Cowl

So this is another sort of carryover post; some of this work I did over the weekend, plus a bit more tonight. Basically, I finally talked myself into sitting down and sanding on the aft edge of the lower cowl. I didn’t go to the trouble of getting the gap exactly right like I did with the upper cowl, though. My reasoning is that things are still going to be somewhat in flux until I have the cowl join line and hinge in place; I don’t want to spend a bunch of time on this edge and then have it distort a bit when I bring everything together. So I just sanded until the cowl was able to slip in place and not overlap the fuselage.

With that done, it was time to work on the attachment hardware. I started by drilling pilot holes for the future screw holes on then bottom flange; for now these are just #40 holes with clecos in them, but later they’ll get upsized for screws. Next, I laid out the rivet holes for the hinges along the sides of the cowl, and then went to town getting those drilled and clecoed.

Once I had that finished up, I was able to remove the top cowl and, for the first time, see the lower cowl supporting itself on the fuselage:

It’s worth noting that all this time I’ve been fitting, I’ve been putting the cowl on backwards. Normally, the lower cowl goes on first, followed by the upper cowl, but since the upper cowl gets fitted first, things have to go in reverse order. The good news here is that now putting the lower cowl on/off should be easier, since I don’t have to figure out ways to hang it from the upper cowl or prop spinner.

Anyway, next on the docket will be replacing the upper cowl and using it to mark the horizontal cut line where the halves join, then I’ll get to make yet another Big Cut and spend a bunch of time sanding.

Once again, this is a post covering a few days. I’ve been trying to be productive on the weekends, but it sure is tough after, oh, about 11 AM or so. Anyway, on Sunday I got the upper cowl in place, snugged everything up, and marked the cut line on the lower cowl. Then I made the cuts on both sides and rough-sanded the cut line just to straighten it.

Another thing I wanted to do before really working on the split line was to shim the lower cowl attach to get a nice fit with the fuselage. My rationale here is that adding the shims will probably shift the upper cut edge around, even if it’s just a tiny bit. So before removing the lower cowl for the trim, I took a look at the join along the bottom and sides. The bottom looks pretty good, but on the sides, the cowl sits a little below flush. A bit of probing with calipers indicated that some .020” stock would work for a shim.

Tonight I made the shims, drilled them in assembly with the hinges, and reinstalled them. That was pretty straightforward, but I was a little surprised to see that the cowl still isn’t sitting flush. I’m suspecting that I may have messed up when measuring this. I think tomorrow I might try making a new pair of shims from .025” stock and see if I can get a better fit out of those.

Once I’m satisfied with the shim situation, I’ll be ready to start working on refining that cut line on the lower cowl.

Short night tonight, but I went ahead and made a second set of shims for the lower cowl using some .025” scrap. This took a little more time than last night, since I wasn’t starting with some nice 1/2” strip stock, but no big deal…just a little more cutting and smoothing. I was happy to see that with the new shims, the cowl sits much more even with the fuselage. You can still feel a little ridge there in some spots, but that’s not worth running down.

Since I was finally happy with the lower cowl fitment, I decided it was time to put the upper cowl on and see how the gap looked. At first it was a little concerning, as there was already a decent gap, especially around halfway between the firewall and prop…more gap than I really wanted. A little manipulating of the cowl halves fixed that, though – this seems to b the result of the cowl sides bulging a little at rest, whereas they were being pulled in tight with a ratchet strap when I did the cutting.

So the good news is that the gap should work out fine…the bad news is that this means I need to find some way to pinch the cowl sides while I’m working on finalizing the lower cowl cut to get the gap where I want it. I guess I’ll get to tinker with that next time out.

It is kinda cool to see the cowl halves mating fairly nicely, with no overlap any more. And the gap issues I’m seeing now don’t even show up in this photo:

Hello, my name is Philip, and I suffer from analysis paralysis. That’s the main theme for this weekend.

After getting the lower cowl shimmed to my liking, it was time to start fine-tuning the joint between the cowl halves. A ratchet strap around the middle of the cowl helped pull things together to make the gap slightly more reasonable, but not perfect by a long shot. One bit of good news here is that some research indicated that it was best to wait on fine-tuning gaps until the hinges were riveted in place. The reasoning is that clecos aren’t perfect, and some shifting around is likely when the rivets get set. So instead of spending a bunch of time carefully sanding the joint, instead I just focus on rough-sanding it to where I could pull the cowl halves together for a reasonable tight gap with no rubbing or interference.

That led to the more fun and open-ended consideration of laying out the hinges at the joint. This seems deceptively simple, but there are a lot of interrelated issues here that all sort of come into pay at once. The first issue concerns the misalignment I have at the forward edges of the cowl. I have a few different options for addressing this. The first and easiest would be to just slap some micro on the upper cowl to build it up, then contour sand to match the lower cowl. The main concern with that approach is that the inside portion of the cowl there won’t line up at all, which could be a concern for dealing with how I want to cover the hinge pin access points.

What I’ve been wanting to do is to make some nice custom flush covers here – an approach I liked ever since seeing this RV-9 build log. However, to do what that builder did, I’ll need the inside portion of the cowl halves to line up, since it requires laying up an inside flange for the pin cover to mate with. If I want to do it that way, I think what I’ll have to do is lay up actual glass cloth to build up the upper cowl areas, then contour sand that on the outside, then sand the inside of the upper cowl to get the inner surfaces to match.

That’s quite a bit more work, but as of this morning I was still intending to go that route. But as I spent time today trying to work through the process of drawing up my pin cover shape, and laying that shape out against the cowl, and looking at how it would interface with the pins…I’ve come to like it less and less. I’m starting to lean towards a modified version of those pin covers, where the cover only mates with the upper cowl, instead of bridging across both halves.

Because of how I’ll be laying out the hinges, the access hole for them will sit entirely on the upper cowl, so really I only need to cover a hole up there. And if I can only have the cover attach to the upper, cowl, then I remove the need for the inside edges to have similar contours, which would make my life a little easier. I still think I’ll want to build up the upper cowl with actual glass instead of micro, and I’ll still carve some material out of the inside, but all I’ll have to do is get a vaguely uniform thickness up there.

So that’s my working idea for now. I think I need to let the idea bake in my head a little before I commit to it, but I do feel like this approach gives me a more definite path forward – which is something I need. I’ve got to find a way to get my brain past thinking 40-some steps in the future, and instead just be able to, say, go ahead and mark/drill the hinge halves to the lower cowl.

I guess we’ll see how I feel about this idea come tomorrow morning – a good night’s sleep seems to often help with clarity on stuff like this.

So while I’m still not 100% sure what I’m going to do for my cowl pin covers, I’ve decided that figuring that out doesn’t need to stop me from moving forward on getting the hinges in place. So a couple nights ago, I measured the hinge locations for the fourth or fifth time, and then cut my hinge stock to make the two pieces. ThenI did some marking on the cowl for the rivet locations, and figured out how I wanted to lay out the rivet lines.

Tonight I actually broke out the rivet fan to mark the rivet lines, clamped the lower hinge halves in place and drilled the ~70 rivet holes for those to mount. Now that that’s done, I get to figure out how to handle drilling the upper cowl hinges. Before I can do that, I’m going to have to go back and bevel the hinge halves that connect the upper cowl to the firewall. Right now, the upper cowl has to be slid in place from the front, but that’s not going to work once the side hinges are drilled – it’ll have to drop in place. I’ve been putting this work off for a little while.

Another fun question is how I’m going to brace the back of the hinge against the upper cowling. At the front, I should be able to reach through the air inlets and get to the back of the hinge, but once I get to where the cylinders are, that’s probably not going to work any more. Maybe if I work front-to-back, I can count on each location being held in place by the preceding cleco? Not sure how I feel about that, even using a wood block to back the lower hinge, the drill bit wanted to push it away in a few spots. I’m not sure I’ll have any choice, though…if I try and do this with the cowl halves off the plane, I lose a lot of control of proper alignment between the two.

I’ll probably just have to sort of grin and bear it in the end…

Previously, I mentioned wanting to work on the hinges for the upper cowl before moving on to drilling the side hinges. To recap, the reason for tapering the hinge eyes is to improve the fit where the cowling curves are the tightest – and more specifically, to make it easier/possible to put the cowl in place from directly above. None of this would be necessary if the cowl could be slid in place from a forward position, but that’s not going to be possible once the side hinges are in place.

So tonight I went to work on those hinge eyes with a file. Overall, it wasn’t as tedious as I was expecting, though it still wasn’t exactly a lot of fun. One thing I realized as I did this is that the end goal here isn’t for the cowl to drop right in place from above – that’s just not going to be possible with the way the sides wrap around and down. Instead, I think it’s fine if I can drop the middle (ie not so curved) portion of the cowl in place, and then work the sides into proper mating just by manipulating each side a bit. This still gets me the real goal, which is to be able to get the cowl in place without having to slide it forward/back.

In the end it took just a couple tries to taper the pertinent eyes, and then I was able to get the cowl to pop in place like I wanted. One more little thing out of the way…

With the upper cowl dropped into place yet again, I took a bit of time to look over the alignment on the sides, and I think I’m going to have another item to cover before I can drill those hinge halves. You see, near the front corners of the cowl, where the upper and lower don’t line up perfectly, that misalignment may make it tough or impossible to get the upper cowl to nest properly with the hinge halves. I didn’t spend a ton of time really evaluating the fit, but it may be that I have to address the alignment of those corners – via adding layup material and/or carving out the inside material some – before I can do the drilling.

Working on the cowl is really an interesting love/hate thing. On the plus side, seeing the airplane with the cowl on it is motivating, because it’s really starting to look like an airplane! On the other hand, fitting this thing is turning into one of those multivariate problems where each little thing you need to do is affected by four other things, with all sorts of cascading effects.

But hey, progress is still happening, even if there’s a lot of rethinking on the fly going on…and in the end, the motivation is definitely beating out the fitting challenges.

So…yeah, it’s me. Last time we were here, I made an offhand comment about the cowl not lining up properly near the front corners. I think I said something about it “maybe requiring a little work” to address. Since then, I have spent a lot of time both thinking of the best way to address this, and trying to avoid the inevitable need to cut out chunks of the cowl and completely redo the layup. It also hasn’t been difficult to talk myself out of working on the airplane with weeks on end of triple-digit heat.

This post is sort of a culmination of a bunch of small assorted work sessions over the past week or two, but with some actual consistent work over the past couple days.

The basic deal here is that I need to find a way to create a mold of the upper cowl as it needs to be shaped – then I can adapt this to the existing cowl and work out scarfing in new fiberglass to fix the areas where fitment is an issue. For while I was intending to start with pink insulation foam board from the home improvement store, carefully sculpted to the right shape. The problem with that is that it would require a lot of work to get the shape right, especially since some part of it would have to match the existing cowl. I did go so far as buying the board and starting to work on it, but that just made me even more aware of how much work would be required.

I’d also considered experimenting with two-part expanding pour foam. I’ve seen folks use this for molds as well, and it has the advantage that I can mold the inside of the cowl as a starting point, then build off of that. So I bought some of the stuff, mixed up a test batch, and…didn’t like it. It wanted to collapse after curing, and didn’t get very hard. I was ready to abandon that, when I talked to someone who said that’s definitely not how it should cure. So I decided to mix up a second test batch, and when I did I saw that one of the component bottles was a lot lower than the other. This stuff is supposed to be mixed 1:1, but apparently I somehow mixed it more like 2:1. No wonder it didn’t turn out well. Test batch #2 was much better, and I finally felt I had something I could try out.

So the next step was to prep part of the upper cowl to essentially take a cast with the foam. This just involved me spending a lot of time laying down plastic packing tape inside the cowl, covering every surface where I’d be molding to keep the foam from, uh, becoming one with the cowl. Then I carefully cut cardboard pieces to create a dam around the area I wanted to mold. Nothing here is straight, so there was a lot of fiddly work to get to this point:

Eventually there was nothing left to do except mix up a batch of foam, pour it in, and hope I’d sealed everything up decently. Then I mixed up a second batch, because I’d vastly underestimated how much I needed to fill this space. Then I mixed up a third batch, because it still wasn’t enough. The result is kinda weird looking, but that’s OK, I expect to cut off a lot of what’s seen here as part of shaping the mold:

Then came the real fun: seeing if this mass would actually turn loose of the cowl. It definitely took some muscle, but it finally broke loose. I also had to spend a bunch of time peeling off all that packing tape, which let go of the cowl in favor of sticking to the foam. I was a little worried about the foam filling in the space around the cowl inlet, but it did pretty nicely:

I did some rough trimming and then decided to leave this piece alone for the night (this was Tuesday, I think). The “outside” of the mold is a little irregular, thanks to the packing tape wrinkling on the various compound curves, but I figure once I’m done shaping this, I’ll hit it with some microballoon slurry to fill out the voids and make a nice surface:

So that was one of the two corners I needed to do this for. Tonight, I spent another hour or so going through the same process for the other corner. So at this point, I have the starter molds for both sides of the cowl. I think the next step will be drilling the upper cowl to the hinges as far as I can, then I can start thinking about the fun part of cutting some big chunks out of that upper cowl.

My current plan  isn’t to completely redo the layup in this area – I’d like to keep those corners with the really tight compound curves, and then basically splice them back into the rest of the cowl. I’m not sure how that’ll work in practice, but I’ll give it a try. The cloth I have should be fine for handling the compound curves, but I’d still rather avoid that fun if at all possible.

So, once again, it’s been a while. I’ve been less able to use the heat as an excuse recently, especially over the past couple days…but I’ve still been putting off this whole cowl rework. Just always seemed like it was going to be a big deal, with ample opportunity to mess the cowl up and so forth.

The thing that finally got me moving on this was that after this week, I’m taking a week off work, with the intent of getting in some good build time. I’ve got a bunch of vacation to burn before the year is out, so a staycation is in order. What I didn’t want, though, was to spend that first day of vacation doing an hour of fiberglass mold prep, and then…having to wait a day for stuff to cure. I really wanted to get working on the mold.

The first practical action for this is pretty simple: I wanted to start drilling the upper cowl to the hinge along the split line, and my intent was basically to work my way forward until I couldn’t easily pull the cowl halves into alignment. I figured this would identify the spot where I’d make the cut in the upper cowl, and that’s where I’d start working on the mold and redoing the layup.

So I started on that today, and…well, I made it pretty close to the inlets, with things pulling into alignment nicely. I started to wonder if I could go all the way forward and skip the layup, so I spent some time messing around. While it’s not really possible to pull the upper cowl in in this area, the lower cowl can actually be pulled out and into alignment without a lot of difficulty.

My last test before committing to drilling the rest of the holes was to try removing and reinstalling the hinge pins. My thinking here was that if there was excessive tension caused by the misalignment – the kind of tension that could lead to the hinges breaking in the long term – then it’d manifest in making the pin really difficult to insert. That wasn’t the case at all, though – I got the pins out and back in with no drama at all.

So I finished drilling. The halves don’t pull perfectly into alignment, as shown in the following photo, but it’s close enough that I think I can make this work with just some shimming, filling, and sanding, which is far less of a daunting proposition than completely redoing things here.

And that’s where the black comedy comes in. I have spent literal months agonizing over how best to redo these layups, how to make the molds, how to avoid making things worse, and just generally finding an excuse not to work on this…and it sort of just came together without a whole lot of effort.

sigh

Once again, we have a post here that covers several days worth of work. Most of this work was from a few days ago, but I just consistently kept forgetting to actually write this.

Fun fact: I took all of last week off work, with the intent of putting in a lot of time on the RV. It didn’t really work out that way, due to a combination of being more worn out than I expected, plus lots of other to-dos and prep work for our upcoming vacation. Still, I did put in some time, and I feel like I’m developing that habit of working on the project again. A lot of that probably comes from pushing myself past the “think everything through” stage into “just do stuff.”

The actual form “just do stuff” took was starting to throw fiberglass material at my shaping/fitment issues. The first question at hand was whether I wanted to do any shim work at the forward edges where the hinges were really pulling on the upper cowl. To evaluate this, I assembled the cowl on the workbench and moved all the hinge clecoes to the inside, so I could actually get a better look at the alignment. That also gave me the chance to deburr the hinges after all that drilling. After getting everything together, I decided that shimming wasn’t necessary.

That left the other two big fitment issues: alignment of the spinner surround and forward corners. I really don’t understand how the spinner got misaligned – somehow everything lined up nicely on the bench, but after all the trimming and hinge fitting was done, the lower cowl sat forward of the upper by about 1/8”. The choices to fix this were to either fill the holes at the firewall on the lower cowl, then re-do the trim and hinge drilling, or just work with the front of the cowl…I chose the latter. It helped that I ended up with a slightly larger gap between cowl and spinner on the upper cowl, so adding some thickness there won’t hurt my feelings.

Overall, the plan is to add about 1/16” to the upper, then sand off about 1/16” on the lower, to get the alignment I want. At first I considered adding a few layers of glass for the initial buildup, but at some point I realized that there was no reason not to just use dry micro here as a filler. I also decided to take the same approach to getting the outside corners lined up.

Before I could do any of this, I had to sand off the gelcoat in all the areas I was going to build up…boy, that stuff is tough. After putting a lot of elbow grease trying to hand-sand this, I found that the Dremel with a flap wheel makes much quicker work of it. It doesn’t leave a nice flat surface, but that doesn’t really matter when I’m going to be covering with micro anyway.

Finally, it was time to actually mix the micro and glop it on. As per usual, this stuff refuses to be coaxed into any sort of neat-looking shape on application, so I just put on more than I think I need and accept all the weird resulting protrusions:

Yesterday, I started working on sanding this stuff. First step is just knocking off the obvious peaks and high spots, and getting things somewhere in the neighborhood of the proper form. That wasn’t so bad on the corners, but the spinner area is different. My approach to sanding this to get the gap I want is to make a couple of sanding “files,” which I can slip between the spinner backplate and the cowl with the cowl installed; this allows the thickness of the “files” to easily control the final gap.

The downside to this approach is that the cowl has to be installed first, which means enough of the micro has to be sanded off for it to clear the spinner. To do this, I once again assembled the cowl halves on the bench and used the lower spinner area as a reference for the rough sanding. Then I went to install the cowl halves on the airplane, and had one of those annoying little sidebar issues…which I’ll get to in a second. For the time being, I got the cowl installed and started working on the sanding…which has been tedious, but I’m making slow, steady progress.

OK, back to that sidebar issue: I had the worst time getting the hinge pins started to join the halves. I tested this back when I was deciding if I could skip redoing the upper cowl, and things seemed to go together well…but not today. I probably spent an hour trying to get those pins in. To make a long story short, I probably need to do some shimming of the upper cowl after all.

But this also got me thinking about how I laid out those hinges. Because I wanted to not have the hinge teeth visible in the gap between cowl halves, I offset the hinges up. This means the hinge pin isn’t in line with the cowl split, but rather about 1/4” above it. This causes a couple of annoying problems: one is that maneuvering the pins in the small temporary access holes I cut is tough, but another one is that my rivet holes on the lower hinge half don’t have proper edge distance. The latter would probably be OK, since I intend to epoxy the hinges to the cowl in addition to riveting.

The other result of this decision was that I was going to make my own custom fiberglass cover plates for the hinge pins; these would be removed to access the pins and remove the cowl. At this point, I think continuing with what I’ve done is just going to add more work and time, and I need a lot less of that if I want to get this project done next year.

End result? I’m going to redo the horizontal hinges. This time I’ll put the pin in line with the split line, and I’ll just use the prefabricated cover plates. Those plates are a little pricey, but I think the reduction in custom work is well worth it. I’ll just have to fill the rivet holes I already drilled with flex and start over…that will probably be my next task once I get this spinner shaping done to my satisfaction.

What’s this, a blog post? No way!

OK yeah, I’ve coaxed myself past the “walk through the hangar and think of what I should be doing” phase and into, well, actually doing something. I’ve been tinkering from time to time with stuff relevant to the cowl, namely trying to figure out the right placement for my hinge pin covers, as well as a trial run for filling the old hinge rivet holes with flox. The latter isn’t really a complex procedure, but since I had some scrap lying around (the cutout for the oil door), I decided that trying a new technique on something other than the really expensive cowl was a good idea.

Turned out it really was pretty foolproof. Mix up epoxy, add flox (flocked cotton fiber) until I got a peanut-butter consistency, then work a bit of it into each hole. From there, I sort of worked the excess from one side of the material to the other to get good filling of the holes, while gradually scraping off excess. Worked out pretty nicely.

That was…a few weeks ago. Tonight, after doing various household chores, I talked myself out of continuing to sit on the couch and into getting those old rivet holes filled. I just had to remove some of the clecoed hinge pine, mask off some close-by holes that I didn’t want to fill, and then get to wiping flox into the holes…a whole lot of them. In addition to the rivet holes, I also filled the two holes I made in the lower cowl way back when, which I used during the fit-up process.

So now the rivet holes are filled and curing. Next I’ll knock down the area with some sandpaper, then I need to finalize the hinge pin cover locations, before hanging the cowl again and fitting and drilling the new replacement hinge halves. Eventually it’ll be back to working on all that micro I glopped around the prop spinner area, which will be lots of fun…