Well, this has been a tedious bit of work so far. After previously doing a test-fit and getting some guidelines to use for working on the cowl halves off the airplane, I’ve been working on getting the forward mating fine-tuned. The first issue was finding the best way to work on the cowl; my previous approach of putting both halves on the floor, aft side down, just wasn’t working for lining them up reasonably.

I decided to try laying the lower cowl up on the work table, and sort of going from there. A jug of checks placed inside the lower cowl, plus some wadded- up old sheets beneath, manages to keep it in a reasonable working position. Then, the upper cowl can sit on top, with the aft ends held together in rough alignment by just a couple cleco clamps. The result actually works out pretty well:

That was the simple part, however. The basic task in getting the forward mating surfaces right is to carefully massage the four points of contact: the two sides of the prop spinner area, and the outboard corners. What I’m going for is having all four points of contact mating once and evenly, while also ensuring that the two air intakes are also the same size. It’s entirely possible to remove material asymmetrically and end up with a lopsided nose, which I don’t want.

The air intake aspect actually made this slightly more straightforward; when I compared the two, I found that the left intake was about 1/4” taller than the other one. Not enough to really be obvious without using a measuring tool, but that meant that I could essentially consider the right outboard corner junction to be my reference point, and do all the material removal on the other three points. This would have the effect of tending to shrink the left intake to make things more even.

The process is simple, but tedious. As long as an unwanted gap exists somewhere, go find the spot where the halves are making the tightest contact, and sand that area until the contact becomes light, or even until there’s a bit of separation for larger adjustments. Then find and sand the new tight contact point, then find the next one…over and over.

At first I was making relatively dramatic adjustments, using a sanding block with 40 grit. As things started to get closer, I switched to using bare sheets of sandpaper, either a bit of the 40-grit – which is relatively thick, cloth-backed material – or some 60 grit I had lying around. The 60 grit is nice thin paper, and works really well for truing up a gap. The approach here is to just pull a strip of sandpaper in the joint to be worked, apply a little pressure to the joint, and work the paper back and forth. An uneven gap will naturally tend to even out, since the tight contact point will put the most pressure on the paper. Once the paper moves with about the same resistance throughout the joint, it’s trued up pretty well.

There’s still a little more work to be done, but overall things are looking pretty good:

Really my only gripe at this point is that the prop spinner joints still aren’t trued up very well – they make good contact on the forward-facing portion, but the flanges have a bit of a gap that opens up towards the back. I’ve about decided that I don’t care about this on the inboard side – this won’t be visible at all when installed on the airplane – but I do want to even up the outboard portion a bit.

Also, the above photo somewhat displays another middle annoying issue I’m dealing with up here – all four points just won’t line up at the same time in terms of the outward-facing areas. If I line the outboard corners up, then the lower cowl sits a little far back on one side of the spinner, and a little bit more on the other side. Note the bit of shadow on the right side of the spinner above. If I close the gap at the spinner, the upper cowl corner overhangs by 1/8” or so, and there’s barely any material left to trim at the back of the lower cowl. On the other hand, if I line up the corner, then I have a similar big gap at the spinner.

In the photo above, I’m basically splitting the difference, which I decided I liked best after trying various approaches. I have to remember that the end goal here is to set the joint on the front of the cowl, so I can then hang them and trim the lower. Sometime down the road, I can worry about fixing the misalignment – I’ll be spending time on the entire exterior of the cowl, and adding some micro up here and reshaping everything will just be part of the fun.

So, after the last session, I spent some time reading up on cowl fitting. One particular point I wondered about was whether it might be a good choice to just cut off the overlapping flange around the prop spinner, in order to simplify getting that interface the way I wanted it. I figured it should be fairly simple to add a flange back later on, and when I found a post from Dan Horton, one of the RV fiberglass gurus, suggesting this same path, I decided to go for it.

So I sawed off those flanges, enabling to me to work on that junction as a simple butt joint. This made things easier, but still not 100% easy –  the new issue was figuring out how to fit the portions of the flanges that extended aft. If I just have the cowl halves on the floor, I have to take a pretty wild guess at where they should be positioned at the aft end, and this in turn affect the mating angle at the front. I felt like what I needed was a way to test-fit the halves on the plane to see how the junction was going.

The problem here was figuring out a way to support the front of the lower cowl. Eventually there will be nutplates holding the two halves together, but not only can that not be done until the fitting here is done, it’s really out of the question now that I removed the flange. Eventually I decided to just drill a pair of temporary holes, one below each air inlet. Each hole has a cleco installed, and then a piece of string can go between the checks and over the prop spinner, thus providing a way to roughly hold the lower cowl in position.

With that done, I made my first fitting of the lower cowl on the airplane. The string+cleco idea worked pretty well up front:

…and to bring the rear into alignment, I used the usual method of wrapping a ratchet strap around and using it to pull the lower cowl up against the fuselage:

Doing this allowed me to check the mating angles at the spinner joint, and make some marks to fine-tune them. I removed the lower cowl, sanded it to shape a bit, and hung it once again. This showed that my joints were getting pretty close, but the right side didn’t quite want to pull flush. I suspect that the inboard part of this is interfering, but it’s impossible to see with the cowl on the airplane…which is a new issue with this approach.

But I realized that the main purpose of this was to figure the relationship between the cowl halves when roughly in place. Being on the fuselage was necessary to establish that relationship, but not to maintain it. With the halves pulled together, I just used a Sharpie to trace the upper edge of the untrimmed lower cowl onto the upper…and now I can use that to join them in alignment without being on the airplane, which should allow me to to the last bit of fine-tuning of this interface up front.

That will be for another night, though.

Today I picked right back up with the fun of trying to hash out the junction between the cowl halves. After looking at some other build logs last night, I worked out a better idea for trimming the forward corners, which in turn allowed for me to make a much better attempt at nesting the spinner area.

This showed me a new problem: specifically, the material thickness in this area on the upper cowl is 1) not consistent, and 2) significantly thicker than the recessed mating area on the lower cowl. So I went to work removing material here; I sanded down the upper cowl mating area a bit, but decided to do the bulk of the removing on the upper cowl. So far, I’ve been through three or four iterations of removing material, smoothing the area out, and trial-fitting the halves together.

Things still aren’t fitting perfectly, but they’re a lot better. I can slide the junction almost all the way together:

But the junction still isn’t quite flush, as a picture from inside the spinner opening shows:

I’ve already surrendered to the fact that this mating area isn’t going to fit snugly at first – there’s just no way I’m going to accurately sand this stuff down to a precise fit. My goal is just to get it to a point where it nests properly, and then I figure I’ll cover one side with tape and mold release, then apply some epoxy-flox mixture and clamp it back together again to fill the gaps.

Similarly, the front corners aren’t exactly meeting beautifully. Maybe they’ll end up looking better once I’m really fitting the halves together, but I’m preparing myself for the possibility of needing to do some significant reshaping there to get things looking good.

OK, so…first order of business today was revisiting the aft flange of the upper cowl. As mentioned yesterday, while I did a lot of fine-tuning of the fit to the fuselage, I still wasn’t completely happy about it. So I spent even more time carefully sanding the flange to further tweak the fit. The good news was that things were close enough that I no longer needed to remove the cowl to sand it. With the spacer between the cowl and the prop spinner, the back of the cowl was pressed snug against the fuselage. That meant that I could wiggle the joint around to figure out where there was contact, then lift that area enough to sand it a bit with some 80 grit glued to a popsicle stick. By doing this, well, a bunch of times, I eventually got things looking really tight:

Since I was finally happy with the fit, it was time to move on to drilling the upper cowl to the hinge halves that retain it. To do this, the holes get laid out and initially drilled in the cowl, then later drilled in assembly with the hinges. As suggested in the construction manual, I laid out the rivets about 1” apart, after marking the ends of the hinge pieces on the cowl. A little work with the rivet fan and I had my holes taken care of:

I also took this opportunity to rough-cut the opening for the oil door; this opening also serves to allow access to the upper cowl hinge pines, so I needed it open for the upcoming work. I didn’t open this up all the way to the plans dimensions; I’m probably going to do something custom with the hinge and latch for this, so I’m leaving extra material for potentially working that out later.

Then the cowl went back onto the fuselage, and I very carefully got it positioned, then started drilling. My approach to managing the gap between the cowl and fuselage is to use that spinner spacer to hold the claw snug against the fuse, then I just manually held the hinge half in the rearmost position while drilling. The intent here was to use the bit of slop in the hinges to allow for moving the cowl forward for final fitting later on. About the only issue here was that while I could easily manipulate the hinge on the right side of the plane – thanks to that oil door opening – the left side was way more difficult. The end result is that I don’t have quite as much room to move the cowl forward on the left side, but I’ll still be able to do even more sanding on that later on to finalize the gap.

Anyway, at the end of all that I had the cowl fully checked to the hinge half: 

And with the cowl off, we can see those newly fitted hinge halves:

There was still more work to be done on these hinges, though; the cowl isn’t as thick as the material stickup in the fuselage, so there’s some shimming needed here to get a nice flush line between the cowl and the fuselage. The plans call for just doing a blanket .032” shim, but I noticed that the gap was more pronounced around the edges of the cowl, and less so in the middle. So I did some trial-and-error with some aluminum scraps; this resulted in using those .032” shims on the outside edges, but then closer to the middle .025” shims made for a better fit.

Once all that was laid up, I got to do the work of cutting the shims, drilling them in assembly with the hinge halves, and finally fitting everything back to the cowl. Once that was done, the cowl went back on – again – to verify that things lined up as expected. The fit still isn’t 100% perfect, but if I want to fine-tune this down the road, I should be able to do it just by adding a bit of micro to the cowl and sanding to fit.

OK, next up – it was time for the rest of the trimming on the upper cowl. With everything else in place, I could use some layout marks I already had just behind the spinner – marking the correct cut point to get things nice and symmetrical with the lower cowl – and lay out the cut lone for the bottom edge of the cowl. My approach for this was to use a laser level to first verify that my reference marks were appropriately level, and then to carry those marks around each side of the cowl.

I started with the fuselage in the level flight attitude – this would have given me a cowl split line parallel to the longerons – but that resulted in the level line dropping off the bottom edge of the cowl at the aft end. So I had to gradually lower the tail until the cut line carried the full length of the cowl. It’ll end up being a degree or two off from the longeron line, but I kind of doubt anyone will notice…

Then I got to go and carefully make Sharpie marks along the level line, along both sides of the cowl… while also making sure not to, you know, lase myself in the eyes:

Then, as with the last trim, I used masking tape to lay out the cut lines all the way around:

Which was followed by rough-trimming, sanding the cut lines close to the masking tape lines, and finally using a long sanding board to finish the entire cut line nice and straight. This was surprisingly not really a ton of work; unlike the aft cut line, which had to be carefully fitted to another part, in this case I’m creating the layout line, so there’s not so much trial-and-error. (That’ll come later when I trim the lower cowl to fit…)

OK, enough of the upper cowl…now it’s time for the real fun, the lower cowl. The lower half mostly butts against the upper cowl, except for the spinner area. Here, the lower cowl has an inset flange that need to fit snugly into the upper cowl…but that there’s no way to test-fit this intersection as provided, because the extra material the outboard cowl edges interferes:

So the strategy here is to carefully and gradually trim the forward outer portion of the lower cowl to hash out this nesting area. The marker line in the previous picture follows the edge of the flange, so that’s hypothetically where I’ll want to trim things here. Thing is, I don’t want to trim any further back than I have to, since I can’t know how what the overlap of the two halves will look until I can fit them together…aka not until I get the front to nest. It’s a sort-of maddening chicken-egg type problem.

The approach so far has been to slowly and conservatively remove material in this area, while repeatedly trial-fitting the halves. Eventually I should be able to creep up on a decent trial fit, but I’m not there yet. I got this far before deciding it was time to call it a night:

So this is where I’ll be picking up next time. I’ll probably need to rethink my localized trimming here to creep up on the fitment…that’s something to mull over for a while. As tedious as some of this is, though, it’s still pretty neat to see the cowl profile coming together.

Once again, this is a multi-day report; I actually started on this stuff yesterday afternoon. I’d actually put the upper cowl on and off a couple times over the preceding week, as I was thinking through my next move for the fitting process. I also swung by Home Depot this past Thursday to pick up a 2’x2’ piece of 1/4” plywood, to make a spacer between the cowl and prop spinner. That work got done yesterday – first, making a hole in the middle of the sheet to match the prop hub, then splitting the sheet into two pieces. The idea was that I could rest it on top of the hub for fitting.

Next up, it was time to start messing with the cowl itself. Before doing anything, I needed to get the fuselage into a level attitude, which meant finding the right combination of boxes and scrap lumber to raise the tail to the proper height. The good news here is that it’s a lot easier to lift the tail with the engine hanging out front – that’s really good since I’ve now put the tail on and off the boxes several times.

Anyway, with the fuselage leveled both front-back and side-side, I got my laser level set up to give me a horizontal reference line. This was mainly so I could similarly level the cowl side-side while doing the other positioning. Then I got to mess around with my adjustable supports up front to get the nose of the cowl where I wanted it relative to the spinner, as well as maintaining the side-side level I wanted. Then I checked, double-checked, triple-checked, and so on, until I was really satisfied that I had the location I wanted. Once that was done, I used the tape line I laid out on the fuselage, measuring 2” forward, and laid out the cut line for the back end of the cowl:

Next, I broke out the masking tape and laid a stripe along the marks to create a continuous reference line for the cut:

Then it was time to take a deep breath and make that first cut. I wanted to creep up on the final size, so I worked to make the cut about 1/8” or so shy of the actual cut line. Getting the final fit was always going to require a lot of iterative sanding, so no real reason to try and get super close on the first pass:

Now, the real fun part here is that before I could get to the iterative sanding, I first needed to install the hinge halves that will support the back of the upper cowl. Prior to trimming, it was resting on the fuselage, but now it would need to butt against the fuselage, and thus rest on the hinges. So starting this morning, I cut all the hinge pieces I’ll need around the cowl, then set about fitting the upper hinges to the firewall. This is kind of fun since the hinge needs to be bent to roughly the same profile as the fuselage before fitting, and then there’s just a lot of drilling in assembly. Then I cut the pins for these pieces, and pinned the other hinge halves in place.

Also, I forgot to take any pictures of this, I guess because I was so excited to lay the cowl in place and see how it looked.

Anyway, this got me into the fun iterative part. All told, this cowl probably went on and off the airplane fifteen or twenty times today. Lay it in place, evaluate the fit, figure out where the interference was, take it off, sand that area, lay it in place again…and so on. I was pleased to see that I started with only about a 3/16” gap from cowl to spinner, meaning that I had plenty of leeway to fine-tune the fit while getting the gap right. Eventually I worked up to fitting the plywood spacer between cowl and spinner while doing the fit-ups, and eventually I had a fit that I (mostly) liked at the firewall:

Up front, I’ve got a nice even gap between the cowl and spinner:

I think I’m still going to tweak that fit at the back a bit more tomorrow; there are a couple spots I’m still not completely happy with. Once that’s done, the next step will be to lay out the rivet holes on the cowl, then use those to drill through the cowl into the hinge and cleco everything together. That, in turn, will make the upper cowl stable enough that I can mark the horizontal trim lines along the bottom, and make those cuts. At least those won’t require a bunch of fitting, since they just need to be straight – the lower cowl will eventually get trimmed to match those cut lines.

The one thing that’s bothering me is that the hinge halves for the upper cowl have some slop in the fit. The plans specify undersize hinge pins for those, which are necessary in order to fit the curve of the upper fuselage. But that also means that the hinges don’t fit tight…and that means that it’s going to be hard to get the gap I want between the upper cowl and the forward upper skin. I don’t want a contact fit here; there will eventually be paint in that gap, and if it’s too tight the paint will get chipped removing and reinstalling the cowl.

My best guess for handling this is to allow the lower cowl – which will attach with full-size hinge pins – to do the primary locating work. It’ll be fixed in place, and the upper cowl will be fixed to it via the horizontal hinges that will join the halves. Maybe the best approach is to get the upper cowl fitting tightly to the fuse, drill the hinges such that the cowl can be pulled forward to establish the right gap, and then get everything fixed in place when I add the lower cowl.

We’ll see how this all works out over the next week or so, I suppose.

So, as implied by the title, today was all about getting set up for the joy of cowl fitting. I started this off last night by pulling out the line material that will be used to attach the cowl halves. This was in line with the first step of cowl fitting, per the instructions, being to cut and install the the hinge halves on the firewall. However, I ended up putting those back and delaying for later, mainly because I can’t really follow the instructions to the letter here. The main hangup is that I don’t (and won’t) have the forward upper skin riveted in place, which in turn precludes having those hinges riveted, since all these items are riveted in assembly.

Instead I moved on to a more fun problem, something I’ve been mulling over for the past few days. Briefly, the cowling halves are provided slightly long, and need to be cut to length in order to fit snugly against the firewall. The basic procedure for this is to lay out a line 2” aft of the firewall flange, lay the cowl on top so it overlaps the forward fuselage, and after doing all the fitting and positioning, measuring back 2” forward from that line lays out the initial cut line.

First problem: The firewall isn’t held in position unless that forward upper skin is installed. That’s not necessarily a show-stopper; I can cleco it in place, measure back from the firewall to mark the line, and then remove the checks so the cowl rests on the skin. But that leads to…

…the second problem, which is the right upper portion of the firewall. The left side is held securely in position by the skin, but the right side has a giant open spot where the forward baggage door will go. As provided, the skin didn’t have this piece cut out, but…I cut it out some time ago so I could match-drill that skin to the underlying structure. In retrospect, I wish I hadn’t done that. If that piece wasn’t cut out, then this skin would hold the right side of the firewall securely as well.

Anyway, it’s water under the bridge, and I’m left with this setup. You might be able to imagine how that portion of the firewall has a lot of freedom of movement here:

To solve this problem, I decided to temporarily add the baggage door to the party. Actually fitting this piece is a whole procedure that apparently gives lots of builders fits, but all I needed it to do was serve as a sort of filler piece to position the firewall. In practice, the top of the door attaches to the forward upper skin and underlying structure with a length of piano hinge; for this purpose, I cut a piece of .063” to sit in place of that hinge and attach the door. That piece of scrap was first drilled to the hinge attach structure, then I worked on putting the door in place on the fuselage before drilling the door to the scrap piece.

While doing this, I reviewed the baggage door instructions for inspiration. It turns out that that procedure calls for a nontrivial usage of duct tape, which I took inspiration for. In order to hold the door in the right position for drilling, I added strips of duct tape on the forward side, which also served to pull the firewall into proper position. Between that, and some more strips of tape to hold the bottom edge of the door in place, I finally got the brace drilled, which fixed the door in place pretty well:

Next, I worked between the clecos to place the 2” guide line off the firewall flange, then added another strip of duct tape to really ensure the firewall wasn’t going to move. Finally, after going all the way around the firewall and making marks for the guide line, I put down a strip of masking tape all the way around, which will be my actual reference when it’s time to mark the cowling:

So that takes care of allowing the back end of the cowl to be positioned and marked – the next task is to allow for positioning the front end. If we just lay the cowl atop the engine, the front edge sits on the crank pulley and is far too low. So I created a couple of adjustable plungers which allow for carefully raising the front. These were made using scraps from my first instrument panel cut; each piece attaches to a baffle attach point on a cylinder, and then has a 1/4-20 tapped hole for a long bolt. The bolts are topped with rubber vacuum line caps so as not to damage the inside of the cowl. I also added jam nuts on each bolt, to ensure I could prevent the bolts from moving, as well as taking some load off the threads in the admittedly thin brackets:

Finally, it was time for a trial fit of the cowl to see how all this worked out. Mostly I wanted to make sure that those bolts up front were long enough to raise the cowl to the level needed…or if I was going to need to buy some longer ones. So while this is in no way a careful fitting, I did take the time to roughly align the cowl with the prop spinner:

In this position, I still have a little bit of bolt length left, so it looks like I’m in good shape here. For the final fitting, I’ll need a way to precisely space the cowl off the spinner backplate, as well as a way to establish that the cowl is level from left to right. Really, I just generally need to review the fitting instructions, so I have an idea of the big picture of what I’m doing. It’s probably not surprising that I want to quadruple-check everything I’m doing before I reach the point of actually trimming the cowl…