What's new
Van's Air Force

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

I Just Converted From Pmags to a CPI-2 Ignition

mfshook62

Well Known Member
I removed both Pmags and installed an SDS CPI-2 ignition on my carburated Lycoming O235-L2C engine in my RV-9a. I know that the vast majority of PMags perform without problems, unfortunately mine didn't. Long term ignition reliability was my main concern and here are my reasons for this conversion:

1. I reached my limit checking bearing integrity at every oil change and sending both Pmags back to the factory five times or about every 100 hours to replace worn bearings and occasionally requiring the replacement of the circuit boards and internal PMag alternator. I see that Emagair now recommends that the Pmags be removed every 100 hours to check bearing integrity. I lost confidence in the robustness and airworthiness of the PMag design. Both PMags have the newer version V.4 firmware.

2. I wanted an ignition that would not lose its reference to TDC, which happened to me on many occasions and 3 times requiring emergency landings on one mag. Once the bearing or magnet integrity is lost then timing becomes erratic and in my case severely advancing the timing curve. I was too slow to do a mag check and didn't turn off the problem mag in time. This catastrophic pre-ignition quickly destroyed my engine. I decided to purchase a brand new Lycoming engine rather than rebuild the old destroyed engine. The new engine came with slick mags and I had no ignition problems with them for about 300 hours. I decided to give the Pmags another shot and re-installed them, and the bearing problems started all over again.

3. Brad could only speculate that my engine might have had an accessory case gear train resonance that was destroying the bearings. It's a brand new factory built Lycoming engine and I had the same PMag problems with my last engine. My plane has a light weight Catto composite prop which possibly may have caused my new and old engine to not smoothly spin this low rotational inertia prop and created a gear train resonance. Brad actively listened but had no answer to the premature bearing failure and was unable to solve these ongoing serious PMag problems.

4. I did not feel comfortable with the long term reliability of having the ignition timing electronics and CPU bolted directly onto a heat soaked engine and having to endure repeated temperature excursions and engine vibration. Emagair recommends keeping the PMags and engine temperatures below 200 deg. Occasionally on hot days my engine oil temperature gets up to 240 deg F and the required PMag blast tubes are useless when the engine is turned off. I feel way more comfortable having the control module and timing electronics mounted in a cooler and lower stress environment inside the cabin.

5. I felt that the canned PMag timing curve of 35-40 degrees BTDC was way too advanced for an engine that was specified at 25 degrees BTDC. I wanted the ability to easily "set and forget" my own designed advance curve while; remaining in my engines power curve, maximizing fuel efficiency, preventing detonation, and lowering CHT's.

6. I was fed up with having to cancel flying engagements with friends while I had to stay at home or in the hanger while my mags were back at the factory.

7. I wanted a more robust ignition with no moving parts that could wear out.

8. I wanted the ability to quickly and easily increase the advance curve during LOP cross country flights.

9. I wanted the ability to view the current firing angle on the controller mounted in the panel to be sure at a glance that both ignitions were firing as programmed.

10. I need to feel comfortable that I am using the most reliable products and getting the best long term performance from every vital component.

11. Excellent customer service is an "absolute must" for me. I read all the reviews on VAF and corresponded with a few SDS customers, and they all confirmed that customer service at SDS was outstanding. Fortunately, I have not called SDS since the installation 4 years ago and 600 hours of flight time.


I am a diligent builder and mechanic. It took me a total of 55 hours to completely remove the PMag setup and retrofit the CPI-2 ignition. This included drilling holes and mounting magnets in the flywheel, installing Hall-Effect sensors, coils, controller, CPU, all new wiring, strobe light engine timing, and controller programming. The firing angle for my engine is specified at 25 deg BTDC. I started out using the CPI-2 conservative RPM-MAP advance curve which at idle fires at 23 deg BTDC and smoothly advances to 26 deg BTDC as a function of MP and RPM. Additionally, by pressing the LOP button the user defined curve advances when cruising at higher altitudes. I made one minor modification to the timing table, and I am delighted that there have been absolutely no ignition problems.

Previously, I had to set the PMag canned timing "A Curve" with jumper wire installed to 10 deg ATDC rather than TDC to decrease high CHTs. My #4 cylinder was always about 30-40 deg F hotter than the rest and baffle changes only had a small effect on reducing temperature. I was always having to compensate for high CHTs by slowly climbing to cruising altitude. My first CPI-2 hour long flight at cruising speed was uneventful. On my second flight I climbed at 1100 FPM from sea level to 5000 feet and the highest #4 CHT observed was 389 deg verses the previous 415-425 deg that I was used to dealing with at slow VS climbs. I was delighted that the CHTs and the temperature range was lower.

I installed a complete Dynon HDX system last year, and I anticipated installing an electronic ignition which also requires a 12 v. BACKUP battery. So, I installed an EarthX, 6.2ah, 2.2 lb, LiFePO4 BACKUP battery which should provide a minimum of 45 minutes of reserve power for the CPI-2 ignition and essential avionics in case of an alternator, main battery, or main buss fault that requires switching OFF the MASTER or ALT. The BACKUP battery is charged and isolated from the master buss and starter circuit by 2 large 10A Schottke diodes. The BACKUP battery is normally switched ON before start to supply backup power to the avionics and remains ON during flight up to keep it fully charged and ready to supply reserve power instantly in case the master voltage is interrupted. The BACKUP battery is switched OFF at shut down to keep it in a charged state and to prevent it from back feeding current into the avionics buss.

I planned this conversion for about a year waiting for the CPI-2 to get released into production and to accumulate some flight time. I called SDS on numerous occasions to ask questions and always received a knowledgeable response. I worked in the electronics industry for many years and felt that this design would be a big improvement over the numerous problems that I had with PMags. I decided to purchase the SDS system and all my previous ongoing electronic ignition issues ended after this conversion. Ross and Barry were very supportive after the purchase answering questions and mailing me parts usually with overnight delivery. I followed the backup battery wiring diagram outlined in the Installation Manual, and I called Barry a few times with questions. Now, after 4 years and 600 hours of use the SDS ignition has been maintenance free, and I have not had to remove or service the CPU or modify the program. Here are a few photos of the installation showing the dual Hall Effect sensor triggered by magnets mounted in the flywheel. The dual ECU is mounted under the throttle quadrant where it remains at cabin temperature and is out of the way and easy to service. I bolted an aluminum block near the sensor wires as added protection from fan belt issues.

IMG-3650.jpg
IMG-3717.jpg
 
Last edited:
How many people have had a pmag loose it’s reference to TDC?

Plenty. It was one of the major operational hurdles the company had to work through. It's caused by both software and mechanical issues - the latter being one of the reasons you are required to remove and inspect the unit every CI.
 
Last edited:
I think (knock on wood) the loss of timing was an issue in early PMAGS that has been documented and corrected with the current versions of hardware and firmware. I have 250 hours on a dual PMAG setup and have been very happy with their performance to date.
 
Thanks for posting your pix and experiences here Mike. Installation looks good and easy to get to the ECU.
 
I think (knock on wood) the loss of timing was an issue in early PMAGS that has been documented and corrected with the current versions of hardware and firmware. I have 250 hours on a dual PMAG setup and have been very happy with their performance to date.

Indeed, that demon seems to have been tamed for the most part, but not eliminated. There are still very real mechanical issues that every Pmag owner needs to check for. Bearing condition and magnet position being the big ones. This requires removal of each ignition to inspect once every year, and this high level of preventative maintenance is simply not required on the CPI system. The essentially maintenance free aspect of the CPI is a product discriminator for SDS.

Pmags need to be removed from the engine and inspected on the bench once a year to ensure continued operation. In contrast, SDS is an "install and forget" system with zero moving parts. That's what the OP is getting at.
 
Do folks up-armor the 2 wires coming off the red pickup sensor to the baffles?

Fan belt failure tearing a wire loose has happened, right?
 
Do folks up-armor the 2 wires coming off the red pickup sensor to the baffles?

Fan belt failure tearing a wire loose has happened, right?

I sure do. There are provisions to attach various clamps and shields to the mount just for that purpose and most that I've seen do just that. The SDS pickup and mount themselves would withstand blows from a framing hammer. It's a simple thing to eliminate the thrown belt scenario as a viable risk.
 
Last edited:
Do folks up-armor the 2 wires coming off the red pickup sensor to the baffles?

Fan belt failure tearing a wire loose has happened, right?

I used to sell Subaru engine conversions and parts. One of the parts was a speed sensor using a hall effect sensor and a magnet wheel on a CV joint. The wire coming out of the hall effect sensor was the weekest link in the whole conversion, but replacements sensors are super cheap.

I will definitely up-armor those wires on the SDS and replace the sensors periodically.
 
There are no totally, absolutely risk-free ignitions. Let's focus on operating and performance differences.
 
How many people have had a pmag loose it?s reference to TDC?

Early on there were issues with them, which is why we developed the EICommander.
Since version 40 came out, I have not heard of a single instance of lost timing.

If you are running a ore version 40 P-mag, get it in for the update at your next condition inspection.

As for the bad bearing issue, the EICommander can and has picked those up. That said, these types of failures are very rare.
 
Annual CI and CHT question ?

What is the rationale for removing annually, it is a major PITA on an RV3. I am following a 100Hr inspect schedule on rev 40 vintage Pmags. I cannot recall where that suggestion came from.

CHT - Just for the record, the advance curve is not a problem for all engines. My 7:1 0320 never exceeds 300F winter or summer.

FWIW, I have SDS on my 0360 / RV4 project for reasons cited above.
 
What is the rationale for removing annually, it is a major PITA on an RV3. I am following a 100Hr inspect schedule on rev 40 vintage Pmags. I cannot recall where that suggestion came from.

CHT - Just for the record, the advance curve is not a problem for all engines. My 7:1 0320 never exceeds 300F winter or summer.

FWIW, I have SDS on my 0360 / RV4 project for reasons cited above.

The jumper in (A curve) timing is perfect for low compression engines, such as the one in your -3. If you had the "stock" 8.5:1 engine, there is a very good chance you would have high CHT's. (Jumper in timing starts at 26.6? and advances out from there. No jumper timing starts at over 30?.)
 
What is the rationale for removing annually, it is a major PITA on an RV3. I am following a 100Hr inspect schedule on rev 40 vintage Pmags. I cannot recall where that suggestion came from.

It isn't mandatory. The website states: "The Installation and Operating Guide has a list of items that can (emphasis added) be incorporated into your annual condition inspection.".

One of the items is: "Remove ignition and examine shaft and drive gear condition. Note: Ignition disassembly is not necessary (and if done may void your warranty). Look for excessive play (lateral and axial). Shaft rotation should be free, with no catching, flat spots, or grinding. The shaft on “P” models (with internal alternator) will have a push-pull rippling effect as the shaft turns and the permanent magnets pass the rotor poles. This is normal and expected. If a P model ignition does not have this magnetic ripple, the unit requires additional (shop) service."
 
Congratulations on your new install, hope your CHT is more manageable once you start the advance curve that is suppose to help with the performance of the engine.

There are no totally, absolutely risk-free ignitions. Let's focus on operating and performance differences.

Now, that is what I call a voice of reason and worth paying attention to.

I simply tune out some of the subjective opinions expressed here design to promote one brand or knock other brands.
 
Congratulations on your new install, hope your CHT is more manageable once you start the advance curve that is suppose to help with the performance of the engine.



Now, that is what I call a voice of reason and worth paying attention to.

I simply tune out some of the subjective opinions expressed here design to promote one brand or knock other brands.


So lets get objective.
If you go back to basics and write down requirement for the ignition system it would include something to the effect that the availability of the system should be no worse than any other cause that would likely result in loss of engine power for a certified engine type and there should be no single point failure in the ignition system that could result in loss of power. With the exception of the P mag all others require an external power source or reliable storage that can self isolate from the main electrical system to protect against a large number of single point failures in the typical electrical system . Designing the electrical system to have two independent paths and still make each path reliable, available and automatically monitored is possible but requires great attention to detail in the design and implementation. There may be advantages in some of the designs to being able to control the spark advance and energy but the big issue is independent power that will keep the engine running and developing power until the fuel is exhausted. All other discussion is marketing and brand tribal loyalty.
I hope that in changing from P mag to SDS there were also changes to the electrical system design to address these issues.

KT
 
So lets get objective.
If you go back to basics and write down requirement for the ignition system it would include something to the effect that the availability of the system should be no worse than any other cause that would likely result in loss of engine power for a certified engine type and there should be no single point failure in the ignition system that could result in loss of power. With the exception of the P mag all others require an external power source or reliable storage that can self isolate from the main electrical system to protect against a large number of single point failures in the typical electrical system . Designing the electrical system to have two independent paths and still make each path reliable, available and automatically monitored is possible but requires great attention to detail in the design and implementation. There may be advantages in some of the designs to being able to control the spark advance and energy but the big issue is independent power that will keep the engine running and developing power until the fuel is exhausted. All other discussion is marketing and brand tribal loyalty.
I hope that in changing from P mag to SDS there were also changes to the electrical system design to address these issues.

KT
Keith,
I agree in principal with your comment but if you are implying that not having a self generating system like Slick or PMAG would mean a single point of failure, than I would have to disagree. Although I have been wondering having a hall effect sensor for both ignition means a single point of failure?!?!?!

As a reference, a friend of mine in our airport had a hall effect sensor for his ignition and it failed on the run up area. The magnet had worked itself out. He was really lucky that he was not airborned yet.

Second incident again in our airport with another RV7A and dual electronic ignition with a redundant power supply, a second battery. His primary power source failed on take off and his second battery was in poor condition (unknowing to him) which caused an engine out at 300-400 AGL. Kudos to his piloting, he landed on the taxi way with no injury or damage and he has since fixed the design of redundant power source. Our airport is not a friendly airport if one has an engine out on take off.

So to reiterate your point about the importance of a well designed and tested redundant power source.
 
Keith,
I agree in principal with your comment but if you are implying that not having a self generating system like Slick or PMAG would mean a single point of failure, than I would have to disagree. Although I have been wondering having a hall effect sensor for both ignition means a single point of failure?!?!?!


So to reiterate your point about the importance of a well designed and tested redundant power source.

A self generating system has advantage because it provides isolation, independence and continuity. It makes it difficult (but not impossible) to experience a sequential failure that takes out both ignition paths, is more difficult to have latent failures that lead to loss of power for an apparent first failure. It is very possible to design a single battery/dual alternator power bus (Z-12) solution that relies entirely on a monitored dual path individually battery backed ignition that has no latent failures, no cascading common mode failures, meets the availability goals and has limited maintenance and component periodic replacement needs. I put together a template system when considering all the trade study options before settling on my preferred choice. How you weigh the relative parameters depends on individual choice and past experience. If protection from latent and cascading failures is important you will end up with a solution that other may consider over designed that adversely skews the trade study. We each have to pick our poison.

KT
 
SDS-CPI-2 Status Report

My SDS CPI-2 ignition has been running reliably now for over 4 years and 600 hours. At every oil change I inspect the wiring, coil and pickup mounts and there have been no issues at all. I have not had to change the programming or call the factory since the installation and flight testing were completed. I did use a timing light during the installation to determine the actual firing angle. I used the LOP function a few times during cross country flights to advance the timing 5 degrees at over 6000 feet to reduce fuel flow to under 4 GPH in my O-235 Lycoming engine at 135 K/H TAS. I change the NGK 3961 spark plugs from RockAuto at each annual and I observed a few small lead deposits inside the bottom plugs. I now change them 2X a year, since so inexpensive ($1.75 ea).

There was one occurrence last year when the master buss voltage mysteriously dropped to zero and the backup battery immediately took over and continued to power the ignition and essential avionics for the remainder of the 15 minute flight back to the airport. The backup battery circuit functioned as designed without skipping a beat. During this flight while trying to troubleshoot the fault, the backup battery was switched off momentarily and the engine quit and it got very quiet, the backup battery was immediately switched back on and the flight continued. I have tested the backup system on the ground, but never fully flight tested it. Now, after this occurrence, I know that this backup system will get me home safely. I tested all the components in the master system and could not find the root cause of the fault, so I replaced the master solenoid, solenoid wiring, diodes and master switch. This problem has not reoccurred. I am very satisfied with this conversion and feel confident that I made the right decision.
 
Last edited:
Back
Top