M20J won't start and my mechanic is stumped

[PT20J]

1 hour ago, MisfitSELF said:

I’ve kept on digging and what I’ve managed to find is that my A3B6 may be properly configured with the jumper removed.  The Champion/Slick site isn’t really clear but my right Mag (different model number than my left mag) may not have the impulse coupler.

Lycoming shipped my A3B6 with Slick mags: 4372 (impulse coupling) on the left and 4370 (no impulse coupling) on the right. That is the condition that SHOULD have the strap installed as the purpose of the strap is to ground out the right mag so it doesn't operate during starting. Since the right mag doesn't have the impulse coupling, the spark will not be retarded during start and it will fire the plugs too early at the low cranking rpm without the strap. You only remove the strap if you have two impulse mags installed or if you have the dual mag on the A3B6D. Check with your A&P.

Skip

[RogueOne]

On 9/4/2016 at 1:04 PM, pfactor said:

I think I'm going to have to decide between spending around the same amount of money for either a SlickStart or new Tempest fine wires. I'm leaning towards the new spark plugs since I haven't seen any support for the SlickStart idea.

Oh, and the shop that overhauled the magnetos last year says they'll re-inspect and fix them for free. So we're going to do that as well.

The Champions are no bueno.  There is an “in between” regarding Champion massive and Tempest Fine Wire.  Tempest Massives.  I installed and they are a vast improvement.  Do you have a lightweight starter and extra cranking battery?  These made a nice difference in getting the prop spinning for start.  Good luck “finding the culprit in starting dilemma.  My bet is on the Slick mags/mag. 

[201Steve]

Resurrecting this thread. 77j w/ a1b6D I am having trouble starting as of today it was nearly impossible to start and I’ve never had this problem. The D3000 “new” from Kelly is about 400 hours. After some research; it appears I can isolate the p leads simply by removing the p leads and installing the  timing posts. Sounds like this should make both mags all time hot. If it starts, focus on the switch and remedy. If it continues to hard start (or not at all); focus on magneto\impulse coupling. (Side not here; make sure the grounding strap is not installed as it was called out in Mooney si m20-59a applicable to my serial number to ensure impulse coupling is energizing both magnetos). 
 

Today; I cold started OK ish. Seemed to fire a little funny but whatever. Trying to hot start; it was nearly impossible; and propellor back spinning was observed on a slight false start. Very similar results ans the original posters complaints.  After much agonizing I finally got a cylinder firing with a lot of follow thru keeping the starter engaged much longer and barely got it going where it then ran fine eventually. It sounds like it started at 25 advanced and not from the impulse coupling based on commentary regarding back spinning. 
 

ok so eliminate the switch with timing posts; proceed from there. I feel like impulse coupling is the culprit. I looked to see if you could buy an impulse coupling assembly; with no results. Is it true that you can only buy parts for the impulse coupling and rebuild? That’s what it sort of sounds like. Is impulse coupling a part of a standard R&R or overhaul? I’ll probably just send it out as it is near 500 hours anyway. 
 

thanks for any feedback!

[EricJ]

21 minutes ago, 201Steve said:

Resurrecting this thread. 77j w/ a1b6D I am having trouble starting as of today it was nearly impossible to start and I’ve never had this problem. The D3000 “new” from Kelly is about 400 hours. After some research; it appears I can isolate the p leads simply by removing the p leads and installing the  timing posts. Sounds like this should make both mags all time hot. If it starts, focus on the switch and remedy. If it continues to hard start (or not at all); focus on magneto\impulse coupling. (Side not here; make sure the grounding strap is not installed as it was called out in Mooney si m20-59a applicable to my serial number to ensure impulse coupling is energizing both magnetos). 
 

Today; I cold started OK ish. Seemed to fire a little funny but whatever. Trying to hot start; it was nearly impossible; and propellor back spinning was observed on a slight false start. Very similar results ans the original posters complaints.  After much agonizing I finally got a cylinder firing with a lot of follow thru keeping the starter engaged much longer and barely got it going where it then ran fine eventually. It sounds like it started at 25 advanced and not from the impulse coupling based on commentary regarding back spinning. 
 

ok so eliminate the switch with timing posts; proceed from there. I feel like impulse coupling is the culprit. I looked to see if you could buy an impulse coupling assembly; with no results. Is it true that you can only buy parts for the impulse coupling and rebuild? That’s what it sort of sounds like. Is impulse coupling a part of a standard R&R or overhaul? I’ll probably just send it out as it is near 500 hours anyway. 
 

thanks for any feedback!

It does sound like it could be the impulse coupling, especially if you got some backwards energy during start attempts.   And, yes, the impulse coupling should be part of a normal 500-hour/IRAN or overhaul.   It is considered part of the magneto.

[N201MKTurbo]

The impulse coupling is a simple mechanism. I would just pull the mag and look at it.

One thing you can check without pulling the mag, is just pull the prop through and see if it still snaps. If it doesn’t you could have a broken spring or the post fell out.

[201Steve]

Well, went back today and it started pretty well from cold. I didn't have a lot of time, but let it warm for about 5 minutes and tried 2 subseqent "warm" starts and it fired up. Will be curious to see if perhaps it is a true hot start issue after the next flight to full operating temps.

Interesting commentary above regarding baking a magneto in an oven for a true stress test.

 

Could be a sign of an internally degrading magneto that's just showing early signs. We shall see.

[201Steve]

Question for the mag gurus in here.... there isn't any good reading about how the mag retardation works on the impulse coupler. I get that it has a spring and spins the magneto shaft at a higher speed, but I'm not understanding how that also creates a retarded timing. Seems like, all other things equal, a faster spinning shaft would simply ignite the spark at the same time as it normally would. Would like to understand this better......

[MikeOH]

19 minutes ago, 201Steve said:

Question for the mag gurus in here.... there isn't any good reading about how the mag retardation works on the impulse coupler. I get that it has a spring and spins the magneto shaft at a higher speed, but I'm not understanding how that also creates a retarded timing. Seems like, all other things equal, a faster spinning shaft would simply ignite the spark at the same time as it normally would. Would like to understand this better......

I believe there is a separate set of points (retard points) that are set to fire later (TDC?) which are active when the key is in the 'start' position, and disconnected when in R, L, or BOTH positions.

[Joe Linnebur]

On 1/28/2025 at 1:07 PM, 201Steve said:

Question for the mag gurus in here.... there isn't any good reading about how the mag retardation works on the impulse coupler. I get that it has a spring and spins the magneto shaft at a higher speed, but I'm not understanding how that also creates a retarded timing. Seems like, all other things equal, a faster spinning shaft would simply ignite the spark at the same time as it normally would. Would like to understand this better......

We just got done dealing with the same issue you’ve been dealing with, and it was after a mag overhaul. Many months later, my suggestions to replace the impulse coupling was taken and the problem fixed. 

The prop kickback is a good sign that maybe your IC spring is losing tension and not retarding the firing (ie. it’s snapping/spinning the mags through before TDC causing a cylinder to light off early and push it towards bottom dead center). Ours was inconsistent in starting issues initially, then became atrocious to start. 
 

You can YouTube some good vids like this. Basic, but a nice visual. If you watch this, the retarding occurs by the flyweight getting caught, and the flyweights are set to release at about Top dead center. The Spring tension then spins the mag through at a high rate of speed creating a usable spark for starting. So, if your spring tension is starting to wane, or internal wear in the coupling is letting that flyweight loose too early, you end up where you (and I) are at. Hope this helps.

 

[201Steve]

@Joe Linnebur great feedback thank you. I did watch that video but it doesn’t say anything about retarding the timing which is the part. I’m not quite understanding, but I’m getting there.

[Bolter]

26 minutes ago, 201Steve said:

@Joe Linnebur great feedback thank you. I did watch that video but it doesn’t say anything about retarding the timing which is the part. I’m not quite understanding, but I’m getting there.

I knew less before I watched that video.  Is the timing a fixed thing based on where that post is located?  The catch is tripped, the mag spins enough to make a spark, and does so based on the mechanical location of the catch pin and the rotating tooth.  When running, the points take over the timing.  

[EricJ]

Because the mechanics of the impulse coupler dictate when the magnet snaps (and all the internal parts with it, including the cams, points, etc.), that sets the timing for the starting spark.   The post holds the timing static, then the release dictates when the spark happens.   So the "advanced" timing is delayed until the release disengages the coupler, so the spark happens at that instant.

 

[A64Pilot]

18 hours ago, 201Steve said:

@Joe Linnebur great feedback thank you. I did watch that video but it doesn’t say anything about retarding the timing which is the part. I’m not quite understanding, but I’m getting there.

At low RPM as in starting the Mag rotation stops around -20 BTC, it stays there and continued engine rotation winds up a pretty strong spring. At or very close to TDC the spring is released making the mag rotate at a high enough RPM so that it generates a strong spark. If you rotate the prop slowly with the plugs out for safety, the “snap” you hear is the impulse letting go quickly snapping the mag until it’s caught up with engine rotation.

At a higher RPM, idle  the impulse coupling is disabled and it doesn’t function so timing is returned to normal.

The impulse coupling serves two functions, it reduces timing to prevent kick back of course, but it also spins the mag fast enough to produce a strong spark.

The retard timing is fixed of course, so if your mag is built for a 20 degree timing, and you decide to increase timing to 25 degrees because you read on the Internet that’s better, you need to change the impulse coupling to one with 25 degrees of retard or you run a risk of kickback that can break things.

Of course if your engine data plate says 20 degrees, then if you time to 25 then you’re not within the TCDS and the engine isn’t airworthy. Usually changing the data plate is a BIG deal involving the manufacturer or an STC or similar. I have not researched the work involved

On dual single mag systems there usually is essentially no difference between the mag with and without the impulse coupling, a dual (combined as one case ) mag of course uses both mags for starting as the impulse drives both simultaneously.

Several aircraft with dual single mags have both mags with impulse couplers, my 1946 C-85 for one, other than cost I can’t see why all mags don’t have impulse couplers myself. It definitely helps with starting. It’s easy to see, rotate the engine slowly with plugs out and you hear two snaps.

 

Edited Saturday at 04:02 PM by A64Pilot

[PT20J]

Aviation Maintenance Technician Handbook - Powerplant Vol 1 - FAA-H-8083-32B has a description of the impulse coupling.

The attached service instruction is approved data for changing the timing between 20 and 25 deg and remarking the data plate. (This SI is still active, but not on Lycoming's website because Lycoming's policy is to only post the most commonly requested service literature available on the website and this one is very old news).

SI 1325 Timing Change for IO-360 Series Engines.pdf

[A64Pilot]

3 hours ago, PT20J said:

Aviation Maintenance Technician Handbook - Powerplant Vol 1 - FAA-H-8083-32B has a description of the impulse coupling.

The attached service instruction is approved data for changing the timing between 20 and 25 deg and remarking the data plate. (This SI is still active, but not on Lycoming's website because Lycoming's policy is to only post the most commonly requested service literature available on the website and this one is very old news).

SI 1325 Timing Change for IO-360 Series Engines.pdf 470.94 kB · 4 downloads

If I read that SI correctly it gives directions and therefore approval to reduce the timing from 25 to 20 degrees, but does not give instructions and approval to increase timing from 20 to 25 degrees.

Oh, and it also doesn’t give approval for (D) single drive mags, if again I read it correctly 

Although it does give an engine model with the D in it indicating the big dual mag?

Either way you can decrease timing back to 20, but can’t increase from 20 to 25, not with this SI anyway.

Edited Saturday at 11:16 PM by A64Pilot

[PT20J]

46 minutes ago, A64Pilot said:

If I read that SI correctly it gives directions and therefore approval to reduce the timing from 25 to 20 degrees, but does not give instructions and approval to increase timing from 20 to 25 degrees.

Oh, and it also doesn’t give approval for (D) single drive mags, if again I read it correctly 

Although it does give an engine model with the D in it indicating the big dual mag?

Either way you can decrease timing back to 20, but can’t increase from 20 to 25, not with this SI anyway.

The TCDS specifies 25 degrees with 20 degrees being optional. Since the SI allows changing the data plate when changing from 25 to 20, there should be no reason not to allow remarking it back to 25 since that's what it was originally per the TCDS.

The dual mag engines must be 25 deg because the mag is not available with decreased impulse coupling lag angle to support 20 degrees engine to mag timing.

[N201MKTurbo]

5 minutes ago, PT20J said:

The TCDS specifies 25 degrees with 20 degrees being optional. Since the SI allows changing the data plate when changing from 25 to 20, there should be no reason not to allow remarking it back to 25 since that's what it was originally per the TCDS.

The dual mag engines must be 25 deg because the mag is not available with decreased impulse coupling lag angle to support 20 degrees engine to mag timing.

If the impulse coupling is set to 25 degrees and you time it to 20 degrees it will fire 5 degrees after TDC. I can’t see how this would hurt anything and should start just fine.

[N201MKTurbo]

BTW, I have run my engine at both 20 and 25 degrees. It runs better LOP at 25 and better ROP at 20. I always run LOP so it is set to 25. There was no difference in starting.

[PT20J]

1 hour ago, N201MKTurbo said:

If the impulse coupling is set to 25 degrees and you time it to 20 degrees it will fire 5 degrees after TDC. I can’t see how this would hurt anything and should start just fine.

1 hour ago, N201MKTurbo said:

BTW, I have run my engine at both 20 and 25 degrees. It runs better LOP at 25 and better ROP at 20. I always run LOP so it is set to 25. There was no difference in starting.

If you tried it and it worked fine, I can't disagree.  I've never tried it. I'm just quoting Lycoming about why they never approved 20 degrees timing for the dual mag engines. 

What was "better" in each case? Smoother, lower CHT, more power? Just curious since I haven't tried it.

[N201MKTurbo]

17 minutes ago, PT20J said:

If you tried it and it worked fine, I can't disagree.  I've never tried it. I'm just quoting Lycoming about why they never approved 20 degrees timing for the dual mag engines. 

What was "better" in each case? Smoother, lower CHT, more power? Just curious since I haven't tried it.

It runs cooler ROP at 20 degrees. Which was the intent of the change. It seems a bit more efficient at 25 LOP. 

[A64Pilot]

On 2/1/2025 at 7:01 PM, PT20J said:

The TCDS specifies 25 degrees with 20 degrees being optional. Since the SI allows changing the data plate when changing from 25 to 20, there should be no reason not to allow remarking it back to 25 since that's what it was originally per the TCDS.

The dual mag engines must be 25 deg because the mag is not available with decreased impulse coupling lag angle to support 20 degrees engine to mag timing.

I guess we could argue that. Once you change a data plate usually there has to be a method as in SI, STC, Custom kit etc to change it back.

Truthfully as long as you could get a data plate I doubt anyone would care, Data plates are usually closely guarded by manufacturers.

[A64Pilot]

Flame front movement is slower LOP than it is ROP, some of the old big radials had a timing switch that would increase timing when you ran the engine LOP when you used the switch, yes even way back then LOP was known and very widely used. So increasing timing when LOP is in effect not increasing timing, but correcting for the slower flame front and putting it back to where it should be. One reason for lower CHT’s when LOP is from essentially reduced timing.

This from this link https://enginehistory.org/Operations/R-4360Ops/r-4360ops1.shtml

5. Manual leaning and the use of spark advance enhanced the range capabilities of the aircraft because of improved specific fuel consumption. By carefully maintaining recommended cylinder head and carburetor air temperatures with stable turbo supercharger operations, charge density could be held at an efficient value. Since flame speed is typically reduced at leaner fuel air ratios below approximately 0.077, advancing the magneto/engine timing from 20° to 30° spark advance maintained peak cylinder pressure at the most effective 15° after top dead center position. This combination, manual leaning and spark advance also lowered exhaust gas temperatures, which favored longer valve exhaust, system and turbo supercharger life. Last but not least cowl flap openings could be reduced because the engine ran cooler.

The issue with bumping timing isn’t that it causes any issue at LOP, it’s that it runs hotter, engine life is therefore shortened etc when run at high power ROP, like in T/O and climb etc.

So you are improving LOP performance slightly, at the expense of engine longevity if Lycoming is to be believed.

This is in the SI linked to above.

Ideally we need manually adjustable timing like was done in some of the old radials, but while the fuel burn difference was significant when you had over 16,000 cu in of engines I’m not sure it’s that big when you have only 360.

But yes as just a general statement advancing timing increase efficiency and power output, way back I think in 1930’s Buick was the first to fit Vacuum advance because at low power an engine could tolerate it and it significantly improved fuel efficiency and of course being vacuum operated it automatically pulled timing at higher MP’s to prevent overheating and detonation. Modern cars of course use knock sensors and run much closer to detonation and of course more efficiently too, but they on average rarely produce full power and when they do it’s for very limited duration

 

Edited Monday at 10:26 PM by A64Pilot

[Pinecone]

16 hours ago, A64Pilot said:

Flame front movement is slower LOP than it is ROP, some of the old big radials had a timing switch that would increase timing when you ran the engine LOP when you used the switch, yes even way back then LOP was known and very widely used.

APS and Deakin, Atkinson, and Braly did not invent LOP.  They just "re-discovered" it from the days of piston airliners and military.  They looked and found the manuals on operating those big radials across the oceans and their "invention" was showing that the same principles applied to our little engines.

[A64Pilot]

7 hours ago, Pinecone said:

APS and Deakin, Atkinson, and Braly did not invent LOP.  They just "re-discovered" it from the days of piston airliners and military.  They looked and found the manuals on operating those big radials across the oceans and their "invention" was showing that the same principles applied to our little engines.

They also named it, usually it was just called Economy cruise.

The P-38 manual has a procedure where you established cruise power, turned off the electric props making them fixed pitch then slowly leaned it out until you got an RPM drop, not sure but I think it might have been 25 RPM, after leaning it out you turned the props back on.

Many think that Lindbergh taught 38 drivers in the Pacific how to run LOP but he didn’t, LOP was in the POH. What he taught was low RPM / high manifold pressure to get the same power as higher RPM cruise. Lindbergh flew about 50 Combat Missions and shot down at least 1 Japanese aircraft, as a Civilian, higher echelon wasn’t pleased and the unit Commander was I think maybe relieved of duty maybe?

Most aircraft types would just lean it out until it got rough then advance just enough to run well, most Radials will run smoothly WAY LOP because they have identical intakes and the Supercharged ones especially had very evenly atomized fuel mixture after going through the Supercharger.

It really wasn’t in most cases until the adoption of fuel injection that flat motors could run LOP, procedure was essentially run them as lean as they would run smooth which often was real close to peak, way I was taught anyway.

Edited Tuesday at 10:49 PM by A64Pilot

[A64Pilot]

On 2/1/2025 at 7:09 PM, N201MKTurbo said:

If the impulse coupling is set to 25 degrees and you time it to 20 degrees it will fire 5 degrees after TDC. I can’t see how this would hurt anything and should start just fine.

That is correct, I was speaking if you had a 20 degree coupling and wished to bump timing by 5 degrees for better LOP performance. Then your firing 5 before TDC, is that enough to hurt anything? I don’t know.

So why don’t cars have to retard timing for starting? Is it because of the mechanical advance is retarded at starting RPM? Old points type ignition.

You do retard our Model T

Edited yesterday at 12:07 AM by A64Pilot