My CPI installation

After an inflight ignition failure with my Laser system, I found myself forced to make a choice--------fix whatever issue had cropped up with the Lasar, go to straight mags, or go to all electronic ignition.

The Lasar is a great concept IMHO, but I am not happy that the sustem was never tested for LOP opperations, and in fact the Slick rep told me not to attempt LOP with it. Plus, support for Lasar is kinda slim. OK, Lasar off the list.

Straight mags-------well, no thanks.

So, that left all electronic. My first choice was Emag/Pmag--------and has been since I first saw the unit at OSH 2011. but------it seems to still be vaporware, and not hardware. Also, the price on their website was on the high side compared to other options.

Due to the success of Mike Robinson and Adam Pontious installations in this thread http://www.vansairforce.com/community/showthread.php?t=132220 and the similar success of one of my neighbors, I decided to go with the SDS/Racetech CPI system. Plus, SDS/Racetech has a really long track record of successful systems in the automotive world.

After making the decision to go CPI, and a phone call to get all the specifics ironed out for the system I needed, I placed the order for a dual system.

The installation of this system in my RV 10 is provided for your entertainment

After removing the control box for the Lasar, I mounted the coil pack for the lower plugs on the firewall ----luckily one of the nutplates from the Lasar install was perfect to use for installing the CPI unit, just needed to drill a couple more mounting holes to get it in place. The location spanned one of the firewall stiffeners so that is a good bit of luck. Also, clears the cowl fastening screw---more good luck.



The coil pack for the top plugs was mounted in a similar way to what Mike R. did on his rocket. I had to make up a custom mount that would span over the fuel injection line, and allow the unit to clear the injection spider too. Ross makes a nice CAD/CAM billet part to do the same thing, but he was out of stock on this when I made my order. Here is a shot of my mount, you can see the fuel injection line going under the mount.



The other side, showing the support for the injector line.



The igniter module was mounted to the baffling with a piece of 1/8" plate to give it a solid mount, and thermal grease was use to make sure of good heat transfer. Also, a stiffener was added to behind the baffle to keep the thin baffling material flat.

Mounting the pickup time.


The flywheel needs to be drilled and tapped for the magnets that trigger the Hall effect sensors, very straight forward and well documented in the instructions, so I did not include any photos here. I did verify the final positioning of the magnets to the pickup by measuring from the flywheel mounting surface on the crank flange, back to the pickup sensors. The actual sensor is pretty small, so I wanted to be sure I had the alignment correct.



The magnets are 3/4" back from the aft edge of the flywheel pulley, so that amount was where the pickup sensors should be on the scale---------right on .



The pickup mount is kinda wide, and to achieve the close magnet spacing called for, I decided to trim off the corners of the mount like Adam did in post 19 of this thread.. NEW datum point--------see post 10 below-------no need to file off the corners.

I also made up a wire guard just in case a belt let go-------mine is made from some scrap 1/8" x 1" x 1" angle. Couple shots of it front and back.





After getting it shaped as needed, I buffed out the saw marks, and mounted it on the engine. I added a stainless tube to protect the run of pickup wiring forward of the baffle.

One thing to be aware of is the spacing between the two case bolts. Measure it before ordering your parts. Do not make the mistake of believing all four cyl engines use the smaller spacing (3 1/4" as I recall) and that all 6 cylinder engines use the larger (3 1/2") spacing. This engine has the tight spacing, and my neighbors 320 has the wide spacing. Go figure.....



The wiring was pre cut to apx length, and the connector pins already installed, so that was a nice bonus. I added snakeskin covering to protect the coil pack wires, and then color coded the top sensor, and top coil pack with red heat shrink so I could keep track of what went where.

Brain box.

The CPI system is pretty simple, really only three major components and some wiring to hook them all together.

The coil packs, shown in post 1. The pickup shown in post 2.

And lastly, the brain box shown in this thread.

Here are the two brain boxes with the supplied wiring pigtails------I made up a loom using the supplied pigtails and additional necessary wires to the breakers and toggle switches. I also removed the ACS ignition switch in favor of a pair of locking toggle switches, and added a simple key switch to supply power to a push to start button, both from Steinair.



I made a mount out of some scarp aluminum sheet, bent it up to sit under the sub panel that holds the throttle/mixture/prop controls and other push/pull operated goodies. Guess you could call it a sub, sub panel?? Used a cardboard box to mock it up, got the fit I liked and started making aluminum chips. Also shown is the loom after I got it built.





And finally, test fitted into the plane. Still lots of wire hanging, but all fit just like I hoped it would.



After getting all the bits and pieces installed, you follow the instructions to set the magnet position in the brain box------this requires using a timing light while someone stands behind the spinning prop and checks the timing mark on the flywheel. My buddy Mark-----Lancair guy, but OK otherwise -----made up this cool setup for using the timing light. When we did his plane, it was a real PITA to see the factory timing mark on the flywheel with the engine running. By setting the timing to TDC and then creating a mark on the flywheel and a pointer on the govoner housing, he came up with a much easier to see way of setting the timing.



All for now, engine runs . After getting the magnet position set, you can program the timing to your hearts content. I used figures Mike Robinson had developed for his Rocket as a starting point------will fine tune after I get it back in the air.

Now time to finish up the punch list, and put everything back together.

Next post will show more of the wiring runs.

Thanks for looking.

place Holder

Thanks for the show and tell Mike. I see the difference in the front baffle design between the -10 and the Rocket which drove you to fab up that "belt guard". Clever. I also noticed you have the 540 with the narrow case bolt spacing which requires the hall sensor bracket normally used on the 4 banger.

Great write up and pictorial Mike. The quality of the components from CPI looks really good in person and you did a nice install.

Hi Mike,

Looking good.

I have arrived home from Oshkosh and will continue with my install this weekend.

Can I ask you to send me Michael's timing map ?
Please send to
apmillerservices(at)gmail.com

I liked the TDC pointer idea , I'll be using that on the weekend

Cheers,

Dustyone said:

Can I ask you to send me Michael's timing map ?
Please send to
apmillerservices(at)gmail.com

Click to expand...

I will have to check with him first-----dont see any issues-----then I would be happy to send it along. Just seems the polite thing to do.

Dustyone said:

I liked the TDC pointer idea , I'll be using that on the weekend

Cheers,

Click to expand...

Yep, sure makes it easy when you put the timing mark where you want it, instead of where the factory put it.

The higher the contrast used for the tape and TDC mark the better. We ended up adding short pieces of white tape between the blue for the second run---even easier to see. Work out a set of signals with your helper before you start, and be careful of the prop.

Ashley,

No problem for Mike to pass along my map, but be aware that there will be a difference due to your use of a single EI vs. my dual. You can expect to add about 3 degrees advance to my values for the cruise condition to get close to parity.

And it goes without saying that my map is just a start - you should fine tune for your specific circumstances.

Hall Sensor Air Gap

Thanks for posting your CPI installation experience Mike. This will be valuable for other RV10 owners.

I think the magnet installation manual may not be as clear as it could be with respect to the Hall sensor to magnet air gap. The manual states a minimum of .025 gap from the edges of the sensor to the flywheel ID and .060 to .100 air gap sensor to flywheel ID (which is the same as the magnet faces since they are flush).

This air gap dimension is measured in the center of the red sensor block. The single sensors are narrower than the twin sensor block so there is necessarily more air gap with the twin sensor when the edge gap is the same.

In any case, the magnets will trigger out to nearly .250 so there are no worries in that department.

Typically with a single sensor block, .025 edge clearance gives you about .060 air gap and on the twin sensor, you have around .090 air gap with .025 edge gap. We provide a variety of std and light washers to go under the gold sensor mount to get proper clearances.

The photo below shows a twin sensor with .025 shim stock in the corners to simulate edge clearance and a #40 drill bit stuck in the center to show the air gap. The actual air gap over the Hall sensor elements is a bit less than that because they are outside the center point. There should be no need to file off the sensor edges.

Ross, got it----looks like there was no need for me to file off the corners. Will make a note about it in my earlier post, maybe safe folks some time/worry.

Ya'know the old saying about a picture being worth a thousand words.........great deal of info in your shot.

Mike,
Great post. Thanks for taking the time to help the group. Could you share your thoughts on how you are addressing the other half of the EI equation, uninterrupted power supply.

Mike S said:

Ross, got it----looks like there was no need for me to file off the corners. Will make a note about it in my earlier post, maybe safe folks some time/worry.

Ya'know the old saying about a picture being worth a thousand words.........great deal of info in your shot.

Click to expand...

I'll put that in the magnet mounting manual and the CPI website page too then. Clarity saves wondering...

Your feedback is helpful to improve documentation. Appreciate it.

I have to admit I was pretty obsessive about that magnet gap too. Trying to satisfy both the minimum gap at the edge and still hit the desired gap in the middle gave me fits.

Turns out its not THAT critical.

Barneybc12d said:

Mike,
Great post. Thanks for taking the time to help the group. Could you share your thoughts on how you are addressing the other half of the EI equation, uninterrupted power supply.

Click to expand...

Sure------I have a dual battery, dual alternator setup in my plane. There is a big 60A alternator that is coupled with an Odyssey 925 to power the main buss.

There is a B and C SD 8 alternator coupled to an Odyssey 680 that powers the aux buss. This could also be considered the essential buss if you wish to use that term.

The ignition, radios, and EFISs are tied to the aux/essential buss. All other electrical loads are on the main buss.

The main buss is tied to the aux buss via a Schottky diode so that if the voltage of the SD 8/ 680 setup goes down, the voltage from the main buss picks up the load-----this happens seamlessly, no flicker or brown out on the EFIS screens.

I power up the aux buss first when getting ready to start the engine, wait until the EFISs are online, then power the main and start the engine. No brown out during the starters high current draw either.

rv6ejguy said:

Your feedback is helpful to improve documentation. Appreciate it.

Click to expand...

You are welcome, I appreciate you mentioning it.

We all help each other here.

I learned a lot from Mike Robinson, now I am just paying it forward to the next person using the system.

Toobuilder said:

I have to admit I was pretty obsessive about that magnet gap too. Trying to satisfy both the minimum gap at the edge and still hit the desired gap in the middle gave me fits.

Turns out its not THAT critical.

Click to expand...

I think you had the original manual which also had some other confusing information which was my fault. This is why customer feedback is much appreciated!

Just got back from the first flight post CPI installation.

I did not write down the numbers, so this is all from memory.

Climb performance has always been limited by CHT concerns------I usually climbed full throttle, 2500 rpm, full rich, and 140 MPH in the past. This yielded around 800--1000FPM, with CHT numbers between 380ish to just over 400 depending on OAT.

My new climb numbers were were 1100 FPM, 120 MPH, and CHT in the 370-380 range, and today is pretty warm air temp.

So, the climb was greater, but lower CHT-----

Note these are not max perf climb numbers, just how I normally fly.

Next, after leveling off at 7500 msl, I leaned to LOP, not sure how lean but I feel in the 50* or more range--(forgot to put the EFIS in Lean mode). 2200 RPM. My CHTs shot up over 400 on all cylinders--------426 on the hottest. OK, time to cancel this part of the "test" and get it cooled off.

All for now, need to do a bit of programing change and fine tune things.

Mike S said:

Sure------I have a dual battery, dual alternator setup in my plane. There is a big 60A alternator that is coupled with an Odyssey 925 to power the main buss.

There is a B and C SD 8 alternator coupled to an Odyssey 680 that powers the aux buss. This could also be considered the essential buss if you wish to use that term.

The ignition, radios, and EFISs are tied to the aux/essential buss. All other electrical loads are on the main buss.

The main buss is tied to the aux buss via a Schottky diode so that if the voltage of the SD 8/ 680 setup goes down, the voltage from the main buss picks up the load-----this happens seamlessly, no flicker or brown out on the EFIS screens.

I power up the aux buss first when getting ready to start the engine, wait until the EFISs are online, then power the main and start the engine. No brown out during the starters high current draw either.

Click to expand...

Great explanation. Thanks for that. Learned something.

Mike S said:

....Next, after leveling off at 7500 msl, I leaned to LOP, not sure how lean but I feel in the 50* or more range--(forgot to put the EFIS in Lean mode). 2200 RPM. My CHTs shot up over 400 on all cylinders--------426 on the hottest. OK, time to cancel this part of the "test" and get it cooled off.

All for now, need to do a bit of programing change and fine tune things.

Click to expand...

High CHT while LOP in cruise is completely unexpected behavior. My flight to Vegas last night was same altitude, and LOP, yet my hottest CHT was 368.

Do you recall the advance displayed on the CPI screen?

Give me a call if you feel the need.

Toobuilder said:

Do you recall the advance displayed on the CPI screen?

Click to expand...

No, but should have been around 34 by your map.

Not sure if you know, but I have 9.2 or slightly higher compression, and a somewhat larger FI servo with ram air.

Mike S said:

Just got back from the first flight post CPI installation.

I did not write down the numbers, so this is all from memory.

Climb performance has always been limited by CHT concerns------I usually climbed full throttle, 2500 rpm, full rich, and 140 MPH in the past. This yielded around 800--1000FPM, with CHT numbers between 380ish to just over 400 depending on OAT.

My new climb numbers were were 1100 FPM, 120 MPH, and CHT in the 370-380 range, and today is pretty warm air temp.

So, the climb was greater, but lower CHT-----

Note these are not max perf climb numbers, just how I normally fly.

Next, after leveling off at 7500 msl, I leaned to LOP, not sure how lean but I feel in the 50* or more range--(forgot to put the EFIS in Lean mode). 2200 RPM. My CHTs shot up over 400 on all cylinders--------426 on the hottest. OK, time to cancel this part of the "test" and get it cooled off.

All for now, need to do a bit of programing change and fine tune things.

Click to expand...

Mike,

I have a similar experience. I am running mid 30's advance at cruise MAP ranges. My CHT's routinely rise some when going from climb to cruise. I consider this normal or expected. ROP with 25* advance (climb configuration) leaves the peak pressure point well ahead (i.e. later in cycle) of optimum and this raises EGT and reduces CHT. This is by design to increase detonation margin and improve cooling. At cruise, I reconfigure LOP and 35* advance. This brings my peak pressure point earlier in the cycle. This increases the power output, but also decreases EGT and increases CHT.

In my case, the CHTs are under 370-380, so I am not concerned. You could try dialing back your advance and see what happens. I understood that you could do this easily in-flight with the CPI system, so worthy of some experimentation to find an advance level that will fit your current cooling capacity. Also note that as you go deeper into LOP, your temps should drop, but so will power. Best to find the advance that works for you at the LOP range that you typically operate in.

In general, you should also be cautious of CHT's that "shoot up." A rapid rise in CHT could be a symptom of mild detonation, though that would seem unlikely at that altitude and LOP.

Larry