MEP 831A Governor Control Issue? (video)

[dxhend2]

Having some issues with a MEP-831A that I think might be the governor control, would like to get some second opinions on this.
After sitting for a couple of years, I got it started for a load test. Ran about 5 minutes at 3000 watts, stumbled and died. Couldn't get it to start again, so went through the tech manual troubleshooting guides for ENGINE CRANKS BUT FAILS TO START. From Table 3-2. Operator-Level Troubleshooting, step 2.c: With START/RUN/STOP switch in RUN position, ensure governor actuator lever (see item 1, Figure 2-4) is releasing from magnet. It wasn't releasing, so...If governor actuator lever does not release, refer trouble to Unit-level maintenance.
Went to Table 4-4. Unit-Level Troubleshooting, for 2. ENGINE CRANKS BUT FAILS TO START. All checks passed.
Tried flipping the governor actuator off of the magnet, then it starts, cycles up in down in RPM, then stalls. Did this a few more times hoping to do more troubleshooting from 4-13 GOVERNOR CONTROL MODULE MAINTENANCE, which requires it to be running, loaded and unloaded. Each time I could start it right up, and it would run for a couple of minutes under load before it would start cycling up and down in RPMs and stall. Each run would go for a bit longer, and would start with the governor actuator still on the magnet with S1 in the RUN position.
I made a video with a split screen showing the governor actuator alongside the control panel so you can see what's happening inside and outside. This time it ran for almost 10 minutes at 3000 watts before cycling out again (I just went ahead and shut it down. Yes, should have opened the circuit interrupter first, but it was well on its way down again).
The real finale starts at 9:15 if you want to jump ahead. There are some other periods of instability at 3:30 and 5:50-7:00. Is this normal movement and positioning for the governor under a constant load?

I don't think this falls under any of the troubleshooting scenarios I could find in the TM. So, the question is...is this a candidate for replacing the governor control (A5)? Or, should I go through procedure 4-14 GOVERNOR ACTUATOR ASSEMBLY MAINTENANCE? I've read a lot of positive comments about @kloppk 's digital replacement board; seems like a better solution in the long run. Or, is there something else I'm missing here?

[video]

 

[kloppk]

1) I noticed the white Molex connector and wires for the actuator were getting caught on the actuator arm throughout the video. You need to reposition them, so they don't interfere with the actuator.

2) There is at least one A5 problem. The actuator should lift off the magnet the moment you move S1 from STOP to RUN. Yours didn't lift off until you were cranking it for a second or two.

3) I spotted that the spherical nut on the bottom of the rod was lifting well off the black governor throttle arm. When that happens, the GC is trying to increase RPM but the black governor arm is not rising further to increase engine RPM. A lot of time in the video this is the case. Could be the black START/STOP knob is not set correctly. Or it could be due to the A5 being bad and trying to increase RPM when it shouldn't be.

4) It looks like the wire and lead seal are missing from the top governor adjustment bolt. If so that factory adjustment has been fiddled with. That bolt should never need adjustment.

5) It sounded like the engine was starving for fuel a few seconds before you shut it off. Might be a fuel delivery issue.

6) It does appear that the A5 is not governing properly from the way it was slamming the actuator around during a constant load.


A) Reposition the Molex connector and wires.

B) I'd recommend seeing if there is a fuel delivery issue. With it under full load and the manual START pin left pinned IN see if the engine starts to sputter during a load test indicating a fuel issue. Leaving the manual start in holds the throttle at near WOT.

C) Have you tried to perform the A5 adjustments and Gap adjustment per the -13&P TM? If not, you might want to give that a shot.
Be sure to check the AC voltage / Frequency across A5 terminals A & B to be sure the RPM is correct for no load (3,050 RPM) and 3,450 RPM for full load. The TM has the correspond AC voltage and Frequency for determining the engine RPM.

 

[dxhend2]

1) I noticed the white Molex connector and wires for the actuator were getting caught on the actuator arm throughout the video. You need to reposition them, so they don't interfere with the actuator.

2) There is at least one A5 problem. The actuator should lift off the magnet the moment you move S1 from STOP to RUN. Yours didn't lift off until you were cranking it for a second or two.

3) I spotted that the spherical nut on the bottom of the rod was lifting well off the black governor throttle arm. When that happens, the GC is trying to increase RPM but the black governor arm is not rising further to increase engine RPM. A lot of time in the video this is the case. Could be the black START/STOP knob is not set correctly. Or it could be due to the A5 being bad and trying to increase RPM when it shouldn't be.

4) It looks like the wire and lead seal are missing from the top governor adjustment bolt. If so that factory adjustment has been fiddled with. That bolt should never need adjustment.

5) It sounded like the engine was starving for fuel a few seconds before you shut it off. Might be a fuel delivery issue.

6) It does appear that the A5 is not governing properly from the way it was slamming the actuator around during a constant load.


A) Reposition the Molex connector and wires.

B) I'd recommend seeing if there is a fuel delivery issue. With it under full load and the manual START pin left pinned IN see if the engine starts to sputter during a load test indicating a fuel issue. Leaving the manual start in holds the throttle at near WOT.

C) Have you tried to perform the A5 adjustments and Gap adjustment per the -13&P TM? If not, you might want to give that a shot.
Be sure to check the AC voltage / Frequency across A5 terminals A & B to be sure the RPM is correct for no load (3,050 RPM) and 3,450 RPM for full load. The TM has the correspond AC voltage and Frequency for determining the engine RPM.

@kloppk thanks for the detailed reply. I got a 1 hr 50 min minute run in today with more data.
For the top governor adjustment bolt (missing lockwire), measurement is 0.840" from top of bolt to base flange. I think that's higher than some of the recommendations I've seen.
Item A - corrected. Molex connection and wire was not interfering with actuator, but needed to be secured.
Item B - ran for 90 minutes at 3000 W as you suggested, steady throughout; clear exhaust throughout run. Before starting all of this, I drained and replaced the fuel, replaced fuel filter, and gave it a dose of Sea Foam. Can hopefully eliminate fuel issues.

Item C - probably best to give the timeline of how this run went.

Started right up and governor began cycling; locked in the manual start pin to keep the actuator off the magnet. Skipped the readings for 4-13 Test 2 unloaded since it wasn't at idle. Loaded it to 3000 W, actuator was cycling. Had to hold the manual start pin in since the actuator moving open would release it. Got it stable without the pin after a few minutes (spherical nut off fuel lever seat). Took A5 readings under load:
Pins F&G: 29 VDC (nominal 24)
Pins J&H: 0 VDC (nominal 11)
Pins D&E: 4.8 VDC (nominal 0)
Pins A&B: 207 VAC (nominal 193+/-2)
Pins B&C: 206 VAC
Pins C&A: 206 VAC

25 minutes - settled out, actuator-magnet gap had slowly closed and was holding at 0.5". No audible change in RPM. Took another round of voltages:
Pins F&G: 28.3 VDC (nominal 24)
Pins J&H: 16.4 VDC (nominal 11)
Pins D&E: 4.8 VDC (nominal 0)
Pins A&B: 201 VAC (nominal 193+/-2)
Pins B&C: 201 VAC
Pins C&A: 201 VAC

30 minutes - minor oscillations in actuator (<1/8").
38 minutes - actuator had steadily moved to full open position (spherical nut about 1" off of the fuel lever), no audible change in RPM, load steady. Voltages:
Pins J&H: 26.0 VDC (nominal 11)
Pins D&E: 4.9 VDC (nominal 0)
Pins A&B: 202 VAC (nominal 193+/-2)

52 minutes - some movement down in actuator position, slowly and steadily moving down.
55 minutes - spherical nut reseated onto fuel lever. Minor oscillations in actuator and fuel lever together (<1/8").
58 minutes - actuator opening back up slowly.
60 minutes - actuator steady at 1/8" off full open.
90 minutes - no change, running smooth at 3000 W with clear exhaust, actuator still 1/8" off full open position.
Dropped load to 1500 W. Actuator moved down towards magnet slowly, stable. Spherical nut engaged fuel level. Minor oscillations (<1/16"), actuator-magnet gap at 0.35".

95 minutes - Went back to full load. Actuator slowly moved up, spherical nut off lever.
98 minutes - Actuator full open.
100 minutes - Went to half load, actuator moved back to 0.35" gap, steady.
101 minutes - Went back to full load, actuator started moving up, spherical nut off lever.
102 minutes - Went to half load, actuator moved back to 0.35" gap, steady.
103 minutes - Went to no load. 0.3" gap, minor oscillations in actuator (<1/8"). Started taking no load voltages.
Pins F&G: 27.8 VDC (nominal 24)
Pins J&H: 9.2 VDC (nominal 5.3)
Pins D&E: 85.8 VDC (nominal 0) (yes, checked this 3 times)
Pins A&B: 189 VAC (nominal 178+/-2)
108 minutes - Before I could get the next ones, it had some instability and dropped to a lower idle speed (audible difference), with some minor cycling (also audible in RPM).
Continued voltage checks:
Pins A&B: 161 VAC (nominal 178+/-2)
Pins B&C: 161 VAC
Pins C&A: 161 VAC

110 minutes - Checked actuator-magnet gap at 0.25", bumped actuator while taking measurement. It destabilized and hit the magnet, shutting the engine down.

I'm trying to make sense of this sequence. It took a few minutes to stabilize, then at 25 minutes it seemed to be working as expected for full load, then drifted off to full open. At 55 minutes it had traveled back down to an expected normal position again, then moved back to full open and stayed there until 90 minutes when I cut the load to 50%. It seemed to stabilize where I'd expect it for that, but when going back to full load, the actuator wanted to go full open (tried that sequence twice). Going to no load at 103 minutes initially was looking good, but then it had that instability 5 minutes later.

 

[kloppk]

"Pins D&E: 85.8 VDC (nominal 0) (yes, checked this 3 times)"
It was probably actually millivolts or microvolts and your meter autoranged.
The signal across D&E is normally almost 0 VDC at No load and just about 5 VDC at full load. It's the load signal from the inverter to the A5 to tell it how much load the inverter is under so that the A5 sets the RPM correctly.

My hunch is that there is an issue with A5 since your GC has random oscillations and erratic behavior.

If you like I can ship you one of my replacement A5's, install it and do the no load gap adjustment per the TM. No charge for now.
If it resolves the issues and you want to keep it you can then pay for it. If you decide you don't want it just ship it back.
Having it will help determine if your A5 is goofy or not.

 

[dxhend2]

"Pins D&E: 85.8 VDC (nominal 0) (yes, checked this 3 times)"
It was probably actually millivolts or microvolts and your meter autoranged.
The signal across D&E is normally almost 0 VDC at No load and just about 5 VDC at full load. It's the load signal from the inverter to the A5 to tell it how much load the inverter is under so that the A5 sets the RPM correctly.

My hunch is that there is an issue with A5 since your GC has random oscillations and erratic behavior.

If you like I can ship you one of my replacement A5's, install it and do the no load gap adjustment per the TM. No charge for now.
If it resolves the issues and you want to keep it you can then pay for it. If you decide you don't want it just ship it back.
Having it will help determine if your A5 is goofy or not.

You're right on that meter. Was a brand new one with autoranging I was using for the first time on this test.

I think we've pretty much zeroed this in to a definite A5 issue. I appreciate your gracious offer to send me a "maintenance assist module" as we used to say, and try it first. I'm certain this is a sale. I've read about how you designed and prototyped a digital replacement board - impressive engineering, and a great service to this community.

I'll send you a DM with my info. Thanks!

 

[CallMeColt]

I had this battle with the one I went through a while ago. kloppk was very helpful. I made two videos on adjusting everything;

 

[dxhend2]

I had this battle with the one I went through a while ago. kloppk was very helpful. I made two videos on adjusting everything;

Colt, thanks. I've been offline for a bit with a cold/flu bug, back up now. Great videos - I've been through those in detail a couple of times - definitely helps make sense of the TM.
Also got some great help from @kloppk . Some more tests to do, may have multiple issues to work on this gen set. Learning a lot. I'll post some more soon.

 

[CallMeColt]

Colt, thanks. I've been offline for a bit with a cold/flu bug, back up now. Great videos - I've been through those in detail a couple of times - definitely helps make sense of the TM.
Also got some great help from @kloppk . Some more tests to do, may have multiple issues to work on this gen set. Learning a lot. I'll post some more soon.

Just saw you are in San Antonio, I am not far away in Floresville.

 

[dxhend2]

Replaced the A5 controller board. Generator started after a few seconds. Closed circuit interrupt switch, got green indicator light. Stepped it up to full load and ran for 20 minutes. Governor was steady throughout. Shut down manually.

Came back the next day to do a longer load test. Had been raining (gen set is in enclosed shop, but very damp weather - tech manual was pretty soggy). Would start it, get an OVERLOAD SHORT CIRCUIT fault within a few seconds and the unit would shut down. Disconnected all wires (except ground), same thing repeated. Found this thread about damp weather issues, sounds like this was the issue (a separate problem to work?). https://www.steelsoldiers.com/threads/831a-starter-issues.181793/#post-2466300

Waited a couple of days for dry weather. Started again right away, no fault indicators. Closed circuit interrupt switch, got green indicator light. Went ahead and ramped up load to 3000 W. Ran steady for 20 minutes, then it tripped the load and shut down - no faults indicated. Trying to restart, engine will start for a couple of seconds then immediately shut down with no fault lights on the fault monitor. (Actuator goes to the magnet).

Went through this cycle a couple of more times. Yesterday (dry weather), it started right away, and I ramped it up incrementally to 100% load (25%, 50%, 75%, 100% - 10 minutes at each step). Could hear a slight adjustment up in RPM at each increment, smooth transitions. Then ran it at 100% for a full 60 minutes. Governor was steady the whole time with @kloppk new controller board. Coming back down in increments, it did fine until I went from 50% to 25% - the generator shut down, and I couldn't get it to restart (would start briefly then immediately shut down, with no fault indicator lights).

Tried again today (damp weather). Looking at the K12 relay per @kloppk suggestion, relay is energized (green LED on) when S1 is in RUN, goes out when going to START and stays out while engine is running. Engine will run for a couple of seconds, then shut down. K12 doesn't energize again (green LED on) until after the engine completely stops. Tried again with K12 relay removed, same thing.

Could this all be caused by damp components in the inverter as discussed in the other thread above?

 

[dxhend2]

Just saw you are in San Antonio, I am not far away in Floresville.

Thanks, I'm in Lakehills, NW of San Antonio, not far.

 

[kloppk]

Don't bother running it with K12 removed. Doing so prevents the Governor Controller from every getting power.

If the set shuts down while running AND K12 isn't illuminated when the engine comes to a stop indicates it stopped for some reason other than the Fault Monitor detecting a fault. Since it shut down with the actuator going to the magnet without K12 illuminating can be due to the idle gap not being set correctly or the magnet has been fiddled with and its strength set to high causing it to capture the actuator when it shouldn't.

 

[dxhend2]

I think I have three different issues with this gen set that have had me chasing symptoms all around - I "think" I have it figured out now. Quick summary:
1) Faulty governor controller board A5 (original issue). Corrected, thanks to @kloppk 's replacement board; closing this thread.
2) Mechanical settings in the governor assembly. 95% done, thanks to a great thread from @CallMeColt in 2019 with some excellent guidance by @kloppk. https://www.steelsoldiers.com/threads/mep-831a-fix-up-thread.190228/
3) Humidity/moisture in the inverter, as discussed by @kendive and @Ray70 (with another reference from @kloppk). Next item to work on this gen set. https://www.steelsoldiers.com/threads/831a-starter-issues.181793/#post-2466300

Details:
1) When I am able to get the set started and ramp up to full load, it runs great with the new A5 controller from @kloppk. Very steady, great solution. Highly recommend this replacement.

2) From @CallMeColt 's MEP 831A Fix-up Thread, the recommended governor mechanical settings were from @kloppk measured on two of his units, all in inches (Recommended). I measured my unit (Observed) and made adjustments below (Final Corrected):

Top adjustment bolt
        Recommended    Observed    Final Corrected
        0.765, 0.793    0.831        0.785
    
Bottom adjustment bolt
        Recommended    Observed    Final Corrected
        0.890, 0.900    0.900        0.897
    
Condition 1: Generator off and target contacting magnet.
        Recommended    Observed    Final Corrected
Distance A     0.00        0.00        0.00
Distance B     4.100       4.213       4.100

Condition 2: Generator off and actuator "pinned" for a manual start. Spring loaded pin in holding actuator in near WOT position for a manual start.
        Recommended    Observed    Final Corrected
Distance A     1.00        1.127        1.130
Distance B     3.52        3.571        3.563

Condition 3: Holding the actuator as far away from the magnet as possible.
        Recommended    Observed    Final Corrected
Distance A     1.265        1.451        1.392
Distance B     3.360        3.581        3.575

I still need to check the gap between the actuator and the magnet (should be 5/16") at unloaded idle, but need to get a stable idle to do that. Will have to go back and check the A and B measurements again after that.

3) Focusing on the possible Humidity/moisture in the inverter. Attempting to start it two days ago, it would start then immediately shut down with an OVERLOAD SHORT CIRCUIT fault. The red fault light on the fault indicator panel would come on, at the same time, K12 would energize (green LED inside on) and then the engine would shut down. On previous day's tests, it would start and immediately shut down without any faults indicated (K12 would not energize until after the engine shut down). @kloppk mentioned above that this could be due to mechanical settings in the governor, so I dug into that one (Issue #2).

Tried again yesterday and today (dry weather), same results each time (start then immediately shut down with an OVERLOAD SHORT CIRCUIT fault).
At this point I think I need to try the inverter fix in the other thread (bake and seal the circuit board).

 

[CallMeColt]

Have you tried adjusting where the thumb nut is tightened down? That solved the final issue I had with mine. Not to say it will solve yours.

 

[dxhend2]

I'll double check that one. I did loosen and re-tighten it before all the way down in the START position, but after changing those other adjustment bolts, it could be off. Good check.

 

[CallMeColt]

I'll double check that one. I did loosen and re-tighten it before all the way down in the START position, but after changing those other adjustment bolts, it could be off. Good check.

I don't remember exactly, but I had to adjust mine slightly off the start position. I know I documented it in either my adjustment videos and/or my thread. I think it was causing a surging issue at no load and once I turned things down, it was good. So many projects since then has made me not remember exactly, I'm sorry!

 

[dxhend2]

I don't remember exactly, but I had to adjust mine slightly off the start position. I know I documented it in either my adjustment videos and/or my thread. I think it was causing a surging issue at no load and once I turned things down, it was good. So many projects since then has made me not remember exactly, I'm sorry!

No problem, that was many months and projects ago.
Found it! Post #99 from @kloppk in your fix-up thread. I'll take a look at that.

MEP 831A Fix-up Thread

 

[dxhend2]

I tried adjusting the Speed Control Knob off of the RUN position, no effect on the issue.
After doing a lot more testing, here's how it seems to be consistently acting :
From the normal starting condition (actuator on the magnet), it will crank but not start (I think I only got one start like this early in the testing). That had me going all through the operator and unit-level troubleshooting procedures for ENGINE CRANKS BUT FAILS TO START. The only anomaly seen was in the Operator-Level Troubleshooting
2. ENGINE CRANKS BUT FAILS TO START
c. With START/RUN/STOP switch in RUN position, ensure governor actuator lever (see item 1, Figure 2-4) is releasing from magnet.
The actuator would not release, however, I'm not sure that would/should prevent it from starting. I couldn't find any issues from the Unit-Level Troubleshooting.

If I kick the actuator off the magnet before trying to start it, the engine starts right away, revs up, produces power (indicated on panel voltmeter) and the actuator quickly moves back to the magnet and the engine shuts down. There are no fault indicator lights, and relay K12 does not energize until after the engine shuts down. (I had a few OVERLOAD SHORT CIRCUIT fault indications on some of the earlier days of testing. The weather was damp; that may have been a separate issue. Now, it's consistently shutting down with no fault indicators). I'm topping off the charge on the batteries between each test set.

I'm stumped on this one. @kloppk mentioned earlier it could be something with the magnet capturing the actuator when it shouldn't. The only change before all this was it was shipped by truck to our current location.

 

[kloppk]

With my A5 in there the actuator will not lift off the magnet until you put S1 in START. The moment S1 goes into the START position my A5 sends a pulse to the actuator to lift the actuator off the magnet. If this is not happening, then it would appear the magnet strength may be set too high preventing the actuator from lifting off the magnet.

Suggestion... Lift the actuator off the magnet. Start the set and hold the actuator off the magnet so it's running at a higher RPM. Then very slowly let the actuator swing toward the magnet and see if it stays off the magnet or get captured by the magnet and shuts down.
If it stays running check the gap between the magnet and the actuator. It should be 5/16".

 

[dxhend2]

With my A5 in there the actuator will not lift off the magnet until you put S1 in START. The moment S1 goes into the START position my A5 sends a pulse to the actuator to lift the actuator off the magnet. If this is not happening, then it would appear the magnet strength may be set too high preventing the actuator from lifting off the magnet.

Suggestion... Lift the actuator off the magnet. Start the set and hold the actuator off the magnet so it's running at a higher RPM. Then very slowly let the actuator swing toward the magnet and see if it stays off the magnet or get captured by the magnet and shuts down.
If it stays running check the gap between the magnet and the actuator. It should be 5/16".

I gave that a try. Every time I'd let it ease down, the magnet would grab it.
I also tried loading it while holding the actuator off the magnet. I could step it up to full load and the actuator would stabilize, but it seems very close to the magnet - less than the elusive 5/16" gap at no load. Quick video clip:

 

[kloppk]

I would suggest adjusting the rod length per the TM and see if you can achieve 5/16" gap at no load and have it remain running. Maybe a tiny bit wider if the magnet grabs it.