What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Talk to me about your SDS CPI2 install (especially in an -8)

N546RV

Well Known Member
OK, so I've got a pile of avionics on hand, and I'm finalizing all my little black box layouts and getting shelves fabricated, all preparatory to starting to build harnesses with a vengeance. The one remaining box I don't have on hand is the CPI2 controller, so I'm looking to get that ordered.

Question #1: What did you go with for your coil locations, and how do you feel about your choice? Currently I'm leaning towards one magneto and one case mount. Any potential packaging issues with this setup I should be aware of?

Question #2: What did your harness lengths end up looking like? Ross suggested 7.5' for the crank sensor harness and 6.5' for the coil packs. Based on my routing in the cabin I expect about a 4.5' run to the firewall penetration on the lower left side. Do 3' and 2' sound reasonable for the run from the penetration to the FWF items?

Question #infinity: Any other things you ran into during your install you wish you'd thought of ahead of time?

Thanks!
 
Mounted one coil on the engine backbone near the back baffle and one on the right side of the engine cage. The latter "mount system" was comprised of 3 Adel clamps.
 
I’ve got one coil pack mounted in the right mag position running the bottom plugs and the other mounted on top of the engine running the top plugs. A bit of extra harness length is better than coming up an inch short. I’d add a foot to whatever you measure. I’m a couple days work from my first engine start with the CPI-2. I’ll post a photo link when I have everything tidied up.
 
Hey Philip,

My engine has a dual EM-5 setup and the coils are on the mag pads, IO-360-M1B. The kit had all the pre-made harnesses from Ross when I bought it. Any reason the CPI2 can’t go on the mag pads too?
 
My preferred locations are one on top of the engine, one on the firewall for the installations I've done. Reasoning is to preserve maximum accessibility to the rear engine area for better maintenance accessibility.
Plan for the future...
 
Hey Philip,

My engine has a dual EM-5 setup and the coils are on the mag pads, IO-360-M1B. The kit had all the pre-made harnesses from Ross when I bought it. Any reason the CPI2 can’t go on the mag pads too?

My first inclination was to just go with mag pad mounts for both coils. Seeing some installs with one coil atop the case got me thinking about that. For one thing, the top plugs now only require a single small bundle through the baffles instead of four plug wires. For another, it would free up some space between the engine and firewall.
 
Wire terminals and wire length

I ask Ross if replacing the terminals on the CPU end of wires (shorten)was possible to get a " tidy" fit on the harness. He suggested the terminals were special enough that you probably do not have an acceptable crimper. So, I conclude one option is splicing the wires after assembly to tidy up the harness.
 
I suggest leaving some excess wiring between components & bundling them up neatly. Later servicing is always easier when you have a bit of slack in the harness, to maneuver the connectors without stressing the crimps. I think cutting & splicing is not a good idea, it compounds the potential for future points of failure.
 
Service loop

I’m with Ralph....I have about 12” of extra harness on my CPI2, and while I do have correct crimper, doing it upside down with the fuel selector in between my two herniated discs had me placing a nice service loop in place, specially since it gets a little busy with the backup battery, if installed. Call me lazy if you like, but I’d do it the same way again. I can’t say enough about their support, btw, and about 30 hours in, it’s flawless.

On coil mounting, I went with one on the engine(H2, one mag hole) and the other in the firewall recess usually used for prop gov (H2, n/a but I stalked anyway. It puts the coils close so routing is easy.
 
Last edited:
Also a good idea to keep a drip loop on stuff under the panel. Last thing you need is a high speed expressway for water to enter the magic boxes.
 
Resurrecting this one for a sanity check, since it's time to place this order. I think I'm settled on the coil locations - one on a mag pad, the other on a top case mount - but the harness lengths are concerning me.

As mentioned before, Ross suggested 7.5' to the crank sensor and 6.5' to the coils. I measure about 4.5' from the ECU location (behind the panel) to the firewall penetration (lower left side).

The big question mark for me is wiring run distances forward of the firewall. In an attempt to estimate require wiring lengths here, I took the overall aircraft length dimension from the plans and subtracted the measured length of the canoe + rudder and the spinner dimension from Whirl Wind; this left me with a ~3' estimate of the cowl length, which I figure works as a starting point for estimating the harness length to the crank. Meanwhile, figuring the run to the back of the engine (for coils) is pretty easy, since the plans call out 13" as the engine mount length.

With that in mind, it seems I probably need something more like 9' and 8' for the harnesses - that leaves 4.5' and 3.5' FWF to route to the crank sensor and coil packs, respectively.

Can anyone comment on the reasonableness of this hand-waving? It's still a bit strange to me that my estimates are significantly more than Ross's suggestions, but maybe most folks are mounting the box closer to the firewall or something?
 
Lengths can depend a lot on routing. For instance if you run the Hall cable down the engine spine and then over to one side to penetrate the firewall, that might add 18 inches.

Then if the ECU is mounted on the opposite side, those wires might need to be another 18 inches longer still.

It's best to plan the ECU location and firewall penetration, take a piece of wire and run it along your planned path. Mark and measure that and perhaps add another 6-12 inches.

We usually only terminate one end so you can get wires through the firewall with the smallest hole and be able to tailor your lengths.

We'll happily make your harnesses to any reasonable length you want.

You can coil and cut but can't stretch or add easily if too short.
 
N546RV
You mentioned your planned firewall penetration would be lower left. It does seem a long wiring run to the Crank Sensor & coils, zig zagging up, down, up & forward. I do hope you are not planning to run the crank sensor wiring below the cylinders (near the starter or alternator cables). Better to run that sensor wiring back above the engine away from any potential electrical interferences.

The Rockets I've installed systems in, although not having the front luggage areas as your plane has, had the firewall penetrations near the top corners or the engine mounts, wiring runs were fairly short because they would follow the top tubes of the mounts to the components or baffling with the crank sensor wiring following the top case joint line (secured with adel clamps) to the crank sensor, again not that long.

The RV SBS models I've installed in had similar wiring runs, all worked out well.

In your case, I'd suggest mounting the ECUs in the well areas below the luggage compartment, either left or right for easy servicing access, resulting in shorter wiring runs to the coils & crank sensor. Again, a bit of extra wire length with excess bundled up is a good thing.
 
Last edited:
Here is where I mounted my SDS equipment. All the cables I had from Ross fit but I had to run them a few times to get them right. Another foot would have made it much easier. I had to run the crank sensor underneath but on the opposite side from the alternator. All my sensor wires go through the FW on the lower left and all my power wires go through on the right about halfway up the firewall.

You need to check the baffle plans before you mount the coil on top. If you have the fuel block on top then there isn’t a lot of room left over, unless the top coil is mount off the centerline. I’m not sure how it mounts on the 4 cylinder engines.

I have two GRT sports and the fastback option which makes getting to the avionics a little easier.



 
N546RV
You mentioned your planned firewall penetration would be lower left. It does seem a long wiring run to the Crank Sensor & coils, zig zagging up, down, up & forward. I do hope you are not planning to run the crank sensor wiring below the cylinders (near the starter or alternator cables). Better to run that sensor wiring back above the engine away from any potential electrical interferences.

The Rockets I've installed systems in, although not having the front luggage areas as your plane has, had the firewall penetrations near the top corners or the engine mounts, wiring runs were fairly short because they would follow the top tubes of the mounts to the components or baffling with the crank sensor wiring following the top case joint line (secured with adel clamps) to the crank sensor, again not that long.

The RV SBS models I've installed in had similar wiring runs, all worked out well.

In your case, I'd suggest mounting the ECUs in the well areas below the luggage compartment, either left or right for easy servicing access, resulting in shorter wiring runs to the coils & crank sensor. Again, a bit of extra wire length with excess bundled up is a good thing.

Good food for thought. You've kind of driven to one of the challenges of this process - namely, that FWF routing is something of a mystery to me right now. Without an engine or even mount on hand, I'm having to use a lot of imagination here, and that's made more fun when considering the complexity here. I probably need to do some in-depth studying of FWF stuff, especially the locations of firewall penetrations and mounted components.

Anyway, to more practical discussions..."lower left" is probably not the most accurate description of the currently-intended wiring penetration. One of the design goals here has been to avoid exposed wiring runs in the forward baggage compartment, both for aesthetic and practical reasons (ie potential damage). So the current plan has been to route sensor wiring (EMS stuff, CPI harness, etc) along the left side of the fuselage, and just below the forward baggage shelf.

It occurs to me, though, that I could also route this wiring just inside the deep well in the compartment, which would save a good 3' of harness length. The only real downside I see is that the bundle would need to be secured to a removable part of the aircraft - either the horizontal baggage shelf or the "wall" of the deep well. I imagine that at least the well is rarely removed, so perhaps this is a worthwhile sacrifice.

As for the ECU location, I think my plan of putting it behind the panel is a solid choice for serviceability. Placing it beneath the compartment would require removing the shelf to access it, or else crawling on the floor. With it behind the panel, I'll be able to get to it just by removing the panel center section (or maybe even just an EFIS screen).

This sure is the fun part of the build - those days when all I had to do was follow directions and rivet stuff together were a cakewalk compared to the endless interrelated decisions I'm trying to weigh out these days. But hey, that's where the "education" part of E-AB comes in, and I'm sure it'll contribute to a nice feeling of satisfaction when I'm cruising around in this thing next year.

:)
 
This sure is the fun part of the build - those days when all I had to do was follow directions and rivet stuff together were a cakewalk compared to the endless interrelated decisions I'm trying to weigh out these days. But hey, that's where the "education" part of E-AB comes in, and I'm sure it'll contribute to a nice feeling of satisfaction when I'm cruising around in this thing next year.

:)

And that's why you see so many completed airframes for sale without engine/avionics. Some folks enjoy the hand-held step by step building part, but have neither the skill nor the curiosity to finish it out. Nothing wrong with that, I suppose, but the FWF and avionics were easily the most fun part of the build for me.
 
I would always try to mount the ECU in a reasonably accessible place as we've had more than a few customers mount them in places you need to be a contortionist to remove once the aircraft is all completed. If your wiring needs to be 9 feet long to reach there, that's no problem.

Many of us learned after the first build to make things more accessible the hard way. As you age, you become less flexible and generally lose your close focus ability. Nothing worse than being upside down, under the panel with your feet up on the seat trying to see something 6 inches away that you can't get your fingers on very easily.

Be sure not to route the Hall sensor cables closer than 1 inch to the plug wires or parallel and close to any other wires with high current/ voltage or pulsing current. Ground the ECUs to separate terminals and don't stack on top of any other grounds.
 
Many of us learned after the first build to make things more accessible the hard way. As you age, you become less flexible and generally lose your close focus ability. Nothing worse than being upside down, under the panel with your feet up on the seat trying to see something 6 inches away that you can't get your fingers on very easily.

Yeh, he’s actually describing me, but forgot to mention the addition of more than a couple of pounds :eek: In more than a few of my posts I have suggested that folks should design their installations for future ease of servicing! You WILL thank yourself later.
 
And that's why you see so many completed airframes for sale without engine/avionics. Some folks enjoy the hand-held step by step building part, but have neither the skill nor the curiosity to finish it out. Nothing wrong with that, I suppose, but the FWF and avionics were easily the most fun part of the build for me.

It's challenging, but in a fun way. I have to engage my brain a whole lot more and think 439 steps ahead, but there's the motivational aspect because I'm doing the stuff that's turning an aluminum canoe into a machine.

Turning these kinds of abstract problems over and over in my head until solutions fall out is the exact thing that I enjoy the most about my day job. The difference is that when I make a misstep in software it's a whole lot easier and cheaper to fix than if I order the wrong part. It doesn't change the fun problem-solving aspect, but does make me care more about trying to think through everything.

I appreciate all the input, sanity checks are good to help me pick out the point in time when I can stop thinking and start doing.
 
Yup - heaven forbid it ever happens, but if I ever take a birdstrike and bust the windshield and I'm forced to tear into the front structure anyway, I'm going to unrivet the top forward skin and rape out my entire panel-forward area and reinstall it the way I want it to be now. :D
 
Back
Top