Found broken start ring cogs

[David-aviator]

Am doing condition inspection and have had to replace the start ring on fly wheel due to 4 broken gear cogs.

The issue is kick back from electronic ignition on start. In talking with Robert Paisley, this is happening because I do not have back up 12 volts for the system. Voltage drops during start and EI looses its brain and fires way off schedule.

I stopped using EI for start some time ago when this happened repeatedly. It does not happen starting with left impulse coupled mag. I have mag hooked up to top plugs so it starts very easily. When engine is running smoothly, turn on EI. No big deal.

It is not a starter issue, am using Sky-Tec NL unit, it spins engine very nicely and its gear cogs are like new.

 

[rv6ejguy]

Unlikely battery voltage is dropping too low as the CPU will run down below 7.5V. Doubtful the starter will crank the engine if the voltage is dropping that low.

You could check to see if the CPU reset due to interference or low voltage by putting the display in gauge mode first, then cranking. If it stays in gauge mode the CPU did not reset. If it goes back to the bootup screen, it's either resetting or losing power/ ground. If the CPU cannot execute all the code, it will auto reset which takes about 0.5 seconds.

I suspect possible interference with the Hall sensor signal, possibly from close proximity to the starter cable which may be inducing a voltage exceeding logic level in the cable.

Do not share ECU grounds with any other terminal/ device. Do not share power from any other terminal, especially coil power or anything that has high voltage/ amperage or pulsing current on it.

Don't tie the CPU and ECU driver grounds together.

 

[DanH]

Observed ECU brownout on the Barrett dyno, first start attempt of John Walker's 540. Another battery was wired separately to drive the EFII system.

 

[Andrew Anunson]

The issue is kick back from electronic ignition on start. In talking with Robert Paisley, this is happening because I do not have back up 12 volts for the system. Voltage drops during start and EI looses its brain and fires way off schedule.

Do the installation instructions require the installation of a back up battery for starting? Some here like to start the engine on the EI, since their backup ignition is a non-impulse magneto. If the flyefii ignition requires a complete backup battery system just for starting... well this sounds like a non-starter to me.

 

[rv6ejguy]

Decent battery/ cable setups should be able to maintain 9 volts when cranking the engine. If not, you should look there first. The controller will work down below 7.5V.

When the engine cranks slowly, the pistons slow noticeably coming up on compression, which effectively advances timing. This can cause a kickback.

Default retard when cranking is 15 degrees. Default timing for cranking is 10 degrees so actual timing when cranking with a good starter battery setup should be around 5 degrees ATDC. If you have V27 or later software, cranking retard is user programmable.

EFII uses the SDS ECU. 500,000+ flight hours, 25 million ground hours, many thousands in service for over 20 years. I've personally got collectively over 10,000 hours on the system in cars and aircraft. I've started our cold soaked shop car down at -34C with this system where it barely turned over. No kickback. You shouldn't need a separate battery to start the engine unless you have some wonky wiring or a poor battery/cable setup.

 

[David-aviator]

Do the installation instructions require the installation of a back up battery for starting? Some here like to start the engine on the EI, since their backup ignition is a non-impulse magneto. If the flyefii ignition requires a complete backup battery system just for starting... well this sounds like a non-starter to me.

With single EI and mag, backup power is not required.

But there is an issue of power available for EI during start without BU power, it is true of all EI systems. I had Electoraire years ago, it too broke start ring. On some starts all is well, others it is not. I believe my system would not reset to TDC if engine spun up and then quit due to inadequate priming.

Does the engine start better using EI, of course it does, there is more voltage to plugs. But starting on mag is a non issue, we've been doing it for over 75 years. I have no problem with it, hot or cold engine. Deliberately have mag hooked to top plugs. Net result is have a 20 rpm drop with both systems during mag check. EI is stuck with bottom plugs, it all evens out.

The Robert Paisley EI system is good, go with it if you are inclined to do high altitude flying. If most flying is below 8000', it does not matter than much. I went with single EI because no back up power required and I believe it provides about 95% of what you get with dual EI.

 

[erich weaver]

But there is an issue of power available for EI during start without BU power, it is true of all EI systems.

This isn't true. Pmags for one work well down to the 9 volt range. If your voltage is dropping this low during start, your battery is shot or something is wrong in the wiring. There are large numbers of us flying on electronic ignition with no backup battery and no starting problems.

Erich

 

[Toobuilder]

...But there is an issue of power available for EI during start without BU power, it is true of all EI systems...

This is a bold and absolute statement. As such, it requires some further exploration.

I've used dual Pmags on two airplanes and dual CPI on the latest and have never had any form of back up, starter boost, auxiliary battery, etc, and I have never experienced a brownout, lost timing or kickback event. Both the Pmags and SDS CPI will fire the plugs at voltages far below that which will turn the engine over. So if your engine spins over "very nicely", then you do not have a EI issue, you likely have a wiring issue. And as wiring issues (specifically bad grounds) are so very common in equipment problems, I'd look hard at your particular architecture and execution to see if things are up to snuff.

Bottom line is that a single EI on a 4 banger pulls about 1 amp at maximum RPM, and will fire at very low voltage. There should be no need for any form of electrical "boost" during start. After all, cars don't run dual batteries to keep the ignition alive during the start event.

 

[rv6ejguy]

The way (angle) that the engine stops is always the same. That means the first FULL compression stroke is the second cylinder to compress on the 6, after that the speed (inertia) yields lower RPM variation.

The 4 cylinder only fires twice a revolution and has high torque demand on the starter each compression stroke as there is no overlap of an expansion stroke from another cylinder. This means the RPM variation is greater on the 4 than the 6 though a revolution.

It should be noted that the 4 cylinder torque requirement on the second compression stroke it is higher than the 6 which gets some help from other cylinders. What might work on the 6 might not work on the 4.

The ECU establishes the time interval between magnet/Hall sensor passage for one complete crank revolution before initiating any spark events. The system is vastly proven with literally millions of starts on 4 cylinder engines.

 

[rcpaisley]

EI start

This is one of the benefits of the Bus Manager - protect the 12v supply to the ECUs during starting.

If an existing install develops this issue, it is usually a battery getting weak.
If an original installation with a fresh battery has this issue, it could be voltage drops in the starter power wiring. Grounding wire layout issues are VERY common
in experimental aircraft.

In any case, it is typically a voltage droop issue during cranking that causes this,

Robert

 

[Toobuilder]

...It should be noted that the 4 cylinder torque requirement on the second compression stroke it is higher than the 6 which gets some help from other cylinders. What might work on the 6 might not work on the 4...

No disagreement there, but is this a practical consideration? No. There are countless examples of 4 banger EI only starts - they are "typical" in fact. David's situation seems to be the exception. Let's try solve his issue rather than create a new one.

The start circuit is a massive draw on the system, yes, but a healthy, well designed and maintained system should not black out the ignition when the starter is engaged. If it does, there's a problem with that particular equipment, not the concept in general.

 

[David-aviator]

Thanks for input.
I am no expert on how these things are designed or work during start event.
All I can do is report what happens.
Everything is new so battery, starter, etc. are not an issue.
In beginning before proper priming was established, engine would start, spin up and quit. Sometimes would kick back during spin down, once ran backward for short time.
Second attempt to start with EI powered resulted in kick back almost every time.
It was as if EI was not firing at TDC but earlier like trying to start with non implulse coupled mag.
My experience is with this system and early Electroaire, Lightspeed and Pmag I know nothing about but thought they had similar issues.
Obviously they don't.

 

[DanH]

Everything is new so battery, starter, etc. are not an issue.
In beginning before proper priming was established, engine would start, spin up and quit. Sometimes would kick back during spin down, once ran backward for short time.
Second attempt to start with EI powered resulted in kick back almost every time.
It was as if EI was not firing at TDC but earlier like trying to start with non implulse coupled mag.

David, you have something funky going on. Let's get some basic info.

Ross mentioned V27 or later software features user programmable cranking retard. What do you have, and what did you set?

Is your single EI wired directly to the battery positive, to the main bus, or where?

What starter are you using?

 

[Toobuilder]

I am not completely up to speed on current Pmag offerings, but I believe they have a healthy ignition retard setting these days to make sure a light prop or a slow cranking starter makes it over the hump before firing. As a data point my -8 has a relatively small PC680 way back in the tail, I swapped out the cables to much smaller #6AWG, 200HP, and one of those power hungry mini starters... Dual Pmags has never been an issue starting, even with a weak battery. I've never seen a kickback event.

As Dan points out, the latest SDS offerings (CPI and EM-5) have a user definable start retard so that they will provide a safe, soft start if you have it set with the appropriate retard. That said, I'm not sure if the EFII configuration of the SDS CPU offers this, but it's something to look into.

OTOH, if you are actually "browning out" the CPU during start, then you need to look at your wiring, IMHO. That should not happen. Agreed... Something funky there.

 

[rcpaisley]

test

It is very simple to determine if you are having ECU reboot during cranking (most likely situation - trust me, I've dealt with over 400 airplanes).

Hook up a separate battery to the ECU power for a test. Make sure to connect the grounds of both battery systems.

See what happens. If you no longer have the issue, you know you need to protect the 12v to the ECU during cranking.

Robert

 

[DanH]

David, re-reading your first post I see you have a Sky-Tec NL. That would seem to make kickback damage to the ring gear a lower probability. The inline NL incorporates a shear pin to fail the starter drive, on purpose, before it can hurt the engine case or ring gear.

Early NL's had a small diameter shear pin, and would decouple the drive shaft if (to paraphrase Monty Python) some farted in its general direction. I can't speak for the torque capacity of the larger shear pin in later models, but I assume it works in service.

Quick check....look at your NL starter. Assuming it is installed conventionally (i.e. with a separate start contactor), do you have a jumper wire between the big terminal and the S-terminal?

 

[David-aviator]

David, re-reading your first post I see you have a Sky-Tec NL. That would seem to make kickback damage to the ring gear a lower probability. The inline NL incorporates a shear pin to fail the starter drive, on purpose, before it can hurt the engine case or ring gear.

Early NL's had a small diameter shear pin, and would decouple the drive shaft if (to paraphrase Monty Python) some farted in its general direction. I can't speak for the torque capacity of the larger shear pin in later models, but I assume it works in service.

Quick check....look at your NL starter. Assuming it is installed conventionally (i.e. with a separate start contactor), do you have a jumper wire between the big terminal and the S-terminal?

Dan, I need a day or so to answer your questions, have to pull the cowl to check this stuff for sure.

This much I can tell you, 4 cogs were broken on the ring and the starter did not sheer its pin.

There is no control over timing of EI, it is preprogrammed. It is as it came from Robert. I have a monitor device but removed it because there was nothing to monitor. Timing is about 24 for take off and advances to max of 30 in flight. Don't need to look at that every day.

I do not have EFII, just EI.

 

[David-aviator]

It is very simple to determine if you are having ECU reboot during cranking (most likely situation - trust me, I've dealt with over 400 airplanes).

Hook up a separate battery to the ECU power for a test. Make sure to connect the grounds of both battery systems.

See what happens. If you no longer have the issue, you know you need to protect the 12v to the ECU during cranking.

Robert

And accept risk of busting another ring?

It probably won't happen with 12v protection to ECU. Its your system and you know it better than anyone. I accept your explanation of why it happened without testing it

I don't want a second battery, this is a max effort light weight RV-8 coming in at 1029 lbs. it flies pretty good as is with Catto.

Do not mind starting on mag and bring EI in when engine is running. Just don't want to change ring again or add weight.

 

[lr172]

Thanks for input.
I am no expert on how these things are designed or work during start event.
All I can do is report what happens.
Everything is new so battery, starter, etc. are not an issue.
In beginning before proper priming was established, engine would start, spin up and quit. Sometimes would kick back during spin down, once ran backward for short time.
Second attempt to start with EI powered resulted in kick back almost every time.
It was as if EI was not firing at TDC but earlier like trying to start with non implulse coupled mag.
My experience is with this system and early Electroaire, Lightspeed and Pmag I know nothing about but thought they had similar issues.
Obviously they don't.

David,

I have one EI (home-grown) and one mag. I also have experienced the kickback on "spin down." It doesn't happen often, so I haven't worked on addressing it. I got a small corner cracked off a gear from this. There is a lot going on during this phase, so many issues to consider, but it is not power related. My guess is that only those of us with light FP props are seeing this with EI. My EI has a setting for what RPM is considered high enough to move from starting to running, but doesn't have a minimum RPM to trigger a starting event. I think that as inertia falls off, with no force from the starter, the piston is slowing down too much on the compression stroke. The EI is using time primarily to anticipate the next tooth (the actual tooth appearance then is used to determine any error) and at the slow RPMs of the taper off it can't deal with the error correction fast enough and we get one or two early spark events. I am guessing there would be a lot of testing to get this figured out.

Ross could better answer, but symptoms point to the s/w remembering something that it shouldn't for the second start attempt. It sounds like first start is good, but kicks on spin down. Then second start attempt kicks back during the spin up and starting cycle. If I had those issues (I don't - my second start attempt behaves just as the first), I would cycle the EI OFF then ON for each start attempt to force it to re-calibrate itself. Your successful first start attempt confirms that it is doing this properly, but something is happening when that start attempt doesn't follow the normal cycle of progress and goes back to 0 RPM and then back into the cycle.

Larry

 

[David-aviator]

It should be pointed out that (EFII) electronic ignition is part of a larger system incorporating electronic fuel injection. Both are controlled by the ECU (engine control unit).

It is different than other electronic ignition systems, a bit more complex.

Software makes it work and also provides for pilot control over mixture and timing, something not available in straight auto ECU operations, like Subaru.

Going beyond Subaru, there was not total agreement with regard to whether a pilot should have control over timing and mixture. The end result was intended to be free of need to mess with either and just go fly. But pilots by nature like to be in control, Ross in Canada has made that possible. Robert's system comes from there. The EI system I have does not have a provision for pilot input, but probably does with total package.

The fly-in-the-ointment here may be having just the EI part of it. The entire package incorporates dual everything including electric power source. I don't have it.

The cowl was just reinstalled after fire wall forward inspection. At the moment I would rather go fly than take it apart again to answer questions about start wiring. I have done a number of these over the years and don't believe it is an issue but will check and report next time I pull cowl.

Robert knows what he has created and I am satisfied with his response.

My procedure to start with mag works and am satisfied it is safe. Have had no issues with EI using that start technique. Engine runs smooth and plugs are in good shape. Am using auto plugs with EI, aircraft plugs with mag. Mag plugs are on top.

 

[rv6ejguy]

David,

Ross could better answer, but symptoms point to the s/w remembering something that it shouldn't for the second start attempt. It sounds like first start is good, but kicks on spin down. Then second start attempt kicks back during the spin up and starting cycle. If I had those issues (I don't - my second start attempt behaves just as the first), I would cycle the EI OFF then ON for each start attempt to force it to re-calibrate itself. Your successful first start attempt confirms that it is doing this properly, but something is happening when that start attempt doesn't follow the normal cycle of progress and goes back to 0 RPM and then back into the cycle.

Larry

Once there is no rpm signal from the Hall sensor, the CPU starts over, waiting for one complete engine rev before initiating the first spark event. No need to reset ECU power. Software does the same thing every time- it doesn't break, otherwise engine starting and running would be a shambles.

Lightweight props have little ill effect at cranking rpms (150-250 in most cases). We've started engines with no props and very light flywheels with no issues. Obviously automotive engines have no propeller and much less inertia than a Lycoming with a 6 foot propeller.

This ECU and software is not some cobbled together, untested stuff as I've previously mentioned. Kickbacks are pretty much unheard of if the engine cranks smoothly and the timing and crank retard values are set correctly.

We had one customer in the bush with an almost dead battery even hand prop his Lycoming to life on 2 occasions. That is worst scenario as far as a stable turnover rpms goes.

I realize David does not want to experiment here and has a method to avoid kickbacks. I suspect that the rpm timing value at 500 rpm (cranking reference value) may be set too high and the MAP advance may be adding to that figure too. There is a simple, scientific reason why he's getting kickback and thousands of other users are not.

We recommend the rpm timing value be set at 10 degrees at the 500 rpm range on Lycomings and we don't recommend using MAP advance on Lycomings. Default cranking retard is 15 degrees and that is user programmable along with all other parameters if you have V27 software or newer. We use MAP retard at higher MAP to keep timing safe at high power. This may not be how David's system is programmed but he can check the values currently entered with his programmer. Also, we recommend all users verify the Magnet Position parameter with a timing light to make sure that programmed timing is indeed actual timing on the engine.

If you have 24 BTDC entered at 500 rpm, you may easily get a kickback. Also, if you have MAP advance in the 20-25 inches range, you'll get additional advance during the cranking phase. Both of these factors together will contribute to the possibility of kickback when cranking. In other words, if you're getting kickback, it's 99.9% likely that the programmed values are causing it.

 

[N941WR]

It sounds like the OP is running something other than a P-mag, but to answer Mike's question.

I am not completely up to speed on current Pmag offerings, but I believe they have a healthy ignition retard setting these days to make sure a light prop or a slow cranking starter makes it over the hump before firing. ...

P-mags with firmware Version 40 fire the plugs at ~4 degrees AFTER Top Dead Center for starting.

Prior to Version 40, P-mags fired at TDC. If the P-mag wasn't set perfectly, then they have been known to kickback, especially with a lightweight prop. That problem went away with Version 40.

Also, the P-mags will fire with less than 9 volts, I can't remember the exact number, but I think we have seen them fire at round 8 volts +/-.

 

[Toobuilder]

...There is no control over timing of EI, it is preprogrammed. It is as it came from Robert. I have a monitor device but removed it because there was nothing to monitor. Timing is about 24 for take off and advances to max of 30 in flight. Don't need to look at that every day...

David,

If this "monitor device" plugs into the CPU and is from Robert or SDS, then it is the "programmer". With it you should have complete control over all the ignition settings.

Is this the unit? ^^^^

Assuming your wiring is in fact adequate and your CPU is not shutting down while cranking, then you most likely have the ability to eliminate this kickback scenario with just a couple of keystrokes.

Any chance you can plug the programmer back in and report back the RPM, MAP and crank retard settings?

 

[David-aviator]

David,

If this "monitor device" plugs into the CPU and is from Robert or SDS, then it is the "programmer". With it you should have complete control over all the ignition settings.

Is this the unit? ^^^^

Assuming your wiring is in fact adequate and your CPU is not shutting down while cranking, then you most likely have the ability to eliminate this kickback scenario with just a couple of keystrokes.

Any chance you can plug the programmer back in and report back the RPM, MAP and crank retard settings?

Yes, I have this device. I used it to confirm timing at take off and in flight.
I have not checked out its programming features because I thought my unit was set by Robert and not programmable. Not sure where I got that information but certainly can get spun up on what I have and check preset values.

Will check back.

 

[Andrew Anunson]

Unlikely battery voltage is dropping too low as the CPU will run down below 7.5V. Doubtful the starter will crank the engine if the voltage is dropping that low.

and

You shouldn't need a separate battery to start the engine unless you have some wonky wiring or a poor battery/cable setup.

although

This is one of the benefits of the Bus Manager - protect the 12v supply to the ECUs during starting.

Am I hearing two different lines of thought from the experts? Many homebuilders fly with 1 EI and 1 Magneto and don't need or want a Bus Manager.

Is the Bus Manager a recommended (necessary) component when installing an EFII electronic ignition?

AND

If I start the engine with a weak battery (that is still strong enough to turn the engine over) and only a single EFII electronic ignition , should I expect a kickback if I don't have a backup 12V source?

Thanks! This is a very educational thread for me.

 

[rv6ejguy]

I'll let Robert comment on his own offerings. SDS does several things differently than EFII.

 

[David-aviator]

I'll let Robert comment on his own offerings. SDS does several things differently than EFII.

Yes and one of them is self programming the ECU.

Robert does not recommend it for pilots.

I tend to go along with that policy, it is easy to destroy an engine if it is not done with care and knowledge.

The premise of the system is ease of use and fun flying without concern for timing or A/F ratio, something theoretically managed by well programmed ECU.

At the moment I do not have a document explaining how to program the ECU, I have the device to do it but no instructions on how to do it.

 

[Toobuilder]

The instructions for the programmer are available on the SDS website.

And yes, it is possible to get yourself into trouble if you stray too far from reasonable settings, but the programming is not difficult by any means. Anybody who built and flies their own airplane has the required skill.

We are here to help, but need to know the values you have set presently.

Id like to see your:

Magnet position (and was it verified with a timing light on initial setup?)

Start retard

RPM values

MAP values

 

[David-aviator]

Once there is no rpm signal from the Hall sensor, the CPU starts over, waiting for one complete engine rev before initiating the first spark event. No need to reset ECU power. Software does the same thing every time- it doesn't break, otherwise engine starting and running would be a shambles.

Lightweight props have little ill effect at cranking rpms (150-250 in most cases). We've started engines with no props and very light flywheels with no issues. Obviously automotive engines have no propeller and much less inertia than a Lycoming with a 6 foot propeller.

This ECU and software is not some cobbled together, untested stuff as I've previously mentioned. Kickbacks are pretty much unheard of if the engine cranks smoothly and the timing and crank retard values are set correctly.

We had one customer in the bush with an almost dead battery even hand prop his Lycoming to life on 2 occasions. That is worst scenario as far as a stable turnover rpms goes.

I realize David does not want to experiment here and has a method to avoid kickbacks. I suspect that the rpm timing value at 500 rpm (cranking reference value) may be set too high and the MAP advance may be adding to that figure too. There is a simple, scientific reason why he's getting kickback and thousands of other users are not.

We recommend the rpm timing value be set at 10 degrees at the 500 rpm range on Lycomings and we don't recommend using MAP advance on Lycomings. Default cranking retard is 15 degrees and that is user programmable along with all other parameters if you have V27 software or newer. We use MAP retard at higher MAP to keep timing safe at high power. This may not be how David's system is programmed but he can check the values currently entered with his programmer. Also, we recommend all users verify the Magnet Position parameter with a timing light to make sure that programmed timing is indeed actual timing on the engine.

If you have 24 BTDC entered at 500 rpm, you may easily get a kickback. Also, if you have MAP advance in the 20-25 inches range, you'll get additional advance during the cranking phase. Both of these factors together will contribute to the possibility of kickback when cranking. In other words, if you're getting kickback, it's 99.9% likely that the programmed values are causing it.

If I read this correctly, it says SDS EFI EM-5 4V22, which probably means V22 software. In scrolling through did see 10 degrees at 500 rpm.

 

[Toobuilder]

It's been suggested that V22 is not (or has limited) adjustable. I'm not sure about that, Ross would be best qualified to comment.

That said David, did you go through the magnet position setup process before flight (external pointer, automotive strobe light, etc) or are you running it with the default values?

 

[rv6ejguy]

If I read this correctly, it says SDS EFI EM-5 4V22, which probably means V22 software. In scrolling through did see 10 degrees at 500 rpm.

Ok, with V22, you don't have user programmable cranking retard, it's set at 15 degrees so your 10 degrees at 500 rpm (the value the ECU uses for everything below 500 rpm) plus the 15 degrees of retard, should fire your EI plugs at around 5 degrees ATDC. This is pretty standard on how we ship out ECUs. Not sure if Robert changes the timing or not.

You are using the same values as hundreds of other Lycoming clients are using. As Mike mentioned, it's an excellent idea to verify your actual timing with a timing light and adjust MAGNET POSITION if they do not agree. To help diagnose your issue, as Mike said, we'd have to see all your timing values.

You could go to 5 degrees of rpm timing at 500 rpm which should fire the plugs at 10 ATDC. Should be impossible to have a kickback when firing the plugs even at 5 ATDC.

IF the engine kicks back, it follows that the EI IS firing the plugs too early BTDC. Your case is not usual at all so there has to be something different about your installation.

I'd be happy to reflash your ECU with the latest software if your plane is down for a few days- if you'd want to be able to program in more cranking retard.

 

[rcpaisley]

ECU

Our standard ignition timing curves work very well for these engines.
There is certainly nothing wrong with pilots making adjustments if they know what they are adjusting.

I have helped several people through the starting kickback issue. It has always been a voltage dip during cranking that was the culprit. As mentioned by another poster, this is a well known potential issue with electronic ignitions.

Sometimes this starts happening after batteries age and lose capacity, sometimes it is grounding/starter wiring/battery issues. In any case, it is never much of a mystery and it is not difficult to cure. Yes it does take an effort to get a result.

We can go around this mulberry bush 100 times, the voltage droop still has to be addressed if this is to be fixed.

Robert

 

[Toobuilder]

We have not established if there is a significant voltage droop, Robert.

We have not established if the magnet position has been set either. The basic timing could very easily be off enough to make the engine fire before TDC - and thats a kickback waiting to happen.

Deductive reasoning suggest to me that since his engine spins the engine easily yet he has experienced kickbacks, Id say the first place to look is the magnet position. If that's off, you'll get a kickback even with a healthy battery (or two, for that matter).

But you are correct in that both possibilities need to be investigated and resolved.

 

[David-aviator]

It's been suggested that V22 is not (or has limited) adjustable. I'm not sure about that, Ross would be best qualified to comment.

That said David, did you go through the magnet position setup process before flight (external pointer, automotive strobe light, etc) or are you running it with the default values?

The magnets were installed in the fly wheel by Robert and the pick up was installed as per instructions. I did not verify timing is correct. Will do so.

Friend Bill Peyton is good at this and we will see what where timing is set.

 

[Toobuilder]

The magnets were installed in the fly wheel by Robert and the pick up was installed as per instructions. I did not verify timing is correct...

Bingo! Though the magnets are placed with CNC precision, and the pickup is installed with the same care, there is no way to gurantee that the relationship is the same from engine to engine. You will be close with the defaults, and maybe even dead on, but you simply have to verify every single installation. This step is not optional. We've done 4 CPI installations (same hall sensor as you have) in my Rocket group and we are all different. A 5 degree swing one way or the other is not unheard of, and thats enough to erase your start retard. The good news is that it's an easy process and only has to be done once. With the magnet position verified, dont forget to document it in the logbook in case the programming is ever lost.

 

[plehrke]

The magnets were installed in the fly wheel by Robert and the pick up was installed as per instructions. I did not verify timing is correct. Will do so.

Friend Bill Peyton is good at this and we will see what where timing is set.

David, if you need any help send me a private message as I just set the magnet position on mine last week. I live in Defiance and fly out of 1H0. SUS is half way between and can stop and help. Very easy to do.

 

[David-aviator]

(Thanks Phil.)

Bill Peyton and I did some testing this morning.

!. Voltage drop during starter engage is from 12.4V to 10.2V. The battery actually read over 13V static and dropped to 12.4 with master switch on. This test with all plugs installed.

2. EI is firing at about 10 degrees past TDC for start. This test with top plugs removed using induction timing light. Starter spun engine at 140 rpm based on GRT EFIS reading.

So what's going on? My guess is is ECU is not resetting to start timing after first start attempt when engine quits and spins down due to inadequate prime.

 

[rv6ejguy]

(Thanks Phil.)

Bill Peyton and I did some testing this morning.

!. Voltage drop during starter engage is from 12.4V to 10.2V. The battery actually read over 13V static and dropped to 12.4 with master switch on. This test with all plugs installed.

2. EI is firing at about 10 degrees past TDC for start. This test with top plugs removed using induction timing light. Starter spun engine at 140 rpm based on GRT EFIS reading.

So what's going on? My guess is is ECU is not resetting to start timing after first start attempt when engine quits and spins down due to inadequate prime.

10.2V when cranking is excellent.

If the plugs are firing at 10 ATDC, it should be impossible to have a kickback. 140 rpm is good too.

I'll have to check the code and see how long the ECU waits before re-initializing but I believe it's less than a second. Wait 5 seconds before next cranking attempt.

 

[Toobuilder]

Looks like you had the timing light out... Did you do the synch magnet position setup process yet, and if so, what was the difference (if any) between the setting you had loaded, and the final required?

If you are actually seeing 10ATDC during cranking, then it looks like your magnet position is off, because we'd expect to see 5 ATDC.

I'd say it once again: before the next flight, you must verify the proper magnet position.

Also, there was little doubt that the battery was healthy based upon your earlier description, but did you take the voltage reading at the CPU or the battery? We know the starter is eager, we need to verify what the CPU is seeing during cranking. That could still be weak due to some wiring issue.

 

[David-aviator]

David, re-reading your first post I see you have a Sky-Tec NL.

Quick check....look at your NL starter. Assuming it is installed conventionally (i.e. with a separate start contactor), do you have a jumper wire between the big terminal and the S-terminal?

Yes on jumper.

Also re ECU power, direct off battery positive pole with spade fuse, ground is off bus with other stuff.

 

[David-aviator]

Stop the Press!!

Another fly is in the ointment!!

During testing of system yesterday the starter did not disengage several times.

The only way we stopped starter was master switch off. It was intermittent malfunction in key start switch or relay on firewall.

In discussing issue with Bill Peyton, we conclude this could be reason for broken gear ring.

I have a little blue light next to key switch, it is on when start relay at starter powered. Problem is it is normally hard to see, but I saw it yesterday and also when Bill signaled to stop rotation it did not stop.

We have more testing to do. It either key switch or relay on firewall.

 

[tim2542]

David make sure you check the magnet position with a timing light even though it appears you have discovered a cause not related to the ignition. Until you do you don't really know what the timing is.
Your magnet position value should be very close to 90, if it's not be suspect of your procedure.
Tim Andres

 

[Toobuilder]

...Your magnet position value should be very close to 90...

Yes, but...

The final value is also dependent on whether there is a dual or single hall sensor. For example, my dual sensor values are 86 and 91. If I had a single mounted, it should split the difference (87/88). That said, my Rocket buddy's with dual sensor pickups are a few degrees off of mine. The 5 degree split is the same because they are manufactured as a single unit, but where they are indexed from engine to engine is the variable.

That said, I know Ross centers the pad between the two case bolts, but I dont know if Robert's solution is the same. There's nothing sacrosanct about centering the pad at the 90 degree point, so it could be almost anywhere the manufacturer wants it.

 

[rv6ejguy]

Another fly is in the ointment!!

During testing of system yesterday the starter did not disengage several times.

The only way we stopped starter was master switch off. It was intermittent malfunction in key start switch or relay on firewall.

In discussing issue with Bill Peyton, we conclude this could be reason for broken gear ring.

I have a little blue light next to key switch, it is on when start relay at starter powered. Problem is it is normally hard to see, but I saw it yesterday and also when Bill signaled to stop rotation it did not stop.

We have more testing to do. It either key switch or relay on firewall.

I was drafting a post yesterday questioning if the starter was disengaging ok but then you said before that it started ok on the mag only so I cancelled that post.This seems like a more likely issue.

 

[Bill.Peyton]

We did indeed check the timing using a timing light. The timing during start was around 8 degrees ATDC. AFAIK the ignition is operating correctly during startup. The ring gear teeth were in all likelihood broken due to the starter motor hanging up. Either the ACS switch has failed or the start solenoid is hanging. In either case, that is causing the starter to stay engaged after the engine starts. Isolating the problem and replacing the culprit is tomorrows task....

 

[David-aviator]

David make sure you check the magnet position with a timing light even though it appears you have discovered a cause not related to the ignition. Until you do you don't really know what the timing is.
Your magnet position value should be very close to 90, if it's not be suspect of your procedure.
Tim Andres

There is no provision or instructions to check magnet position with EFII single EI as you describe. This is not SDS install although it comes from Ross.
There is gap measurement between magnet and pick up but no checking magnet position, it is precisely installed in fly wheel and that's it.

 

[Toobuilder]

The CPU, programmer and hall sensor are SDS products. There is no secret sauce that anyone has spread on the components to allow a universal fit on all engines without verification. If you scroll through the settings in the programmer, you should see a "mag pos" window. There will a number there. This number is the default magnet position in degrees that someone input as a best guess for your engine. The magnet position check is nothing more than verifying the timing the computer thinks it should be, with the actual timing on the flywheel (with a timing light).

For example, if the programming expects 10 degrees BTDC at 1000 rpm and below, then you simply mark the flywheel at 10 degrees and add a temporary pointer. Run the engine below 1000 Rpm and shoot the strobe light at it. If the 10 degree mark aligns with the temporary pointer, then you are good to go. If not, then go to the mag pos window and key the up or down arrow until you walk the mark into alignment with the pointer. When it aligns, then the value displayed in the window is your individual magnet position. Write this value down in the logs. Aside from marking the flywheel with a sharpie and adding a pointer, this step takes about 60 seconds even if you have to make adjustments. The instuctions can be found on the SDS site under the CPI or EM-5 pages.

You MUST verify this value, because all timing is based on this reference position. This check is just as important as verifying magneto timing or idle mixture settings on a new engine. You wouldnt slap a new overhauled engine on the airplane and fly without this setup step, right? Same thing.

Mike S has a good illustration of the pointer setup on his CPI thread, you can check it out if you need help. Of course, you can call me if you'd like. PM me and Ill walk you or Bill through it.

 

[David-aviator]

The CPU, programmer and hall sensor are SDS products. There is no secret sauce that anyone has spread on the components to allow a universal fit on all engines without verification. If you scroll through the settings in the programmer, you should see a "mag pos" window. There will a number there. This number is the default magnet position in degrees that someone input as a best guess for your engine. The magnet position check is nothing more than verifying the timing the computer thinks it should be, with the actual timing on the flywheel (with a timing light).

Michael, all what you say is relevant. But I say again, this is not a SDS install, it is EFII.

I bought the single EI system from Robert Paisley, EFII founder, a guy I have know for over 15 years going back to the early effort to use Subby engines and Subby ECU. I asked about seeing timing advance as we did in early days and his response was it is not necessary with this set up, the magnets and sensor are precisely positioned and it works. That is true. The purchase price of single EI does not include the programmer. Robert is stand up guy, knows his stuff, will back it up and I trust him.

But I wanted to see basic timing so bought the programmer from Ross in Canada, hooked it up and sure enough timing was 24-25 on take off and advanced to 30 in flight. It was working as advertised. End of story, I removed the device from panel, it was crowded.

There is a fundamental difference from SDS to EFII. It is philosophical and gives builder a choice from total programming control (SDS) to somewhat less programming control (EFII). The history of this goes back to Eggenfellner days when he wanted no pilot control of system (total factory set up) to Ross's much more liberal control of system.

Anyway, back to broken ring gear. We don't know what caused it, having discovered intermittent starter hang up engaged. The EI system is ok. It works as advertised.

But that being said, I have decided to remove it and go back to left-right mag operation. Probably first guy on planet to go from EI to mag.

There three fundamental reasons for doing so.

!. I do not need EI for type of flying I do these days, low level river valley patrol. EI is good for cruise above 8000', seldom get up there.
2. I need low level visual recognition and want Vans wing tip wig-wag LED light system for RV-8.
3. Resources are limited.

So EI is coming out and will trade for Vans LED light system or approximate value in money (around $800). EI stuff includes everything I have including programmer, plugs, inserts, wires, and O2 sensor system.

PS Incidentally, O2 sensing NOT required for EFII FI system. Another good feature of EFII, IMHO.

 

[rcpaisley]

timing

Just a clarification on engine timing with EFII.
We position the timing magnets in our machine shop with a precision indexing machine. The is no variation from engine to engine on magnet position with EFII systems. This is not a customer performed operation. Customers do not need to do machining or fabrication with EFII systems. This is all done in our CNC machine shop.

The timing magnets are in exactly the same position on all EFII equipped engines. The ECUs are also preset with the correct parameter for the timing of the crank trigger - again nothing here for the customer to do - it has already been done for you.

We don't believe the customer should have to do engineering to install their EFII system. That is our job.

Robert

 

[David-aviator]

Found the culprit....

Bill and I discovered faulty part with intermittent starter engage hang up. It is the relay on firewall, B&C part S702-1.

We proved it conclusively after getting it to hang up with many start attempts.

Bill disconnected 12V wire from start switch to relay and it did not open, stayed closed. It is hanging up mechanically or self powering internally.

Called B&C, they are sending a new one.