the front legs so that you can easily attach them using woodglue & screws.

• Installing the front legs first is recommended because the back legs, by virtue of being longer, are heavier and make the whole thing more unwieldy to move about.

5. Rotate the assembly again so that the backs are on the ground, facing down. Install the rear legs similar to how you installed the front legs.
6. While the desk is on its back, install the front wheels. Then rotate it again and install the back wheels.
7. Rotate the desk so that it’s upright.
8. Cut 2 1.5“ x 3.5“ tabs into the back of the tabletop such that it’ll fit nicely into the top of the desk with no overhangs on the front or rear.
9. Get a friend to help you as you screw in the pegboard into the top of the rear legs.
10. Install the “backguard” piece by screwing it in. You can add woodglue (or just actual caulk) as caulk to make sure that any small pieces that land back there don’t fall off the back.
11. Screw the tabletop into the top frame. Don’t glue it in so that you can easily replace it in the future.
12. Paint or finish it as desired. Be sure to sand down the rough edges prior to use.

Last updated: 2022-05-07 19:40:39 -0700

Fiberglassing Techniques

Various techniques used in fiberglassing.

BID Taping

Don’t apply BID tape like a standard layup. You want to apply it already-wet and pre-trimmed. This is faster and has a cleaner result.

1. Create a plastic backing that is a little oversized for the final tape.
2. With a marker, draw the desired length & width of the BID on that plastic.
3. Layup the BID inside of it. Layup all of the plies you’ll need in this. Be sure to squeegee out the air bubbles between layers.
4. Once you’re done with the layup, put on a layer of plastic. So that you have this BID tape sandwiched between the plastic.
5. Cut the plastic to the desired shape, following the lines you previously drew.
6. Pull off one side of the plastic, and gently press the tape into place. Remove the plastic backing, and go over it with a clean squeegee to remove air bubbles.
7. Add peel ply another plastic wrapping to finish.

Applying Fiberglass to vertical pieces

Whenever possible, you should layup fiberglass on top of parts that are horizontal (or mostly horizontal). Sometimes, that’s not possible, and I’ve found that using the BID tape technique with those works incredibly well.

Last updated: 2023-08-12 07:19:09 -0700