Avionics

I was really hoping for some cooler weather this week – we’ve been hearing about a potential “cool front” – but it looks like it’s not to be. And so today was just a normal stupid-hot Houston day, but I worked through it (mostly in short bursts, as per usual). I’d intended to get going on fabricating the lower shelf behind the panel – the one that will house the fuse blocks and other primary distribution stuff – but I ended up going in a different direction. As we’ll see later, this was actually a good thing.

Anyway, to start with I wanted to lay out the location of that lower shelf. The intent here is for the shelf to start at the lower edge of the cross-cabin brace behind the gear towers, and extend straight forward until it encounters the baggage bulkhead. The first thing was to figure out the vertical location at that forward end that would result in a level orientation of the shelf. This I worked out with a combination of a square and ruler, using a rivet line as a guide. With that, I was able to draw a level line on the baggage bulkhead, setting the forward attach point for the shelf.

Next, I wanted to get that bulkhead into the fuselage, so I could measure the distance from the brace to the bulkhead – which would determine the depth of the shelf. This sent me into a side trip of removing the rudder pedals, which was needed to get all the baggage floors in place, which were needed to locate the bulkhead precisely for measuring.

At some point, though, I changed and started working on the upper shelf. I don’t even remember why. The design here is similar – this shelf will start at the forward cross-cabin brace, which is higher than the aft one. At one point I was considering making this shelf actually slant from one brace to another, but that seems overly complex – so instead I decided to just have the shelf go straight back, and then have a 90° bend to a vertical portion that can attach to the aft brace.

So I did the measuring routine again, figuring out the depth of the horizontal and vertical portions. Then I cut off yet another section of my trashed wing skin, and got to work laying the part out. I wanted to also allow material at the ends of the shelf to bend down for stiffness, so that had to be accounted for. (I’ll likely eventually add another stiffener or two, once I figure out the device locations).

Then it was time for bending. Here’s the sheet clamped to the edge of the workbench, ready for me to apply some targeted physical violence with a hammer:

 After making that bend, along with the two stiffening flanges, it was time to take a deep breath and go for a test fit. Whew, it fits nicely!

 At this point I couldn’t resist trying a layout of my cardboard faux avionics boxes:

 This led me to an interesting consideration: namely, packaging things around the brake lines coming out of the gear towers. I haven’t really thought of these at all while conceptualizing all the avionics stuff. My main concern was having the EMS connectors hang off the edge of the shelf and interfere, so I grabbed the brake lines and installed them. To review, these lines go from the top of the gear towers to a point on centerline, just under the baggage bulkhead.

This highlighted an issue with my lower shelf that I hadn’t considered. My original plan was to simply rivet a section of piano hinge to the baggage bulkhead and the shelf, allowing it to pivot down and forward and allow access to the fuses. But with these brake lines running beside the bulkhead, that means the lowered shelf will interfere with them. Possibly not a huge deal – this shelf isn’t intended to be accessed in flight – but I’m not sure I like the ides of the shelf potentially swinging down and hitting the lines repeatedly over the aircraft’s lifetime.

The routing of the lines is visible here. Note the Sharpie line on the bulkhead; that would be the location of the forward edge of the lower shelf:

So yeah, I’m glad I didn’t go ahead and do the lower shelf and then discover this. It probably wouldn’t have been a huge deal – I’d probably just need to add some stops fo keep the shelf off the lines. I’ll probably still need to do that, no matter what I do, but at least I can plan for it instead of having to be reactionary. I’m thinking what I’ll do here is to rivet a length of angle to the bulkhead, and then rivet the piano hinge to that – that’ll offset the hinge point for the shelf by 3/4”. Combined with some sort of stop, that should make things work OK here.

Anyway, back to the upper shelf – I still need to lay out and make the attach holes on the braces, which will use screws and nutplates. I may wait on that until I get my avionics on hand – I’m a little concerned that my packaging designs from before might not work. One thing I didn’t account for was the mount points for boxes interfering with where the shelf overlaps the forward brace. I might end up having to do some creative stuff here, like mounting components vertically.

Oh, and I’ve got a bunch of tentative orders working right now. I need to get some quick-disconnect fasteners to hold up the aft end of the lower shelf – I think I’ll end up getting some quarter-turn captive screws from McMaster-Carr. I also spent way too much time looking at switches last night – I want to nail down an order for those this week, so I can have the switch panel(s) finalized before I get to building harnesses. And I guess I might go ahead and order my ELT from Spruce this week, and see about tacking on some assorted hardware since I’ll be getting free shipping.

…so much to do…

Sort of a two-parter here. Tuesday night, I did a pretty extensive inventory of all the avionics, in an attempt to document all the assorted fasteners I’d need to get things mounted. Once I had that done, I assembled a decent-sized Spruce order, and added the ELT because 1) I figure I’ll want to mount it semi-soon and 2) it allowed me to get free shipping. Now all that stuff’s on the way.

Tonight, I got started on the mount setup for the ADAHRS units. Originally I’d planned to build a sheet shelf like I have elsewhere, and put it right behind the baggage bulkhead, but I’ve rethought some of that. For one thing, I decided to move the mount back to just ahead of the next bulkhead. The reasoning here is to get some separation between the units and the baggage area. A concern here is ferrous material that might affect the magnetometer (aka electronic compass), especially ferrous material that moves around or isn’t always there – for example…baggage! This location will make maintenance less fun, like if I ever need to service or replace one of these, but that’s better than having them be unreliable.

I’m also fabricating the mount out of a combination of angle and thick stock. This is to 1) provide really good rigidity and 2) provide a mounting surface thick enough to tap for screws. In keeping with the ferrous-material concerns above, these units must be mounted with non-ferrous hardware, such as brass screws. I really don’t want to do screws and nuts here – I’m imagining trying to work a screwdriver and wrench back here to service one of these – and while stainless nutplates do exists, they’re harder to find and still not necessarily non-ferrous. So I’m just going to tap the mount and have the screws thread directly into it.

Anyway, all I did tonight was set the location of the rear angle piece and get it cut and trimmed to fit. Next comes the more fun part of laying out the second angle at the precise distance forward, and figuring out the fit of the plates that will tie them together. Tomorrow…

Not a ton to report tonight. I laid out and cut the second angle piece that will constitute the ADAHRS mount platform. Next up will be deciding how to add the thick plate material that I’ll be tapping for the attach screws. Originally I was going to bridge the angles together with that stuff, but I don’t really feel like there’s a need to do that, and it’ll just add weight. So now I’m leaning towards just cutting two bar-shaped pieces the same width as the angles and riveting those in place.

It was fun to do a little mockup at the end of the night though:

The first and most interesting question of the evening involved laying out the mount holes to match the ADAHRS units. Normally, I’d use the actual unit as a drill guide for something like this, but that wasn’t going to work for this – because the holes through the mounts need to be undersized, since they’ll get tapped for the 10-32 attach screw. Laying tham out manually seemed fraught with peril, especially since the hole spacings are 4.2” – not exactly a nice even division.

Eventually, though, I realized I could use the unit itself for this. I’d already drawn a line along the length of both angle pieces, marking where I wanted the mount holes to lie; I just needed a way to draw an intersecting line to mark the hole centers. So I clamped an ADAHRS in place on the angle, butted against the mark at the lateral center (ie where it would butt against the other unit), and then traced the inside of the mount holes. The outboard intersections between those circles and the previous mark would be my hole centers, and I could know they were the right distance apart since I used the unit as a guide.

So with those marks, I got the four holes drilled in one angle, then clamped both angles together and drilled through both of them to make the matching holes in the other angle. For the moment, I only drilled them to #30 – the final hole size before tapping will be #21, but I want to drill that final hole in assembly with the bar stock pads I’ll be adding to provide adequate thickness for the tapped threads.

As for those pads – last night I’d figured on just cutting 3/4” strips, making them the length of both ADAHRS units together, and riveting them to the angles. I decided instead to just use short pieces as needed; here again I can save a bit of weight and material. So there end up being three mount pads per angle piece: the two outboard ones cover the single outboard holes, while the center ones cover both inboard holes for the units. Each pad will be attached with a pair of flush rivets.

Anyway, there was just a decent bit of measuring and cutting and deburring and laying out holes and so forth. Also a brief break after my bandsaw blade decided to break – good thing I have a spare, or I would have been rather annoyed. I got to the point of having all the rivet holes drill between the pieces, and called it a night:

So tomorrow I’ll just need to mark the pieces so I can keep track of which goes where, get the rivet holes deburred and countersunk, then shoot primer on all this stuff before getting the riveting done. I’m going to wait until everything’s riveted together before I final-drill and tap the actual screw holes – I really want to make sure those are lined up properly.

I guess I need to think of my next mini-project – which I think will be putting together the switch console. That should provide a good solid couple days’ work…

Well, not the most productive day in terms of shop time, but progress is still progress. The first thing I attacked today was some reorganization. With the recent electrical stuff I got, plus some other hardware, I was kind of outgrowing the parts bins I had, and I picked up another one when I was at Harbor Freight last weekend. What started as “find homes for this stuff” eventually turned into a pretty involved shuffling of parts.

Among other things, there was a lot of ground to be made by consolidating hardware – because of the way Van’s ships hardware bags, it’s very possible to have the same item split up in multiple places, That was doubly true since I had some hardware I’d bought independently. In the end, I emptied out probably ten or so assorted hardware bags, either by combining items or just being able to dump the contents of s bag into a dedicated bin.

With that done, I was once again left to figure out something to work on, and I settled on fabricating a mount for the two Skyview backup batteries. I’ve had this idea for a while, as a way to simplify packaging behind the panel, and I even had a rough cardboard template that I made up a while back. I’d been unsure whether I wanted to do this or not, but there’s been a little too much imagination involved in considering the layout behind the panel, and I felt I needed something physical, even if it ended up not getting used.

The fabrication was pretty straightforward, again using some scrap .025” sheet that was lying around. Even though I had that cardboard template, I laid out the piece here from scratch, using an actual battery (unlike the template, which I made based on dimensions from the Skyview install guide).

After the first rough cut, I had the general shape – a central rectangular part that the batteries would rest against, two triangular ears to provide support, and tabs to hold the screw holes that will secure the thing:

A bit later, after cutting out the area between the ears, finishing all the edges, and laying out and drilling the mount holes:

Then it was time to break out the dead blow hammer and commence physical violence, using the workbench as usual as a bending form:

Some final tweaking with the hand seamers, and it was done. I’d intended to go ahead and screw the batteries in place, but…I need some longer 8-32 screws than what I have right now. So I had to make do with clecos instead:

The mounting of the batteries to the piece will use the aforementioned machine screws, along with nylock nuts. That’ll be a little tedious to install and remove, but the idea is that if I ever need to service these, the batteries and mount will come out as a unit. I also like how this setup puts both wiring harnesses right on top where they’re 1) easy to access and 2) don’t take up any of the shelf footprint.

I do think I’m going to add another screw hole in each mounting tab though, now that I look at this. Otherwise I’m imagining the center web of the mount gradually abrading the shelf since it’ll probably be able to vibrate a bit. I may even go wild and cut a lightening hole in the center web to save a teensy bit of weight…

With that done, it was time for yet another Tetris session with the avionics shelf. This time I ended up breaking out some clamps to accurately envision parts that will go in the baggage bulkhead. That matters since that’s where about half the stuff may end up going. And after a bunch more shuffling, I think I may finally have a workable arrangement here:

 Most of my previous plans had the EMS module (second from the right on the vertical bulkhead) on the shelf, to the far left. That module presents fun packaging challenges, since it has a DB9 Skyview network connector on one side, and two more D-sub connectors (37- and 25- pin) on the other. Having it on the edge of the shelf seemed like something that might bite me later; in that same area I’ll have scat tube coming up for the fresh air vent along with the brake line from the gear tower. It seems prudent not to pack this area too tightly with giant harness plugs as well.

This layout leaves lots of rook for those big connectors, and I think overall makes better use of the baggage bulkhead space than previous ideas. There are still a couple things to consider before I finalize this. First, I really need to get my CPI2 ignition ordered so I have the real unit to play with. I’m not too concerned about the actual control module (the cardboard piece far right), but the backup battery mount may not fit where I think it will (but boy I sure hope it does). Second, I also need to consider the various D-sub connectors on the stuff in the panel, especially the displays. Which makes me think that maybe I really need to get the panel cut.

So as I mentioned last night, I think it’s high time to go ahead and get this panel cut so I can convince myself for once and for all that my avionics shelf layout is solid. Plus I’ll need this done before I can start finalizing the harness behind the panel as well.

A perennial question has been whether to cut this by hand or have it CNC’d, and I’ve decided to go with the former. My motivation here is that I feel like I’ll spend as much time making a precise CAD drawing (mostly figuring out how to do so) as I will just doing everything by hand. Also pertinent here is that I haven’t finalized the exact location of some things, which I think is easier done drawing on actual parts.

So with that in mind I printed out assorted layout sheets and went to work with a ruler and a Sharpie. The first order of business was to figure out the vertical location of the whole group of stuff; basically, I want to place everything as high on the panel as possible. The limiting factor there ends up being interference between the brace behind the panel and the upper outboard corners of the intercom and knob panel. Those flank the GPS-175, which is centered on the panel, so it was pretty easy to draw out the horizontal edges of the bezel outlines and then figure out the highest point the bezels could be.

Once those three items were laid out, the rest of the design flowed from there, and I got the bezel outlines traced out for the two displays and the radio and AP panel. I took some time at this point to hold various components up to their outlines just to sanity-check that I hadn’t done anything grossly wrong.

Next up was working out the actual panel cutouts. The displays are pretty easy; they just have slightly smaller rectangular cutouts. The same is true of the intercom and GPS-175 cutouts. The other three Dynon units are a little more fun, as they require sort of cross-shaped cutouts – I guess they needed extra space in the modules to fit more stuff? Whatever the reason, those were a little more tedious to get done, but not horrible.

So now I’ve got all my cut lines all laid out. I haven’t bothered to mark any of the attach screw locations; my intent is to use the actual components in place to mark and drill those. That way I can clamp everything in place where I want it and ensure it all lines up properly; there will be enough room to work with the cutouts to adjust positioning as needed.

Yay, lines:

The remaining question is how I want to actually cut this. Typically I’ve done stuff like this with a Dremel cutoff wheel, but it’s not especially precise and will probably not like the thick material here. I’ve seen where people mention using a jigsaw with a metal cutting blade, which I have, but I’ve never been a huge fan of jigsaws for making precise cuts. The other option, which I looked into last night for a bit, is to buy something like a Dremel SawMax, a compact high-speed circular saw. They’re supposed to work well for making straight cuts, though I think it might be a bit much for the smaller cutouts. I’m somewhat tempted to get a cheap vibratory saw and see how it does against aluminum.

So…we’ll see…

Yup…today it was time to start making a whole bunch of cuts. Last time I was trying to decide what tool to use, and after some research, I ended up picking up an oscillating Dremel multi-tool. I was far from certain it would work well for this purpose, but based on some feedback from friends, I decided it would be worth having one around in general even if it didn’t work for this. Which is a good thing, because after some testing, I decided it wasn’t the right tool for cutting the panel. It’ll go through thinner sheet quite well, and is very controllable, but for thicker material it’s slow going…and I figure the slower the going, the more time I have to mess up a cut.

So in the end I went with the good old jigsaw. The main downside to this was that it meant having to make a lot of big pilot holes in which to start the saw – unlike the plunge cuts I’d hoped to make with the oscillating tool. So the first order of business for the day was to lay out and drill a bunch of holes – first, #30 holes at all the inside corners I’d be cutting, then 1/2” entry holes for the jigsaw blade. Essentially, I spent about the first hour of the day just drilling.

Then, with all the various holes done, it was time to go to work with the jigsaw. Not much to go on about here, just lots and lots of careful cutting, plus frequent pauses to sweep/vacuum up the copious aluminum dust I was making. Here’s a look at the state of things after making the first couple rough cuts:

Eventually, I had all the rough cuts done. I intentionally cut a bit away from the final cut lines, wanting to carefully sneak up on final dimensions instead of risking overdoing anything. Another fun part of this was dealing with clearance issues that affected the jigsaw. For the lower right cutout below, the stiffening lip at the bottom edge of the panel made it impossible to effectively cut the two outside vertical lines – hence why that cutout is a bit rougher looking than the rest. I got creative to remove as much material as possible with the jigsaw, but there was just going to be more to take off later for that one:

Next came the tedious part of the process: getting those cutouts up to size. I started with the large display cutouts, which are pretty straightforward but also pose a bigger challenge in getting the long edges straight. The first display cutout took me probably an hour or more – I was really being conservative in sneaking up on final size. The second went a lot faster, since I had a better idea of how the cut lines I’d drawn related to final size. Basically, it was a case of removing the lines entirely (vs removing material right up to them) for a good fit.

The tedium paid off with some motivating test-fit action:

Then it was on to the real fun: working with the sort of cross-shaped cutouts for the AP/knob/radio heads. Working up to the fit here was a whole lot more fun, since it wasn’t just a matter of “the cutout’s too narrow here” – there was also the question of whether it was the inner or outer edge that was too small. And here I repeated the previous performance with the display; I probably spent an hour and a half getting the first piece (the AP panel) to fit, and afterwards decided it was time to break for dinner. After that, the other two I got done in less than half an hour each.

And at this point the test-fit action was even more fun to look at:

The good news is that this is probably the worst of this work. The intercom and GPS-175 cutouts are fairly normal squares, none of the crazy stuff, and should go faster. The one thing I need to figure out is how I’m going to attach the GPS-175 rack to the panel – probably something to ask around on VAF about.

Once I get the other cutouts done, I’ll do a final test fit of all the components at once, and then I’ll actually drill the screw holes and get the nutplates installed; by doing this last, I can ensure that everything line up really nicely while I’m doing the holes, though to be honest everything is looking great so far just fitting snugly in the assorted cutouts. Then I can actually hang this thing in the fuselage and evaluate my behind-the-panel layout.

(I’d also like to climb in the thing and play with the avionics while making airplane noises, but that would require taking the fuselage off the rotisserie…not really worth it. Later…)

So this is sort of a catch-all post, and also includes some work from a few nights ago that I never got around to writing up. The first thing one might notice is that this post is not the “panel cutting part II” that one might have expected, and that’s where the story begins.

Before I even went out to try and work Monday evening, there were some questions about whether my panel cuts were going to work. One thing I didn’t think through before I started cutting was how I’d attach the GPS-175 rack to the panel. For all the other components, the attachment is really simple; the GPS rack is a different matter. Basically, it’s a sort of sleve that the GPS unit slides into, and how it gets attached to the panel is up to the installer. The most common way would be a short piece of angle on each side of the rack, one side screwed to the rack, the other riveted to the panel. But since I put the knob panel and intercom right up against the GPS, there was no room for that. But I had a couple idea I was going to tinker around with.

That all became a moot point as I was preparing to finish the last two cutouts, because I noticed that I bungled the intercom cutout. Basically, when measuring in from the device outline to lay out the cut lines, I switched the measurements, and as a result cut a hole that was too narrow and too tall. And because of the way this worked out, I’d cut out the material needed for the attach screws. No coming back from that one, unfortunately. But it didn’t bother me too much, since I was already considering that the GPS rack might be a deal-breaker.

I still got some stuff done that night though, just went looking for menial stuff, which resulted in me digging out the antenna doublers I made a while back and getting their rivet holes dimpled – that’s in preparation for the fuselage riveting session that I really hope I can get done soon.

That brings us to tonight. I’ve got a new panel blank on order from Van’s, which should be here early next week. In the meantime, I needed to find stuff to work on – in fact, I need to generally get in the habit of finding stuff to do even if my “main” desired task is held up by something. I’m never going to finish this if I can’t find ways to be productive outside a narrow area of focus.

So I started by going ahead and jabbing up the GPS rack attach stuff. Nothing too exciting here – I cut a couple short pieces of angle, shortened one leg from 3/4” to 1/2” to minimize the amount of dead space I’d have to put between the GPS and its neighbors, marked and drilled it for the screw holes in the rack, and installed nutplates in the angles. Got those installed in the rack for a test-fit, and they look good:

Note that they’re still oversize; I’ll need to remove them and trim to final size before I fit them to the panel for real.Also worth noting here is that the real fun is positioning these angled properly relative to the forward end of the rack. The rack needs to be carefully located relative to the panel surface for the GPS unit to sit flush like it should. I ended up here with the angles a bit further back than I want, but that’s the error I’d prefer; I can add shims in between the rack and panel if needed, and I expect I’ll end up doing just that.

OK, now what to work on? Well, there was a new service letter released for most of the recent tailwheel RVs a couple months ago. Not a mandatory service bulletin, but something recommended to be done if desired. Basically, there have been instances of deformation of the last tail bay, believed to be due to stresses from the tailwheel, specifically if the tailwheel hits a significant bump while turning – thus putting a strong side load on this area. The fix is to add a diagonal stiffener on each side of the skin in this area to provide some reinforcement.

I decided this was worth doing preemptively, and so when I ordered some parts from Van’s a while back, I included the two stiffener pieces, and I decided tonight was a good time to take on this project. Overall, it’s pretty straightforward.

First up, the two end rivet holes for each stiffener are located. The service letter provides instructions for these; basically, you draw an arc a fixed distance from a few reference rivets, and use the intersections of those arcs for the rivet holes (note the marks upper left and lower right):

These holes are used to initially locate the stiffener on the outside of the skin, where it’s used as a drill guide for the other three rivet holes. After some demurring, it’s time to cleco the stiffeners in place inside the skin and check for interference. It may not be super obvious from this photo, but the aft end of this stiffener overlaps the aft bulkhead flange, so it’ll need to be trimmed:

After trimming both stiffeners (they both had the same interference with that bulkhead), doing some trial-end-error to get everything right, I got them all polished up on the bench grinder and dimpled. Then it was time to dimple the holes in the skin. The upper/forward three were easy enough to do with a 3” yoke, but the other two on each side required breaking out the pop rivet die set – which is really hardly up to the task of dimpling this thick skin. I ended up dressing those dimples with a countersink bit to get the rivets to sit properly flush – the dies themselves didn’t quite cut it. Now the stiffeners are clecoed in place and ready for riveting:

Since it was nearly 10:00 now, and since some of these rivets will have to be done with the gun, and since I want to be nice to my neighbors, that’s where I called it a night. I figure I should be able to finish this up tomorrow.

New panel blank came in today, so I decided to get right to work on it this evening. First step, like before, was to do the layout of the actual devices and the required cut lines. Adding a slight twist was the need to modify the layout a bit; as mentioned before, I need to provide some empty space to the sides of the GPS-175, instead of putting the two Dynon units right up beside it. I decided to complement that bit of empty space by also separating those two units from the screens below; this will probably be more clear as we get to the work here.

First step is to mark the no-go areas of the blank. The removable panel attaches to two side “wings” with screws; those wings are in turn secured to the fuselage. The wings sort of arch together in the top center area, and the end result is an overlap where I can’t really safely cut any holes. So I just checked the wings into place and traced the edge with a sharpie:

Next up is working out the locations of the top three units; I want them as high as they can go without interfering with the wings, and their position essentially drives everything below. So I start out by marking the lateral extent of these. First up is a centerline, then I line out the outside edges of the GPS-175, followed by the empty space required for the rack mount, and finally the knob panel and intercom units. Before going any further, I laid the actual units in place to sanity-check my marks:

And then the intersection of the outermost marks and the “no-go” zone becomes the upper alignment line for these three things. With that as a baseline, I can then draw in the lower edges of the GPS and Dynon units. Additionally, the bottom edge of the GPS becomes the reference line for the top of the displays and other stuff below.

Here the spacing under the two Dynon units becomes more obvious; they’e spaced out about 1/2” from the edges of the GPS, and then by aligning them with the top of the GPS, I get about 1/4” margin between them and the screens below:

Next up is marking the edges of the two screens, plus the radio and AP panel alongside. I also paused to clean up some of the superfluous Sharpie marks to remove any confusions. This is required a folded paper towel soaked in acetone and a very steady hand:

This finishes up outlining the actual bezels of the units; that i the physical space they’ll take up on the front of the panel. Next up is the real fun, drawing the actual cutouts, using the outlines as a reference. It’s just more of the same as before, lots of careful marking and measuring with a ruler and Sharpie. Once that was all done, I did another cleanup session of superfluous marks. And now it’s all ready to go (side note: it sure is fun trying to take decent photos of a reflective surface):

Next up, I get to repeat the fun exercise of doing rough cutouts, followed by lots and lots of filing and such.