Firewall Forward

Well, that was quite a productive day…seems like as of late, even when I work on stuff on the weekends, I can only squeeze out a few hours. Today I managed to get in basically a full day’s work, and ended up with a nice payoff.

I’ve been sort of wracking my brain over today’s assembly for some time. I knew I needed to mount a few fuse holders, plus the E-bus relay, on the firewall, and I was pretty sure I wanted to mount these items to a panel, and then mount the panel to the firewall, rather than just direct-mounting everything. The main reason for this is to have only four new holes in the firewall instead of ten. A side benefit is that if stuff needs to be modified down the road, it’ll be possible to modify or replace the panel, instead of turning the firewall inset Swiss cheese.

About a week ago I made a cardboard mockup just to test the layout of everything, and ensure that it’d all fit where I wanted it. Today I actually got to work on building the panel. It’s fairly straightforward – just a rectangular piece of sheet, to which I added 1/4” flanged top and bottom for some stiffness. It mounts to the firewall at the four corners; the top two screws go through one off the angles, while the bottom two are just through the stainless.

Actually building this was more involved – I worked out an order of operations for things to go relatively smoothly, cutting the panel piece, laying out the flange bends, then the attach screw holes, then transferring those holes to the firewall. That part required some creativity, because things were tight in a few places thanks to the engine mount tubes.

I didn’t bend the flanges until I’d laid out the attach holes for all the components, and drilled the rivet holes for the nutplates. Only then did I bend the flanges in my little sheetmetal brake, followed by riveting a bunch of nutplates, mounting the parts, and doing the most fun part – building out the various wires. I did what I could with the panel on the bench, before mounting it in the airplane and working on integrating it with the battery/contactors/etc.

Here’s the final product mounted on the firewall. From left to right, we have:

• An ANL fuse holder for the main bus feed. This will protect all the circuits downstream of the battery and alternators.
• The battery bus – just a small fuse block with a few always-hot circuits
• The battery bus then feeds a relay, which can be activated by a switch in the cockpit. This provides a way to power essential equipment in the event of a master contractor failure
• Another ANL fuse holder for the alternate feed

Obviously we’re still missing some things here, like the alternator feeds, and the cables from the fuse holders feeding across to the firewall penetration and into the cabin. This long runs will come later.

Here’s a slightly wider shot, showing how the panel sits above the contactors:

Finally, a closer look at the contactors, where I also fabricate a few of these cable/wire runs today. Visible here is the cable from the battery to the master contactor, a smaller tap from the hot side of the contractor to the battery bus, the big jumper from the master to starter contractor, and the tap from the hot side of the starter contractor up to the power distribution panel:

Of course, as I type this, I realize I made at least one mistake. I have a Hall sensor that I want run the battery bus tap through, so I can see how much current it’s drawing at any time. This will work in concert with a shunt on the main bus feed; the two together will give me a full picture of current flow.

Problem is…those terminals on the battery bus feed are too big go go through the Hall sensor. I needed to feed that cable through the sensor before terminating both ends. So I guess I’ll be prefabricating that cable at some point. Maybe tomorrow, assuming I have enough of the big terminals…

So, the first order of business after the last work session was to fix that battery bus feed cable. It occurred to me that I didn’t need to completely re-make the cable; there was enough slack that just clipping off one terminal and crimping on a new one was perfectly fine. I did check for the proper orientation of the sensor about 47 times before I crimped on the new terminal, though.

That also led me to think about connectors; my intent has been to put in a three-pin Deutsch connector for the hall sensor, but if I do that I think I want to mount the connector to an engine mount tube, instead of having it hang around in midair. So I ended up ordering some little pads for that purpose, which should be here tomorrow.

From there, it was on to the sort of vague task of how to best organize the wiring runs, and how to actually work things into nice bundles. I decided a good way to make practical progress here was to start picking off the various wire groups, running them to somewhere near their sensor, and let that sort of flesh out the layout. To do that properly, I decided it was time to cut the hole in the right rear baffle where the coil pack and fuel flow sensor wires will go through.

After some thought, I decided the best way to do this was to drill two adjacent 3/4” holes and then open that up to a nice oval shape. That cutout is large enough to allow the coil pack plug to pass through, but will still be covered up by the grommet I bought. Then I laid out the screw holes for the grommet, drilled them, and installed nutplates. I got a picture of the hole itself, but I don’t have the right screws to actually install the grommet…plus that would have been purely for the photo op.

Next, I needed to figure out where to mount the manifold pressure sensor, along with the vacuum block that came with the CPI2 ignition. I ended up mounting those sort of back-to-back, since they’re connected together with a vacuum hose. I didn’t get a good photo of this, but if you look carefully in the following photo, you might see the gold-anodized vacuum block hiding back there.

With those items mounted, I could finally pull wires around. That still didn’t give me a great look at what the routing would look like, so I decided to use some temporary zip-ties to better mock up the harness run, along with where stuff would branch off. The result still doesn’t look super organized, but hey – it’s still a work in progress:

Of course, about the time I was taking this photo, I realized I forgot to include the CHT and EGT sensor wires in the bundles. Those should be fairly straightforward, though.

You might also notice the two stray wires at the bottom of the photo. These are the 5V and ground for the various sensors; they’ll need to serve three and four sensors, respectively, so I’ve got some window-splicing work to do. The general idea will be to pull them along the temporary bundles I have in place, figure out where to put the splices, and then I should be in a pretty good place to maybe start lacing some things up here and doing some terminations.

Electrical stuff is fun. Really.

So as I’d hoped, I was able to start getting into some of the satisfying wiring stuff – routing wires to their final locations, trimming and terminating them, and lacing up bundles. It wasn’t all fun and games, though – as seems to be inevitable at this point, I had instances where I needed materials, or discovered a misstep from earlier. Nothing awful, just minor annoyances.

The first question was a classic one: where to start. It seemed to make sense to work from the firewall out, picking off branches as I got to them and working on the branches. Accordingly, the first thing I came to were the wires for the battery bus and contactors. On Tuesday I was able to get the four battery bus circuits taken care of, along with the switched ground wire to the E-bus relay. That was when I hit a misstep – I need a +12V source for the other side of the relay coil. Looking at my schematics, my intent was just to jump a small wire from the hot relay input. It’s an easy enough fix – cut the terminal and crimp on a new one, with a 22AWG wire tucked in alongside the 10AWG supply – but I don’t have the right terminal to redo that relay connection. First item on the order list right there.

OK, well at least I can pull the switched wires for the contactors while I’m in the neighborhood, right? They’re close enough that they might as well just get bundled along with some of the battery bus stuff. Except…I don’t have the right ring terminals for this. The ones I have are too small for the contractor studs. Yet another item on the shopping list.

But those were easy to tuck out of the way so I could move on. Next step before starting any more lacing was to splice in the +5V and ground wires for the hall sensor – which will also get wired alongside the other battery stuff. Getting those window splices done was a little fun, since I was working really close to the firewall fitting. Once they were done, I went ahead and laced the battery bus/contactor/hall sensor branch, then started working my way down the line.

The next junction was for the manifold pressure sensor. I was hoping to go ahead and get the termination done here, but the connector for this sensor is odd. After some research I determined it was a Weather Pack connector, and while I could probably do OK crimping the pins with what I had on hand, I decided to go ahead and order a specific crimping tool. At least Amazon had that for delivery tomorrow, is it skipped right past the running shopping list.

To make a long story short, I ended up adding the +5v and ground branches for both the MAP sensor and the fuel flow (well, fuel flow only gets ground, it runs off +12V). Those are all the branches I need on the right side of the engine, so I was able to lace things all the way up to where two branches split off – one for the fuel flow sensor and top mount coil pack, and another for the main ammeter shunt.

Before I really go any further, I want to get the CHT and EGT sensors installed so I can properly handle the wiring routing for those. So I put a top to lacing for the night, but did decide to go ahead and secure what I had laced to the engine mount tubes. A couple adel clamp pairs took care of this, and as of now the harness is looking pretty nice for the first couple feet:

Tomorrow I figure I’ll see about getting the MAP termination done, and maybe get those CHT/EGT probes mounted. The only thing that might delay me here are the various crimp terminals I need to order – I doubt I’ll get those this week, so it might be next week before I can sit back and feel like the FWF wiring is largely done.

That reminds me, I also want to order some standoff pieces from Van’s for routing wires along the sides of the engine forward – like the alternator/starter cables. The shopping list grows again…

Yeah, I’m getting behind on build log posts yet again. Haven’t had any real extended work sessions, but I’ve been chipping away.

Last week, after getting my new crimpers, I got the connector done for the MAP sensor – which gave me the opportunity to fire up the panel and see an actual functioning engine readout. It’s not much, but it’s something!

Beyond that, I’ve been a little hung up on some parts on order, and I kept forgetting to place those orders, which didn’t help. But now  I have some terminals coming from Stein, plus an order with Van’s for the standoff pieces I’ll use for routing wiring near the sump. Not sure how long it might be before I get the Van’s order, though – I’ve heard they’re running behind on shipments, and they had messaging on their site to that effect as well.

I did get the CHT sensors installed; these are pretty easy. There are threaded holes in the cylinders to accept fittings, to which the actual sensors mount bayonet-style. I had to go in and clean up these holes, thanks to mud dauber activity…which in turn required removing the exhaust, as well as the #2 intake tube.

Then there was the matter of the EGT sensors. To mount these, I need to drill a hole in each exhaust pipe. Which means I have to decide where to drill that hole. After a fair bit of research, I ended up going with a position about 2” from the exhaust flange. Then I had to ensure that the probes – which project out a few inches – wouldn’t interfere with the spark plugs or the lower cowl. So I got to hang the cowl again briefly just to check clearances, which turned out to be fine.

So finally after a lot of thinking I took the drill to my expensive exhaust. In the end it was undramatic, but it’s always fun going with these sort-of irreversible actions. So now I have the EGT sensors mounted all around:

Finally, I did some other prep work for that upcoming sump routing. I decided it would be a good idea to put the prop governor oil line back in place, since it’s a potential conflict for the wires I’ll be running. When I removed that before, it was a real pain to snake it out around everything. I was smarter this time, and just went ahead and removed the exhaust and intake pipes on that side of the engine, which made things go a lot better. I’ll have to remove the line again at some point when I’m finalizing the baffles, but such is life.

Now I’ve just got to figure out what I can work on while waiting for those standoffs to come in. I suppose one thing to be thinking about is the lower spark plug wires – those will have relatively long runs from the mag pad on the back of the engine. Plus there are some other terminations I can still work on in the meantime.

Not a ton to report right now. Still waiting on a couple parts orders so I can deal with routing wires by the crankcase, along with spark plug wires. I did decide to go ahead and knock out some EGT stuff.

The EGT probes come with armored cables, terminated with tab connectors. I’ve heard from a couple people that they’ve had problems with these terminals, so I followed someone else’s lead and decided to replace them with Deutsch connectors all around. This was my first time using the Deutsch kit I bought, so there was a bit of a learning curve, but in the end it went pretty quickly.

The original tab connectors (BOO):

And the Deutsch connectors (much better):

Yeah, that’s it for now.

So…I’d originally planned on being at Oshkosh this week, but those plans got changed. I could have just cancelled my vacation time as well, but instead I decided to make it an airplane-building staycation.

I’ve been all over the place as to hat to work on next…previously I was looking at routing the EGT/CHT leads, but decided I wanted to see the spark plug wire routing first…and then I wasn’t sure about making the plug wires until some other stuff was done…finally I settled on making the starter supply cable and the ground cables.

So that’s what I’ve been working on. These are fun tasks where the actual cable fabrication is pretty simple, but there’s lots of thinking and trying to figure out the best way to route things. For example, the starter cable pretty clearly needs to route behind the engine, then above the intake runners. The fun part is figuring exactly how to route it across the back of the engine.

In the end it just becomes a trial-and-error exercise; crimping on terminal (the starter end n this case), securing it in place, then wrestling the cable around and just figuring stuff out. I also decided to firesleevea this cable given its proximity to the exhaust pipes, especially right at the cylinders. That added another fabrication dimension, but at least it looks neat:

Then there were the ground cables; here I decided to go with two cables. First there’s a heavy 2AWG cable from the negative battery terminal to the right side of the engine case, which is big and fat since it’ll be handling the starter current. Then there’s a smaller 6AWH cable from the left side of the case to the firewall ground bus; this one is smaller since it’ll just be handling all the other loads in the airplane.

For the engine case connections, there are two bosses near the lower engine mounts that work well for this purpose. The real fun was figuring out how to get in there with a wire brush and remove the paint so I could get a good ground bond. It’s also necessary to allow for the engine moving relative to the mount here; a bit of an anti-strain loop is necessary, as seen here on the right side:

Then the cable runs down the lower mount tube to catch the battery terminal:

Things are a little tighter on the left side, and I had to bend the terminal just to get some reasonable semblance of an anti-strain loop. I also decided to add some snakeskin wrap to the cable there just for some additional anti-chafe protection. It’s not obvious in this picture since it’s kind of dark, but the heat shrink near the lower adel clamp is there to secure one end of the wrap:

Then, as with the other cable, it follows the mount tube to the ground bus lug:

I’m thinking that tomorrow I may go ahead and fabricate the last two heavy cables, which will run between the alternators and the power distribution stuff.

That said, the more of this I do, the more I think that I should maybe switch back to finishing up the engine baffles. A lot of the questions around where to route stuff have involved baffle-related stuff that I can’t quite be sure about. I’m also concerned that the more wiring stuff I put in place, the harder it’ll be to remove baffle parts, which I’ll need to finish that up. If nothing else, I should at least figure out the oil cooler mount…

So waaay back when, I had a post or two about fabricating a guard for the SDS crank trigger wiring. The purpose of this is to prevent a broken alternator belt from knocking out the wires to the sensor and thus making the big fan out front stop. Later on, when I was working on the baffles, I had to significantly modify that guard, and then it still wasn’t enough once I bent the intake ramp up to the proper angle.

Today started out with planning cable runs under the right side of the engine, but it became clear eventually that I’d be wanting to figure out the sensor wiring run at the same time, which meant finalizing some stuff, including removing more material from that intake ramp and fabricating a new cable guard.

So I went to work. The previous guard was made of aluminum angle, which was kind of a pain to work with. This time, I decided to use some scrap from my first failed panel cut, specifically the 90° bend at the base of the panel. This is still 0.063” alclad, so it’s pretty beefy, and I think perfectly workable for this purpose.

I don’t really have any progress pictures, but suffice it to say there were a couple false starts before I got something I was sort of happy with. Though now that I look at it, I’m second-guessing myself again, and thinking that the opening in the guard to accommodate the wires is way too big. So…it’s entirely possible that I bin this one and start over yet again tomorrow. We’ll see how I feel about it in the morning.

For now, though, here’s the new thing in place alongside the newly-trimmed intake ramp:

Another multi-day entry here; been kind of tinkering with this stuff intermittently over the past few days. On Friday, rather than going back to messing with the crank sensor cable guard again, I decided to get back to my primary goal, which was to figure out cable/wire routing under the right side of the engine.

First up was the big item, namely the primary alternator feed. The big item of concern here is that the fuel line to the injector servo runs in this same area, and by necessity it’s going to exist in close quarters with the alternator feed. Since the fuel line is metal braid, and will have a path to ground, the concern here is the possibility of that alternator feed ever being able to ground fault to the fuel line and become a nasty fire hazard. This is one of those items where I feel like the word “overthinking” doesn’t really apply – it’s worth being really careful.

To get better access to tinker with this, I removed the exhaust and intake pipes. Doing trial-and-error fitting while working around those was going to be no fun at all. After a good bit of that trial-and-error, I came up with a routing that I was happy with. At the point where the cable and line are closest together, I used a pair of Adel clamps to ensure they’d keep their distance. Further back, I anchored the cable to a sump bolt. Past that point, it goes away from the fuel line, but then it has to pass close to one of the engine grounds I added, so that spot has a dedicated standoff as well.

As with the starter cable, I elected to wrap this cable in firesleevea since it’ll be sitting close to the exhaust. After the requisite fitting and experimenting, I got all the fabrication done and routed the completed cable. Here we can see how it’s secured and stood off from the fuel line under the cylinders:

Further back, it crosses the ground cable, then route goes up the engine mount ring, and has one more standoff to the engine wiring bundle before making it to the junction with the battery and standby alternator feeds:

Then came the next moment of truth: I reinstalled the intake pipes to see how everything fit together. After looking at this, I made some routing modifications. Previously, the fuel line was secured to the #4 intake pipe, but that I didn’t like how that setup handled the proximity of the cable and line. So I ended up switching up the mounting; the cable is now secured to the intake pipe, and there’s another pair adel clamps keeping the fuel line at a distance. Kind of hard to see the latter here, but it’s there:

Then came the consideration of the other wires here. Initially I’d planned to route all four thermocouple leads inside the intake pipes, but given how things are a little crowded there, I think it’ll be smarter to route them around the outside. That just leaves the alternator field wire and the crank sensor leads to get routed alongside the alternator cable.

But before I could do much with those, it was time to confront the real fun of routing the crank sensor lead up by the intake ramp. Routing under the cylinders isn’t bad, but figuring out how to handle the lead close to the sensor itself is giving me fits. The leads are pretty stiff, and the angle at which they exist the sensor really want to have them rubbing on the edge of the ramp, which is no good at all. And that’s before I even get into how it should route relative to the prop oil line and the alternator bracket, all in the same area:

So far I’ve experimented a bit with an adel clamp or cable guide on the back of the cable guard, but neither of those made me particularly happy. After looking at that last photo, I’m considering maybe having an adel clamp or two under the ramp for routing, but that’s going to require more experimentation…which will be a job for another day.