Cowl

So the general to-do item right now is to redo the cowl hinges. But before I can figure out the length of the hinges and their layout, I need to decide where to locate the hinge pin covers. This has been a source of quite a bit of head-scratching and second-guessing over the past, well, six months. All that time when I wasn’t making blog posts, I was still occasionally taking a look at this issue and trying to figure out how to solve it.

These nice machined covers from Aerosport Products are intended to match the profile of the cowl, but that kind of depends on putting them in the right place. And while they have specific instructions for locating the covers on some RV models, the RV-8 is not among those. I’d previously tried using the RV-7 instructions as a starting point, but they didn’t really work due to apparent differences in the cowl design. So that left me with trying to reason the whole thing out.

To save some time, I’ll skip to the end here – I decided the best way to experiment with location and fit was to make a female mold of one of the mockup covers (these are used for fitment instead of the actual covers). So I covered the, um, cover with plastic wrap and then laid up a few plies of fiberglass over it. Once cured, I sanded the mold to shape, and now I had a way to truly visualize how the perimeter of the cover would rest against the cowl.

This still wasn’t the panacea I might have hoped for – while it gave me a good look at fitment, it also showed me that there was apparently no location where the cover would perfectly match the cowl. That at least allowed me to stop trying to think the problem into submission, and I just chose the best apparent location and moved on. Later on, I may add some micro around the covers to make them blend in better, but I’ll cross that bridge when I get there.

In the meantime, I moved on to laying out the hinge on the lower cowl. I figured out the overall length, laid out the end locations and the outer rivet holes, and marked my guide points for laying out all the other rivet lines with the rivet fan. I shifted all the holes a bit in order to try and put them in between the previous holes that I filled. This is probably overkill, but…well, I am still me, so here we are.

Drilling will need to wait until another night; first I have to remove the lower cowl again and modify the vertical hinges that attach it to the firewall. Those will now interfere with the new horizontal hinges, so I’ve got to solve that problem before I can get to actually drilling the horizontal hinges – but I have a good sense of how I want to handle the situation.

Just more chipping away at the cowl hinge. Last night, I worked on modifying the lower cowl-firewall hinge. In order to accommodate the new horizontal hinge location, I needed to remove one eye from the cowl half of the firewall hinges. Once those were trimmed, cleaned up, and reinstalled, I got the two lower horizontal hinge halves cut to length and those cut ends cleaned up as well. I also broke out the rivet fan and marked all of my rivet hole spots on the lower cowl.

Tonight was drilling time – just clamping each hinge half in place and doing a lot of drilling, moving clamps, adding clecoes, and so on. Once the two lower halves were drilled and flecked, I marked the upper halves and got those cut to length as well. Last order of business for the evening was to lay the upper cowl in place again and mark the area where I needed to remove some material to accommodate the hinge pins during fitment – that’ll be where I pick things up next session.

In the meantime, hey look, it’s a hinge:

Straightforward stuff again this time. Last night I opened up a small passthrough on the upper cowl for the hinge pins, then laid out the rivet locations. Tonight I made some bracing blocks from wood, to support the aft end of the upper hinge half while drilling, then got the upper cowl in place and did the actual drilling.

I’m now trying to decide whether I want to go ahead and mark/cut the openings for the hinge pin covers, or if I should go ahead and rivet the horizontal hinges in place first. I’m kind of leaning towards the latter approach – I’ll likely want to do some sanding/shaping at the horizontal cowl joint, especially towards the front where the covers will be, and I can’t really finalize all that sanding/shaping until the hinges are permanently attached. The main reason for that is that things always fit differently with actual rivets instead of clecoes, and I don’t want to do any finish-fitting (eg setting the gap between cowl halves) with temporary fasteners in place.

On the other hand, just cutting the cover openings is pretty low impact, and would make installing/removing the pins a lot easier for the time being. The point at which I’ll want something resembling final fit is when I add the backer flange for the covers…but that can wait a little while. Maybe I’ll rough-cut the covers tomorrow and then switch gears to attaching the hinge halves.

So…I was prepared this morning to go forward with my plan to rough-cut the pin covers, then pivot to working on the hinges. Instead, I took a brief detour. As I was considering the placement of the covers and marking the cut lines, I decided that it’d be a good idea to patch up the previous pin passthrough cuts I made. These cutouts are just a bit forward of where the new cover cutout would be, close enough that I figured I was likely to mess up the cover cutouts if I went back and patched the old cutouts later on.

So I pivoted to working on those. I sanded the perimeter to a decent taper, then laid up four strips of cloth across the inside of each cutout. I pressed the cloth just a bit into the cutout, mainly to ensure good contact with the taper I’d sanded. The basic idea was to bridge the gap with cloth, then come back later and add flox to fill things out (the latter bit was done this evening).

In the meantime, while the first layup was curing, I pulled the horizontal hinges and deburred all my rivet holes. I also countersunk the rivet holes in the lower cowl…upper cowl will wait til later, ie once the current layup/patch work has cured.

So tomorrow I’ll get to do more sanding (it never ends with fiberglass work), then I’ll probably take an initial look at the gap between the upper and lower cowl. While I don’t want to try to get it precise until the hinges are riveted, I’d at least like to get a rough fit done. In particular, right now there are a couple spots where the upper and lower cowl halves make contact and interfere a bit.

I think I’m also going to add some material back to the upper cowl and rework some of the gaps there. There’s one spot in particular where I ended up with a much larger gap than I really wanted.

Worked on a few different things today, all of it looking towards getting those hinges riveted in place. First item of business was laying out and rough cutting the cowl pin covers. With both cowl halves in place and pinned together, I used the cover templates to trace the outline off each side of the cowl:

Then I super rough cut the openings with a spiral bit in the Dremel – that worked mostly OK but boy does that thing like to wander around. Next I used a rotary file bit to shape the edges and get them a little closer to the cut line. The final cut will wait until after I’ve riveted the hinges so I know everything is fixed in place:

Well, mostly. After working on the layup on the upper cowl (more on that in a moment), I went ahead and sanded the cutouts in the lower cowl to snugly fit the template covers. These cheap covers are slightly oversize compared to the actual ones; a snug fit for these templates will allow for a nice paint gap around the final covers:

OK, so about that layup on the upper cowl. I decided I didn’t like how the gap ended up between the right side of the upper cowl and the firewall, so I wanted to add some material so I could rework the gap. The basic idea here was to sand a scarf joint and lay up a bunch of layers of cloth to overhang the edge. To get the contour of the cowl right, I clecoed a piece of aluminum scrap in place on the outside and covered it with packing tape to keep the fiberglass from sticking:

Then I laid up eight layers of cloth on the inside, which should give me about the same thickness as the original cowl edge. Once this cures, I’ll have a lot of sanding to do to clean up the scarf joint here, before I even get to working on the actual edge:

Finally, while working on other stuff, I had an idea for improving the cowl alignment at the firewall. The basic issue here is that the very upper edge of the lower cowl is unsupported, along with about a 6” portion of the upper cowl in the same area. So while everything else is pulled into nice alignment the upper edge of the lower cowl sits a bit proud of the fuselage skin:

The construction manual suggests possibly drilling a hole in the firewall for the hinge pin to slide into, this providing some additional support for this area. That seems like a decent idea, but I’m not a big fan of unsealed holes in my firewall, no matter how small they are. But I realized that I might be able to add some aluminum blocks inside the firewall flange, which could in turn force the pins inboard and pull the cowl edge into alignment.

Some experimentation with scrap from my panel cutout looked promising, so I cut a couple pieces and match drilled them. I also added a taper on the forward edge of the spacers, to help guide the pins into place as they’re inserted. These pieces still need some final trimming and cleanup, but the idea seems sound:

Much better alignment with the spacer in place:

I guess the next order of business will be cleaning up that flange extension – as mentioned, lots of fun sanding in the works there. Then maybe I’ll see about riveting all these assorted hinges in place…

So the last two nights’ work has been doing lots and lots of sanding on that extension piece I grafted not the cowl. The basic idea here was to sand a scarf, lay up multiple layers of cloth, and then sand the overlapping area back down flush. This was, to put it lightly, quite a bit of sanding. Especially since resin and cloth makes for some pretty hard stuff that doesn’t exactly sand easily.

It also makes it more fun that this is a contoured surface, though at least it’s not a compound curve. Still, that means my normal tools for trying to sand a straight line, like a nice sanding block, are no good here. I ended up wrapping sandpaper around a deep socket – yes, you’re reading that correctly – to make a sort of round sanding block. Then I went to town with that thing, for…a lot of time. I think it was about 45 minutes in that I really wanted to go tell Past Philip that that old gap really wasn’t that bad after all…too late now, though.

Tonight, I finally got things sanded down to a relatively flush point, so I switched to a soft foam block and 60-grit paper to finish the blend between the old and new glass. The result of all that work was this:

Next up was getting started on re-trimming that extension. I basically repeated the initial trim steps – laying the cowl in place, marking a conservative initial cut line, then gradually sanding the edge to fine-tune the fit. After a couple iterations, I was able to pin the two cowl halves together again. The gap still needs some work, but I think I’m going to go with the approach of riveting all the hinge halves in place before I remove any more material:

Oh, and I’m still not done sanding here, even ignoring the gap itself. I took a wild guess that eight layers of cloth would be about the same thickness as the rest of the cowl, but that looks to have been too many layers. The new grafted edge is a decent bit thicker than the rest of the cowl, to the extent that I can’t pin the upper cowl to the firewall right now – too much misalignment. So I’m going to have to work on thinning that extension down some before I can put this whole task to bed.

In retrospect, it probably would have been smart to do some test layups to get an idea of the finished thickness of different layer counts. But, much like the whole “that gap wasn’t so bad after all,” that ship has sailed, and there’s not much to do except…more sanding…

I’m back with more cowl updates, after a May that included vacation followed by all sorts of annoying adult issues, including a broken-down car, a broken-down home AC unit, and that one nasty storm we got to enjoy. But I’ve been back at work this past week, and today I ended up with some nice milestones taken care of.

Earlier this week, I picked back up on sanding down that cowl extension. Previously, I’d gotten rid of the extra protruding material, but the thickness in that area was preventing the upper cowl from fitting in place. So I spent some time working on that area and getting the thickness back in line. To do this, I set my calipers to the “factory” thickness, and used that as a sort of go/nogo gauge as I sanded excess material from inside the cowl. Once all that was cleaned up, I rehung the cowl again and verified that I could get all the hinge pins through.

At that point, I felt like there were no remaining excuses for not going ahead and getting all the various hines riveted in place. So I started working on prepping the holes, starting with a whole bunch of countersinking. Originally I was hoping to epoxy and rivet the hinges in place yesterday but then I did the wrong thing…I went online to get advice on this procedure. Before long I was swimming in opinions: Normal epoxy isn’t OK here, I should use special flexible stuff! Or maybe just Proseal! Also the normal rivets are too hard, they’ll crush the fiberglass, there are special soft rivets! And on and on it went…

Here’s the thing, the -8 plans and manual don’t even say anything about attaching the hinge pieces. Nothing about epoxy, they just mention drilling rivet holes and I assume after that they expect you to just rivet the things. Even my idea of adding epoxy for reinforcement is going the extra mile, but as per usual, I got all tied up in going with the right extra mile.

Anyway, I eventually mentally slapped myself in the face a few times and decides to just go with an epoxy/flox mixture. So today I got to work with the real prep – sanding down the mating surfaces (especially on the smooth fiberglass) for better adhesion and cleaning everything up. Well, the adhesion promotion issue brought forth another concern – the firewall joint hinges all have shims underneath them. Should I epoxy between each layer? I decided that was overkill, and instead went with a “flox rivet” technique. This is where you drill large-diameter holes in the hinges, and fill those holes with a thick flox putty, this providing some extra mechanical bond. So I also made those holes in the firewall hinges and then had to deburr those as well.

Finally, it was time for the fun. I mixed up a thick but still somewhat runny flox mixture – I want to have some bulk for adhesion, but I don’t want the stuff so thick that it doesn’t want to squeeze out. Next, I used a brush to paint the mixture into the mating areas in the lower cowl, then I got the hinges clecoed in place.

Now, originally my intent was to just cleco the stuff together and let the epoxy cure a bit, then come back and rivet. But i noticed pretty quickly that my epoxy squeeze-out was filling my countersinks, which would mean having to redo them, which seemed annoying. So I decided on the spot to go ahead and rivet while the epoxy was wet. Fortunately I’d already determined the rivets I needed and had them laid out, so this wasn’t a total scramble. It was still a sorta messy experience, though, and there was lots of tool cleanup afterwards.

I also discovered that some of my countersinks on the cowl split line were a little too shallow, leaving the rivets proud. Normally this would be a show-stopper, but in this case I went ahead and set them. I’ll come back later and sand them down, and eventually the whole cowl will be getting a skim coat of epoxy and primer that will smooth everything out. But I’m definitely going to recheck the countersinks before I repeat this procedure on the upper cowl.

So at long last, I have a lower cowl that’s entirely devoid of clecos:

I guess tomorrow or Tuesday I’ll get the upper cowl assembled, and then I can look forward to the fun of working on finalizing all the gaps everywhere. That’ll be followed by the real fun of getting back to work on matching up the spinner and forward corners, which I left off on so long ago…

Tonight was basically a repeat of Sunday, but with the upper cowl this time. The only real change was that I made sure to check every single countersink to make sure I wasn’t going to have a bunch of proud rivets that needed sanding down like the upper cowl. After that, I went through the same routine of sanding mating surfaces and drilling 3/16” holes in the aft hinges to allow for some flox bleed through and hopefully extra adhesion.

I did take a different picture this time, just to show how I mask around the hinges to make it easier to clean up the epoxy squeeze-out. I’m trying to keep the inside of the cowl halfway decent looking, after all:

One other thing I decided to do tonight was to assemble the cowl halves before finishing the work session. I noticed that the lower cowl got a lot stiffer where the hinges were bonded – not too surprising, I don’t think, given the extra material and tight bond – and I thought it was worth making an effort to have the upper cowl cure in assembly, to help both halves line up nicely going forward. Of course, I was really careful to clean up any epoxy squeeze-out here, lest I accidentally bond the cowl halves together.

It never seems to get old seeing an assembled part without clecos for the first time:

Now I’m really anxious to fit these halves back to the fuselage and see how everything looks together…then I guess it’ll be on to the fun of starting to work on cleaning up gaps. Or maybe I’ll leave that alone for a bit and work on the pin covers instead. Or maybe I should clean up that micro on the front corners first. Ah, decisions…

So, with the hinges taken care of, I decided that my next step would be working on finalizing the hinge pin cover cutouts. Previously, I rough-cut these, and sanded the lower cowl cutouts to fit the cover, but I needed the hinges in place before I could put the cowl halves together and work in finishing the upper half of the cutout.

At first I started gradually opening up one of the cutouts, and got fairly close to getting the cover to fit, but then I started looking more closely at the front cowl corners. There were two issues here that I felt needed to be remedied before finalizing the cutouts. First, the upper and lower cowl halves were making contact here, especially on the right side, when there should be a bit of a gap there (for paint). Second, the previous misalignment between the halves was still there, particularly in the area leading right up to the cover cutouts. It seemed counterproductive to carefully work on those cutouts and then slap some filler in the area.

So I switched gears – first, I worked on reopening a minimal gap between the halves in the corner area, then I switched to working on the filling. First task before filling was to do some more rough-sanding of the last micro application. For one thing, I needed a relatively even surface to put more micro on top of, but I also wanted to sand roughly to the lower corner contour so I could have a better idea of where I needed to build things up.

The latter turned out to be a good idea – once I’d sanded roughly to match, I could see that the contour leading up to the corner os each air inlet was a little funky looking. Basically, the area near the split was built up to match, but moving upward from the split, there was a pretty obvious slope back to the original contour, which looked a little funky. So I decided that this next application would cover more area vertically, to provide a better chance for contouring things.

I also decided to put in some flox inside the lower corners. Doing this gives me some structural filler, so if I decide I want to sand back the lower cowl a bit, I’ll have more room to work down there without worrying about thinning things out too much. But before getting to that, I made up some forming aids. Last time, adding the micro was tough, since I wanted enough excess protruding beyond the cowl cut line to allow me to sand it back to a nice flat profile, but that also made getting a good buildup tough. This time, I took some scrap posterboard type material and covered it in packing tape. The idea was to pin the cowl covers together for this work, and trap the posterboard between the two, thus giving a nice flat edge for the flox/micro to “mold” against.

I did the flox layup first. I think I could have added more flox to the epoxy; it seemed stiff in the cup, but kept wanting to sag after I’d applied it. The taped-up cards came in handy here; in order to address the sag issues, I flipped the lower cowl upside down, with the cards under the corners, trapped against the work table. This way the flox mixture was sagging into the joint, not out of it. I left the cowl like this for about 45 minutes, which gave the mixture some time to set up; after that, the sagging wasn’t a problem any more, but the epoxy was still malleable enough that I was able to shape it the way I wanted, before pinning the cowl halves together again, with the cards in place. I ended up flipping the assembly upside down just in case I got any more sag.

A few hours later, once I was confident the flox was definitely cured enough to not sag, I flipped the cowl again and started working on the micro. I drew a reference line on each card, to give me an idea of about how far out I wanted to build with micro. Then I started by using a small popsicle stick to work the micro mixture into the “corner” formed between the cowling and the cards – I figured this would be a likely spot to create a void if I wan’t careful. Then I worked on just slapping on more micro and trying my best to same it into reasonable shape.

As is always the case, the end result looked pretty crappy at first, but instead of focusing on trying to make it super smooth, I instead focused on trying to get rid of any potential voids from trapped air. Then I let the micro cure or about half an hour – like with the flox, I came back out and massaged it into a more pleasing shape while it was still malleable.

Still looked kinda crappy, but the card confined it nicely:

After waiting another couple hours for the micro to firm up, I pulled the halves apart and got the card out of there. This time I definitely got plenty of micro in place…I don’t think I’ll run into the problem of not having enough material to sand down to the right contour. Instead, it’ll be having to sand a lot off:

I actually got a head start on the sanding tonight. Since the micro was stiff, but not super hard, I tried shaping it with a Surform tool, kind of like you might do with Bondo. It worked pretty well, and while there will still be plenty of sanding, I at least took off a fair amount of material and got a head start.

Tomorrow: sanding time, HOORAY!

This is yet another multi-day post, actually covering a few work sessions over the past week or so. The basic deal is that I’ve been working on sanding and shaping those cowl corners, generally working in fairly short sessions because it’s been in the 90s almost every day. However, tonight things cooled off nicely and I had what was actually a fairly pleasant work session.

I’m pretty happy with how the corners have shaped up, though there’s still some fine-tuning I want to do, plus the usual pinhole filling to take care of. But I think I’m at a point where I won’t need to to another layer of micro and another round of shaping, which is a win all by itself. Here’s a look at the right corner, showing a void (“pinhole” just doesn’t cut it here) that I uncovered:

Most of what I worked on tonight was the prop spinner area, where I slathered on a ton of micro a while back. Initially I had the idea of sanding that in place, using sandpaper glued to pieces of aluminum, to ensure nice spacing from the spinner backplate, but previous attempts at that were…let’s just say not fun. Instead I worked on shaving down the micro to get it roughly to match the lower cowl, then hung the cowl on the plane again to check the fit. It’s actually looking pretty good, but I think I am going to revisit my sanding-in-place approach to get the cowl-spinner gap fine-tuned the way I want it.

Once that’s done, I think my next move will be to lay up the flange where cowl mates in the spinner area – this will effectively lock the cowl halves together so I can finalize the corner shaping. Right now, the cowl halves can move laterally relative to each other, which makes the corner shaping a bit imprecise.

I’m also considering adding flanges where the corners mate, which I’ve seen other builders do, but I’m not entirely sure it’s necessary, plus I think it might be kind of a pain to deal with in close proximity to the backing layups for the cowl pin covers. So I suspect I’ll end up not doing that, but you never know what strange decisions I might make down the road…