CO Detector, Cowl Fit

The miscellaneous accomplishments of the night:

I don't like that the backup latch at the top of the canopy swings under the roll bar and doesn't have any barrier to keep it from banging against the paint.  Since this is just a backup up catch to keep the canopy from lifting if the main hooks fail, I thought about just setting it up so it didn't touch the roll bar at all in normal use (although if the canopy does try to depart, I don't really care if the latch destroys my paint keeping everything in place!).  I can see myself accidentally swinging the latch around before the cockpit is fully seated though, and in that case it would gouge the roll bar pretty good.  So I dropped the latch down about 1/8" below the roll bar height and installed a strip of UHMW on the aft roll bar channel.  This at least gives me a little protection.

Next up was installing the Aithre CO detector.  It's a super lightweight box that can go anywhere in the cabin.  At first I was looking for a place to bolt it, but it's so lightweight that I decided a strip of adhesive velcro was plenty (and in turns out, that's an option in the install instructions).  I put velcro on the cross bar forward of the subpanel and slapped it in place.  I actually prefer this approach because for anything hidden by the subpanel, maintenance involves crawling underneath the panel.  If the CO monitor has to be taken out, it's going to be far easier to just pull it off the velcro than contort myself to undo nuts and screws.

I spliced the power, ground and data wires onto longer pieces and ran them up to the Backup EFIS Dsub connector on the ACM.  Getting to the ACM connectors is easier said than done with all of the wires in place now, so I took the easy way out and just removed the PFD screen.  That gives easy access to working behind the panel.

Once it was all wired up, I configured the MFD for the monitor (serial port 2 on the MFD) and set some basic alert levels.  Anything in the caution or danger levels will have alerts that show on the screens, as well as provide audible warnings.

I believe that's actually the last bit of wiring that has to be done behind the panel itself!  There's still all of the engine sensor wiring, but that's already in place and is just a matter of cutting wires to length on the engine side and connecting to the sensors.

Now comes the part I've been dreading...working on the cowling.  I made the first cut on the top cowl to cut it to length so the fit is fairly close to the firewall outline.  I came very close on it, but it looks like I got off about 1/8" in one section where the firewall changes angle.  So unfortunately I'll have to do some surgery there and add a little fiberglass to extend it again (it's only about 3").  Given the horrid state of the cowl, I'm going to have to get used to doing fiberglass repairs.  The fiberglass parts are the laughing stock of Van's, and I see why.  It's going to be a nightmare to make it all fit together nicely.

I spent an hour slowly sanding away the flanges of the nose bowl where the top and bottom cowl nest together.  While I got it close-ish, it's still horrendous and has a long way to go.  I just need a decent fit for now so I can start to work on the hinge connections and firewall connections.  Once that's all done I'll come back around to making it all pretty.

Hurry Up & Wait

The ELT has a little beeper that goes behind the panel to alert when the ELT is going off.  It is connected to the ELT and to the panel activation switch with standard household telephone wire, which is admittedly a little goofy.  Had I known how this was set up, I probably would have bought a different model, but hindsight is 20/20.  This box has a battery that'll have to be changed out every so often.  It's super light, so I just put some adhesive backed velcro on it and stuck it to the back of the subpanel.  That'll make it easier to just reach behind the panel and pull it down when I need to change the battery.

I attempted to install the backup alternator, but no joy.  The oil filter extension that Titan gave me that bumps the filter aft a few inches to clear the engine mount gets in the way of the alternator.

You can see that the alternator won't seat completely against the accessory pad because the bumps for the bolts that hold the alternator case halves together rub against the oil filter extension.  I got in touch with Titan, and they said they do have a 1.4" extension (vs this 2 1/2" one).  I took some measurements, and that will be right on the edge.  It still might be about 1/16" long, but we'll see.  Worse case scenario, I can take the smaller extension to a shop and have them mill about 1/16" off.  Now I just have to wait for the new part to arrive.  I have to hand it to Titan though - they didn't ask any questions and gave me a UPS tracking number within minutes, no charge.

Panel Power

Most of my work these days is spent fiddling with individual wires here or there, so not a lot to take pictures of.  I do have what I think is the final shot of the area behind the panel though.  I had forgotten to pull the two wires for the defrost fans that live in the canopy frame.  I didn't feel like undoing a bunch of zip ties, so I just added them to the bundles and ran them to the center area of the panel.  I have to provide enough play that they can move around as the canopy is opened/closed.  Once I get the canopy back on, I'll put some protection over them.

I also ran the static air tube to the G5.  I put a T on it to take it to the alternate static switch first.  With the switch off, the static air is pulled from the ports at the rear of the plane.  If I ever need to (unlikely), flipping the switch up opens the valve at the back of the switch and allows cabin air to feed the G5 instead. Once I'm flying I'll have to document what the difference is between the two so I know what to expect in terms of changes in instrument indications with the different air sources.  The only time it would ever be used is if the static ports ice up or get clogged with something - again, unlikely since this plane is not one that I'll get anywhere near icing, and there is a port on each side of the plane, so clogging one doesn't necessarily mean a complete clog.

With all of that base wiring done, I decided it was finally time to flip the switch and bring the panel to life!  I still have to do all of the FWF sensor wiring, but want to at least make sure the rest of the wiring is correct before I get too deep into the rest of it.  When I first flipped the master switch....nothing.  Turns out it's important to actually connect the master relay wire to the relay itself.  Duh.  I had left that off until now, not wanting to accidentally let power through when I wasn't ready.  Once I remembered that and connected the relay, everything powered up right away and all of the smoke stayed in the wires!

So at least I know that all of the devices work.  Now the big hurdle is configuring them to actually talk to each other and read all of the sensors.  The big upside to the AFS setup vs Dynon is that it has a lot more serial port options to allow for more peripheral devices (5 per screen).  The downside is that those serial ports aren't all shared between screens, whereas with Dynon (which only has something like 5 ports total) shares all of the port data with both screens.  That means with Dynon, if you lose one screen, you still have control over all of your peripherals via the other screen.  With AFS, if you lose one screen, you lose the peripherals that are linked to that screen and not shared with the other screen.  It gets even more confusing, because that's just the serial ports.  With AFS, all of the devices that use the "skyview network" are shared between screens, regardless of which screen dies.  So that means you'll always have ADAHRS, com panel, autopilot panel, and engine data info on both screens, even if one screen dies (so basically, all of the "get me on the ground" info is available on both screens - plus also having the backup G5).  Risk mitigation with AFS is just about ensuring you configure the serial ports on each screen so that you have necessary controls on the other screen should one die.  For example, once I put in my IFR GPS and second radio, I will configure one screen to be connected to comm 1 and the Dynon GPS, and the other screen will connect to comm 2 and the IFR GPS.  That way if one screen dies, I still have a comm and GPS input coming through the other screen.  Likewise, if both screens were to die, then the G5 would still show all flight attitude info, and I could still use the radio control on the panel itself (vs controlling through the EFIS).  The only things I could potentially lose completely would be my transponder and ADSB, if that screen they are linked to went belly up.  Neither of those present a flight risk though.

All of that is a long winded way to say that my next step will be to sit in the plane for a few hours and configure each screen.  I'll have to take the install manual out on my ipad and just go page by page through it.  It's a lot of configuration!  Everything from entering in the tail number so it can generate the right ADSB code to entering in all of the flight envelope speeds, sensor types, and hundreds of preferences.  One thing I noticed is that I haven't been able to get a GPS signal in the garage.  I temporarily attached the GPS pucks on the plane, but at least with the garage closed, they haven't been able to connect.  I'll open the garage and see if that makes a difference.  I'll probably go ahead and connect the comm antenna as well (even though all of the antennas will have to come off when I paint), just because I don't want to accidentally hit the transmit button without an antenna attached.  That can blow the control module.

Wiring Never Ends

Most of the wire routing directly behind the panel is cleaned up now.  The only remaining connection to finish is the "front" airframe connector to the ACM.  I just need to pull the field wire from the alternator and pin it in that Dsub, then it'll be all set.

The area forward of the subpanel is getting close.  There are just a handful of random wires to run yet, then it will be as organized as it can be.  While it's not pretty because of the various wire lengths I had to deal with (lesson learned - the next plane will be better), it's all solidly mounted.

To do the rest of the firewall forward wire routing, I needed to get the scat tubing for the heater muff inlet in place.  I temporarily put the right aft baffle in place and got the scat tubing routed around the engine mount.  No matter what I did I couldn't keep the tubing from rubbing against the engine mount though.  I used a couple of adel clamps and put RTV between the tubing and engine mount to minimize chaffing. 

Routing the alternator power and field wire back to the firewall is a little awkward.  I made some steel standoffs and attached them to the engine using the oil sump bolts.  Then I attached the adel clamps to the standoffs.

I used the same approach on the other side of the engine for the starter power wire.

More Panel Wiring

There isn't much I can share that looks different in photos, but I'm continuing to truck through the panel wiring.  Even though I have most of the wiring sorted out now in terms of connections, the fiddly part is in figuring out the routing.  I'm trying my best to keep it as organized as possible.  I'm succeeding in some areas and failing miserably in others.  Except for a couple of remaining connections to the fuse block on the right side, most of the wiring behind the panel is in place and solidly supported.  The left side between the ACM and the PFD is incredibly tight, which makes clean routing next to impossible, but I think I've got it about as good as it can get given the space limitations.

The area forward of the subpanel is a different story.  I still have a lot of work to do there.  Part of the issue is a lesson learned from a year ago when I ran my wiring from the other areas of the plane.  Based on what people had said, I made sure to make the wire harnesses just a little longer than necessary to account for service loops behind the panel.  Knowing what I know now, I don't think I'd do that again.  For one, with modern electrical designs, nearly everything in the panel is easily removed just by unhooking Dsub connectors.  With a slider canopy where the back of the panel is only accessible after you've removed the glass screens, a few extra inches is all that's needed to unscrew them, then reach up and remove the connectors.  With a tipup canopy, I don't even have to do that!  All I have to do is open the tipup, and I have unrestricted access to everything.  So I think this "service loops for everything" idea is kind of antiquated at this point, depending on the aircraft design.  For me, the service loops are causing me nothing but grief.  I don't have any room immediately behind the panel to coil up all of the extra wire from my harness runs to other parts of the plane.  That means that I have to bundle up all of that extra wire in forward of the subpanel.  It creates quite the mess!  I'll be able to tidy it up once I sort through my engine systems wiring and all of wires that need to traverse this area to go to the aircraft ground, but in hindsight, I'd cut all of my harnesses shorter and just make them the right length to get to the panel.  Service loops that have to be forward of the subpanel means that if I ever need to remove or change a wire, I have no choice but to crawl under the panel and cut a bunch of zip ties to find my wire in the sea of loops.  It would be far easier not to have the loops to begin with and have no extra wire length in inaccessible areas.  Oh well, lesson learned.  There's no shortening things up at this stage, and it's fine.  It's just an assault to my senses and an extra pound to the plane.

Taming the Rat's Nest

It seemed like there was plenty of space behind the panel until I actually started filling it with wire harnesses today.  I don't think I quite grasped how full the ACM would be.  By the time I was finished plugging things in, it was so full I had a hard time even tightening the Dsub screws.  The downside of having harnesses from AFS is that they aren't the length I would have made myself.  So some fit perfectly, and others have enough extra that I have to make a tiny loop.

I used the panel stiffeners as a place to put adel clamps and run most of the wires from the right side of the panel to the ACM.  That bundle is going to be really close to where the canopy frame swings when closing and opening, but I think it should clear.  If I find it doesn't, I will rotate the adel clamps around and put the wires on the aft side.

I installed the headset jacks.  The lemo plug is the black one on the outside, and the standard 2 plug headset jacks are the silver ones.

Next up was the wiring for the Pmag switches.  I still need to clean up the routing of those wires, as well as run the wire for the master switch. 

Panel Assembly!

The day has finally come to put the panel together!  I'd love to be able to route all of the wires before anything is on the panel and there's a lot of space to work with, but it's too hard to know exactly where everything will set.  I screwed the panel to the longeron brackets and then screwed the audio panel tray in (screws/nuts). The tray for the backup battery and comm transceiver went in next.

The brackets for the audio panel tray stiffen the panel up quite a bit, but the top still has a considerable amount of give when pushing on it, like you would do when pushing the buttons on the radio or autopilot panels.  Since the panel is already painted, the last thing I want to do is put more screws through it, so I decided to make some stiffeners out of aluminum angle attached to the subpanel and the top reinforcement of the panel.  I countersunk the screws on top - the glare shield sits down flush on it when the canopy is closed.  This really stiffened up the panel, and as an added bonus it gives me more places to attach wires routed from one side of the panel to the other.  I had to cut off a big chunk of the angle of the stiffener on the right side to allow clearance for the Dsub that goes into the autopilot panel.  Not ideal, but also not really a big deal given that this isn't a part that will really handle much stress at all.

Then the fun part came that I've been waiting months for!  I put everything in place and locked it down with black button head screws.  I have to admit, it looks pretty cool!  It's especially cool to know that I did it all myself, whereas most people these days just pay a shop to make the panel, from cutting to labeling and wiring.  The only thing of note that I'm not sure about is the fit of the audio panel.  It was very tight going into the tray and is canted just a tiny bit, meaning I got the tray angled down just a little bit.  I know it can go in perfectly straight since it was when I first fit the panel.  I may pull the panel out tomorrow and loosen up all of the screws of the tray to see if I can tilt it back to level.

The remaining things to do are to install the headphone jacks (wired into the audio panel by AFS, so I couldn't put them in before installing the audio panel tray) and 3D print a cover to go where the future IFR GPS will go.  I also need to route the cabin heat control from the heater box up to the panel.

Once everything is in place, the space fills up pretty quick!  Routing wires is going to be interesting.

Misc Wiring & Exhaust

Doing some catch up documentation from the last few days of 10 minute here and there build sessions.

I can't remember if I mentioned before, but my 3D printed fan guards turned out very nice, but when I screwed them down onto the glare shield, they curled up around the screws because I had made them so thin.  3D printed guards are probably not a great solution since I want them to be low profile.  I think I'll probably end up getting them cut out of aluminum instead.  I'm just waiting to see if I have anything else I need cut before sending the design to sendcutsend.com.  I went ahead and spliced the fan wiring together so I could just run one pair of wires to the panel.  Once the panel is in place and the canopy back on, I'll figure out how to deal with the slack needed in the wires for the canopy opening/closing arc.

I hooked the fans up to my bench power to ensure they worked.  I'm glad I got the thicker fans.  They actually move a lot of air!

I had already finished up most of the switches in the panel, but there were still a couple of oddballs left to do - the Pmag and Master switches.  Neither of those go through the ACM.  The Pmag setup took a bit of head scratching.  The use of a three way switch for their OFF-ON-TEST functions is a little confusing to wire (mostly because unlike standard mags, the Pmags have a self powering function, which means they have to be able to be grounded in different ways to allow testing of both internal power and ship power scenarios).  One of the requirements of the Pmag is having either a 3A circuit breaker or fuse in the mix.  Usually a circuit breaker is used when a key switch is used for all of the runup testing, because then the interrupted power test can be done by pulling the breaker for each Pmag.  Since I'm using toggle switches, the circuit breakers ($$$) are not really all that necessary.  Pmag specifies that fuses can be used as long as they are the slow blow type, which allows for the potential initial startup surge.  I don't need access to the fuses in normal operations, so I put them on the subpanel above the ACM.

For my future self - these are the fuses I'm using:

I believe all of my panel wiring is in place at this point (but not routed nicely - that's for after I finally install the panel).  That means I can start creating a new rat's nest of wiring for all of the firewall forward connections.  I started pulling wires to/from the EMS module, as well as continuing with the main wiring of the alternator, starter, etc.  Still more to do there because part of cutting those big wires means figuring out their routing around the engine.  I'm getting there, just mulling it over before committing.

One thing I've procrastinated on is installing the tailpipes on the exhaust.  I was waiting on that because I needed to have an idea of where the lower cowl exhaust exit opening would set to ensure I had good clearance around the pipes.  The pipes have an angle in them, so spinning them in place means there's a huge range of where the ends can exit the engine compartment and what the clearance to the fuselage and cowl is.

After a lot of iterations, I got both tail pipes exiting in a fairly symmetrical way.  Perfect symmetry isn't possible because they both have a slightly different angle where they mate up to the rest of the exhaust.  The tail pipes get connected to their feeder exhaust pipes using a clamp that has a 1/4" pin welded on the inside of it that keeps the pipes from moving out of position.  I drilled a 1/4" hole through the pipes to accept that pin and clamped the pipes down.



The right tail pipe will be where the cabin heat muff is.  I need to get this in place next because the bulky scat tubing to and from it will dictate where I can route some of my wires.  The heat muff is just a thin cylinder that goes around the tail pipe, with fresh air feeding in one side and then going out the other end to the firewall cabin heat box.



One of the complaints about this design is that in cold regions it doesn't provide quite enough cabin heat.  Some people put a second muff on the other tail pipe, but I've seen those installs and it adds a ton of scat tubing to an already cramped area.  Instead I'm going with the increased surface area approach.  A common solution to increasing surface area is to stuff steel wool into the muff around the the pipe, but the complaint with that is that it decreases the volume of air that can get through to begin with.  I saw a builder who instead added stainless rings with fingers to the pipe.  He said it has worked really well, once he found the right number of rings to add on (he had too many at first and was cooked out of the cockpit).

A while back I drew up a design and sent them to sendcutsend to cut out.  The muff sets off center around the pipe, so my rings have to account for that.  One thing I didn't account for is a change in design of the muff from when I first saw it - it now connects the two end caps using steel rods that go inside the muff.  Those would get in the way of my rings, so I'll have to chop of some of the fingers to allow the rods to pass through.

My next step is to do some tweaking of my ring material (I didn't get my measurements perfect when designing them), bend them around the exhaust and bend the fingers up.  I think I'll start with 6 of the rings.  Those fingers should pull a lot of heat off of the tail pipe and still let the air pass through pretty easily...I think.

Wiring Continues

It was a little tough to get a picture, but if you squint just right, you can see the work I did on the battery.  The first issue with the battery is the fact that it is a new lithium design (not the spontaneously combust kind, but aviation specific lithium chemistry), which means it's a different size than the old school battery.  The bar that holds the battery down doesn't fit very well with this design, so I modified the thickness of it.  I also had to buy some modified terminals that screw to the standard terminals and allow the wires to be attached from the front side vs the top (since the hold down strap covers the standard terminal location - had I know this battery was a different size, I would have changed the entire Van's design to make the hold down strap work better).  The hold down strap sits about 1/16" above the brass bar that relocates the terminal.  I don't love that.  It seems ripe for a short from a dropped screw or washer.  I think I'll pull the strap off and paint some plastidip over that brass bar.

I used 2 AWG welding wire for the positive and negative battery connections.

My wire connectors arrived this afternoon, which allowed me to finish up the switch wiring.  The last thing to do before pulling everything off the panel and painting is to figure out the length of the switch wire harness.  I put the panel back in the plane to sort that out.

I decided to make the harness really close to the exact length necessary vs adding in extra length for a service loop.  The area behind the panel is going to fill up very quickly, and for a harness this simple, I don't think the extra wire is necessary.  If there's ever an issue and I need to replace wires, the access is not a problem and swapping wires would be pretty easy.

The remaining switches either don't run through the ACM (ex: PMags and Master) or tie into other connectors on the ACM (the passenger PTT and starter switch).  All of that can be done with the panel in place.  I pulled all of the switches out, so now I'm ready to prime and paint the panel.  If the weather holds tomorrow, I might actually get that done!  My dry transfer labels are supposed to arrive early next week, but I'll give the paint a week or so to cure before I start messing with the labels.

Panel Wiring

 Before jumping into more of the panel tonight, I decided to do something I've been putting off for months and months.  When I originally put the firewall insulation and stainless foil in place, I made a template and cut it to the exact size of the firewall.  I THOUGHT that the cowl hinges that are riveted to the firewall flanges would be able to be put in place with the insulation trapped behind it.  It turns out the insulation is just a smidge too thick, so it pushes the hinge out far enough that it can't be riveted where it needs to be.  So all of that "get it the perfect size" energy was for nothing.  The only option is to cut the insulation and foil away from the firewall flange about 1/8" so the hinge can slip back against the firewall.  That seems like no big deal, but let me tell you, getting to the foil and cutting it with snips was nasty business!  I finally got it inch by inch and then ran a wood dowel over the upturned and ragged foil edge to squish it back down.  It's pretty ragged looking, but once I have the hinges riveted on with the top skin, a bead of fire barrier will cover the exposed edges and clean it up a bit.

Back to the panel.  I finished up the shelf for the com transceiver and backup battery.  I needed to see how far back the MFD display would project behind the panel to know where to locate the transceiver, so I put the MFD in place.  It's starting to look like a real panel!

The MFD is fairly shallow, and also has a few inches on the bottom side, so I was able to set the transceiver relatively close to the panel underneath the bottom edge of the MFD.  I needed to sneak it forward like that so I could also fit the battery on the shelf.  The pictures are a little deceiving.  It looks like the battery could have been slid further outboard, but if I did that it would trap the transponder on the subpanel in place.  One thing I did do with the battery was flip it end on end so the cables come into it from the outboard side.  It would have been nice to have them come in from the inboard side to make the run to the ACM shorter, but that would put them uncomfortably close to the center stack and make it all fit tighter than it needs to.

I connected the cables to the MFD just to verify I had clearance for everything.  There's obviously a lot of cable management that's going to have to happen once everything is in place.

I'm at the point where I can start wiring the switches on the panel.  I'm going to wire the switches in place, then remove them all and paint the panel.  I figured there'd be less destruction of paint that way vs flipping the panel around on the bench after it is painted.  I loosely inserted all of the switches.  I was originally just going to have the alternate static air toggle (upper left corner) left open just like the rest.  It's very sensitive and easy to flip up though, so I think I'll put the red guard on it.  While unknowingly bumping to the cabin air source wouldn't be the end of the world, it would throw instruments off enough to be a little confusing, especially if flying IFR.

The switches are a combination of spade connectors, screws for ring terminals, and solder on (for the cheaper non-aviation specific stuff).

The ACM only has 4 ground pins in the connector for the switches.  I would need 3 or 4 times that many if I grounded everything separately.  I verified with AFS that the switches carry very little current, since they are just used to send the signal to the ACM, which then does the heavy lifting of opening and closing circuits (not all switches go through the ACM - the master relay and PMag switches work differently and carry the full loads).  That means that I can combine a bunch of the grounds together to consolidate them on just a few of the connector's ground pins.  AFS suggested daisy chaining switches together to cut down on the amount of wire needed.  All of the center stack switches can easily share a single ground this way.  I chained the center switches together, then started running the power wires as well.  Unfortunately I ran out of spade connectors and had to stop.  I had originally thought that most of the switches would need ring connectors, so I ordered a ton of those and very few of the spade connectors.  Got that backwards!  These connectors fit a lot tighter than the automotive ones I have, so I'm going to order more vs just grabbing some locally.

Panel Supports

The next step to finishing up the panel was to cut the center stack opening for the IFR GPS.  It'll be a minor miracle if all of this is accurate enough to not require some additional surgery down the road, but hopefully it'll be fairly close.  The standard cutout dimensions that Garmin provides must assume that the panel is being cut by machine, because the tolerances are miniscule. Cut just a hair off the mark and the bevel of the face plate won't cover up the cutout.  Because I'm cutting by hand, I modified the dimensions so the rack will set behind the panel face instead of inside the cutout space.  This will allow more wiggle room so the face plate will cover any less than perfect cut lines.  Now that I've experienced panel cutting, I can say for sure that it'll be worth a few bucks to have my next panel cut by CNC.

The audio panel and GPS are deep enough that the subpanel has to have a cutout as well.  I made the cut just a hair larger than the racks will require, then made an aluminum doubler to go behind the subpanel to stiffen the area up a little.

The racks need a way to attach to both the panel and the subpanel.  Because my cutouts were made by hand and not perfect, getting two pieces of angle where they needed to be so the audio panel face is square and flush was easier said than done. 

The audio panel rack has holes for two screws on each side that will attach to the angle on the panel.  There is also a single screw on each side towards the back of the rack. This rack was kind of a nightmare to fit.  I emailed the company to see if I got a bogus rack, but never heard back.  The dimensions are such that there's literally no way to set the rack against the panel and still have the audio panel itself cover up the hole. This seems to be a common theme with panels!  They essentially make the outside rack dimension the exact same dimension as the face plate of the unit.  Why they don't make the unit 1/16" bigger all the way around is beyond me.  I'll probably regret it later, but I made my cuts to ensure all of the cutouts would be hidden.  That meant making a smaller cutout and offsetting the racks behind the panel instead of thru the panel, as normal.  It makes getting everything square a lot more difficult, but worth it in the end.

Like on the panel side, I made a a couple of angle brackets for the subpanel side.  Then, measured 100x to figure out how to create angle brackets that go from the panel to subpanel.  These brackets have to set at a slight angle to capture one of the rack's front screws, then the back screw, then the subpanel bracket, so it took a bit of experimenting to get them right.  The GPS rack is larger and will also need some sort of support, but until I have the actual rack, it's just too hard to fabricate anything ahead of time.  I'll just have to suck it up and deal with the pain down the road when I add the GPS.

I pulled out the center stack to give myself a little more room to work on other panel related stuff.  Next up is finding a home for the com transceiver, fuse block and backup battery for the PFD/MFD.  I was initially going to put the com transceiver on the subpanel, but they've changed models since I first laid everything out.  The newer transceiver is not as deep, but it's taller and wider.  It just doesn't fit very well on the subpanel.  So instead, I'm putting the fuse block on the subpanel.  Here you can see it next to the big subpanel cutout.

Since the transceiver lost its home on the subpanel, I need to make a tray of some sort to mount it and the battery in between the panel and subpanel.  I screwed a couple of pieces of angle to the bottom of the panel/subpanel.  I've got some spare 1/16" sheet that I'll rivet to these to make a tray to hold the transceiver and battery.  The transceiver is just shallow enough that it can set on the tray without interfering with the back side of the MFD.