MEP-803a not making full power/tripping overload

Hi all, first time posting, but have been lurking and reading through the wealth of information on this site. I have worked on several of the 803 & 802s now, and have learned a lot. This particular unit will not make full power without tripping the overload. I'm using an old air handler with a 10kw heat strip with the squirrel cage fan. If I disconnect a few strips it will run those for over an hour (about 7kw checking with clamp on meter). Anything past that it trips the overload. With the battle short on, and adding a few small heaters, I have had it up to 13.5kw for about 15 minutes. Overload light comes on at about 1-2 minutes.

Things I have checked/done. Cleaned the S8 and S6 over and over, cycled the S8 100x or more, tried Deoxit, and a couple other contact cleaners.

Swapped the K8 with another one, same result. Also did the K8 test per the TM, light came on at about 10 seconds (manual says approx. 1 minute, both relays turned overload light on right about 10 sec)

Ohmed burden resistors, disconnected 1, 2, & 3 from K8 and 14 from the S8, all were right about 7.9, book says 7.8 is upper limit, is this enough to cause concern? While running with a full load, 5.6v on R-10 R-11, 8.3 v on R-12

CT coils appear to be correct, I believe it was 8 loops per circle.

Gone through and tightened/cleaned connections in the compartment where the S8 is and behind the control panel. Found several screws on the S8 that took a good half turn to snug up.

Compared wiring with another good unit, everything around the S8 appears to be in the correct spot.

S8 has been out of this unit before, rear mounting plate only has 1 bolt, and 1 corner is busted. All 4 little ones are in on the front, and the 2 larger studs are missing the nuts. Doesn't look recently as everything was equally dirty/dusty.

I ran the same 10kw-13kw load on the 3 phase setting and it lasted longer, but still tripped the overload.

Interesting thing I found was, on the burden resistors, R-12 gets noticeably hotter than R-10 & 11, I was watching with an infrared camera, a good 100 degrees hotter than the other 2. Also on the CT coils, the one closest to the S-8 was slightly hotter than the other 2 (10-15 degrees, just enough to show a difference on the camera. So I'm not sure if theres something, say a bad connection in the S8 that is causing the extra heat, or if the burden resistor is bad and theres my issue?

Unit was wet-stacked a little, at first it would pull the 10kw load with a good bit of smoke and carbon sparks, but have run it for about 2-2.5 hours at the 7kw load plus several 15 min runs at 10-13kw and it seems to have cleared up a lot. I don't think that is my issue at this point.

Unit was rebuilt in 2010 by the Albany, Ga Marine Corps base, according to the date/hours written on the filter, it was not used after rebuild, only a couple hours on the meter over what was written on the filter, date on filter matched date on rebuild plate.

Initially the unit ran longer than 1-2 min with the 10kw load, maybe 5-10 minutes. Unit needed a new fuel pump, so stopped, ordered a pump, replaced, and now it is tripping the overload every time. I did cycle with the S8 switch in that time, so though I just stirred up some dirty contacts, but cleaning it has not changed much. Have 2 other units that I cleaned the switch 1 time and that fixed both of them.

Any help will be much appreciated!

Sounds like a wire is not connected correctly. Double check the wiring on S8, S6 and the main contactor that sends the load to the lugs. I had a unit that was rebuilt similar to your unit and would trip the overload the same way. Turned out to be a wire hooked incorrectly to S8.

Welcome to SS.Triple check the wire locations by number, against the schematic. All it takes is 1 in the wrong position. Have you tried load testing this in any other voltage position? (like loading it up all on 120V, and setting the machine to output only that)?

Also have you tested this problem with the battle short? Its a useful little widget for problems like this. Get the set going, and get it lightly loaded under power. Flip the battle short up, and continue your loading. See if only the light comes on, but the machine continues to make power... or even with the battle short ON, does the machine jump offline?

I saw your post/comment about the wire being in the wrong spot on the S8 on your unit and checked all mine, S8 is good but need to get a little more thorough on checking the rest, will work on that today.

Light in the Dark said:

Welcome to SS.Triple check the wire locations by number, against the schematic. All it takes is 1 in the wrong position. Have you tried load testing this in any other voltage position? (like loading it up all on 120V, and setting the machine to output only that)?

Also have you tested this problem with the battle short? Its a useful little widget for problems like this. Get the set going, and get it lightly loaded under power. Flip the battle short up, and continue your loading. See if only the light comes on, but the machine continues to make power... or even with the battle short ON, does the machine jump offline?

Click to expand...

Thank you. SS has been incredibly helpful in learning about these units.

I have tried the same 240v load on the 3 phase setting, using the L1 & L3, similar result, did last a little longer before light tripped. I don't have anything big enough to put enough load to test on only 120v output, my air handler is 240v and the majority of my load tester.

It will load test to 13.5kw no problem with the battle short on, overload light comes on 1-2 min in, stays online no problem. Have done this for 15-20 minutes several times.

I would disconnect every wire in the AC portion of the set, and then reinstall them IAW the schematic. Someone else has been inside the set. They could not make it work right. Did you buy it this way?

Guyfang said:

I would disconnect every wire in the AC portion of the set, and then reinstall them IAW the schematic. Someone else has been inside the set. They could not make it work right. Did you buy it this way?

Click to expand...

I haven't had time yet to disconnect and reconnect every wire, but I have gone over all the wire positions for the AC side and they are all in the right spot. Only thing I found different was the T1-T12 wires are in numerical order on the terminal block vs. the diagram having them in a different order.

Yes I purchased the unit this way from GP. I did notice the zip ties around the harness near the S6 are gone, and the harness is very loose there vs. how it is normally very tidy.

I noticed the CT block is upside down on one unit, I have 2 one has all the part number stickers facing up, other is facing down. Does this matter? Wire orders are correct on both for CT coils.

S6 switch had a connection that was dirty. Cleaned that up, tightened every connection I could find, cleaned up the K1 terminals (they were pretty dirty) and now I can go about 8 minutes before overload trips with 10kw. The S6 switch feels "gritty" and not smooth sometimes, should I swap in a known good one to rule it out?

Using an infrared camera, the CT3 coil is 20-30 degrees hotter than CT1 while running with the 10kw load, would this indicate anything obvious?

Havent run into that problem in a mep but in my (rather cluttered) head I would think either the reconnection switch isnt hooked up properly, a winding in the genhead is bad, etc... Id say because the winding is heating up something is shorted or out of range to draw that current.
I am sure somebody has a better method than mine because theres manuals and schematics and stuff, but I grew up having to fix stuff without schematics so ill share what I would do.
I would start ohming the genhead since thats easy and just 12 wires that are hard to mess up (unless labels are gone). If that passes then make sure what Guy suggested is performed. After that Id check the reconnection switch for proper operation and ohm it in each setting using the s8 switch schedule

Theres further testing that can be done with the unit on but if its either of those I would think you would see something. That testing is more in-depth than I can articulate here while half asleep.

BTW Im in St Pete. Not bad if you end up not being able to fix it and want a 2nd/3rd set of eyes.

If the S6 switch feels strange, gritty, it's got dirt, sand, dust, whatever, in it.

In your first post, you said 8 loops through the CT's. Your set is a 10 KW. I am not in front of my TM's, but seem to remember 10 KW is 4 loops. Or am I mixing it up?

TM for the 803 says:
Note: Each wire makes 4 turns through the current transformer.

4 loops and 2 wires per CT right? I can see how one would see 8 loops.

I was kinda hinting that CnP should re check it. Having encountered this several times before, it can happen. I once saw a 30 KW that had the CT's wired up like a 60KW. It was rather impressive when we dumped load on it. If only for a second or two.

Demoh said:

Havent run into that problem in a mep but in my (rather cluttered) head I would think either the reconnection switch isnt hooked up properly, a winding in the genhead is bad, etc... Id say because the winding is heating up something is shorted or out of range to draw that current.
I am sure somebody has a better method than mine because theres manuals and schematics and stuff, but I grew up having to fix stuff without schematics so ill share what I would do.
I would start ohming the genhead since thats easy and just 12 wires that are hard to mess up (unless labels are gone). If that passes then make sure what Guy suggested is performed. After that Id check the reconnection switch for proper operation and ohm it in each setting using the s8 switch schedule

Theres further testing that can be done with the unit on but if its either of those I would think you would see something. That testing is more in-depth than I can articulate here while half asleep.

BTW Im in St Pete. Not bad if you end up not being able to fix it and want a 2nd/3rd set of eyes.

Click to expand...

Connections on S8 are correct, and good idea on the windings, checked those, all good there. Need to ohm connections on S8 still.

I appreciate the offer, I'll keep that in mind.

Guyfang said:

If the S6 switch feels strange, gritty, it's got dirt, sand, dust, whatever, in it.

In your first post, you said 8 loops through the CT's. Your set is a 10 KW. I am not in front of my TM's, but seem to remember 10 KW is 4 loops. Or am I mixing it up?

Click to expand...

Couldn't get the S6 cleaned up, so it has another one in it for now, no change. You are right, I mixed it up, was typing from memory on that post.

Demoh said:

4 loops and 2 wires per CT right? I can see how one would see 8 loops.

Click to expand...

Yes I think that's where I mixed it up, I will go back and double check, but I believe they are all correct.

Guyfang said:

I was kinda hinting that CnP should re check it. Having encountered this several times before, it can happen. I once saw a 30 KW that had the CT's wired up like a 60KW. It was rather impressive when we dumped load on it. If only for a second or two.

Click to expand...

That's interesting! I will double check though.

So for an update, I ran with the 240v single phase load on the 3 phase setting on L1 & L2, and it did good. Using the infrared camera again, the coils on the CT were nice and even heat wise, same for the burden resistors. When running the same load on L1 & L3, CT3 and R12 get hotter than their neighbors. So it seems the problem is in the L3 circuit. Using an amp meter at the K1, the L3 wire is pulling about 2 more amps than L1, (40 & 42 amps), is it normal to be that far apart?

On the K1, in there anything in the bottom that can be cleaned? I took the little cover off the top and cleaned all the terminal connections (they were dirty), but couldn't get anything else open.

I had a 003 like this. Turned out the brand new AC recconection(I believe S-6) switch was wired wrong at the repair depot. I ran out of patience after about 10 hours of trying to sort it out and reading the reconnection switch diagrams in the TM, I ended up swapping out the recconection box, and all was well.

Doing this I did find out two things, there were two different models of switch, S-6, and what might be relevant to your case, the MEP 002 and 003 have the same reconnection box, except for the number of winding around the CVT, and these sometimes get mixed up.

dependable said:

I had a 003 like this. Turned out the brand new AC recconection(I believe S-6) switch was wired wrong at the repair depot. I ran out of patience after about 10 hours of trying to sort it out and reading the reconnection switch diagrams in the TM, I ended up swapping out the recconection box, and all was well.

Doing this I did find out two things, there were two different models of switch, S-6, and what might be relevant to your case, the MEP 002 and 003 have the same reconnection box, except for the number of winding around the CVT, and these sometimes get mixed up.

Click to expand...

I did check all of those and everything is in the right place, and the CT loops are all correct also. I ended up going over every wire on the AC side and all is in the right spot. And this one has the older or original style switches too, all matches the diagram

Update on the unit, I have found that if I put the S6 switch in the 3 phase setting, making the gauges dead (I'm running 240 single phase), it will not trip the overload, the R12 burden resistor is not energized when the gauges are off like that, and that seems to be where the issue is.

Burden resistors were ohmed via the method found on this forum by disconnecting a couple wires on the K8 and S8, and they were 7.9 (TM says 7.9 ). Is that really an accurate way to check them or do they need to be unsoldered as the TM says? R12 also has 8v going through it while running (with S6 in the correct position) while others have 5.4v.

Have you tested CT3?

I am new to the MEP generators and considering purchasing one as a standby for my home. There is a local 802A advertised at a great price ($1000) with the description " It runs but has a couple minor issues. If you run a constant load around 4000 watts the breaker trips and overload light comes on. Smaller loads seem not to bother it ". Sounds very similar to the issues with the gen on this thread, perhaps a wiring issue and specifically S6. I am not a electrical technician, but like to work on cars and boats (I am retired from mechanical engineering). This seems like a tedious problem to troubleshoot, but not a serious problem with the gen set, would you agree?

Usually not a serious issue. Usually due to poor contacts in either S6 and/or S8.
Can usually be resolved using some contact cleaner such as Deoxit.

kloppk said:

Usually not a serious issue. Usually due to poor contacts in either S6 and/or S8.
Can usually be resolved using some contact cleaner such ad Deoxit.

Click to expand...

Thank you, I am going to pursue it. I love reading this forum, and find all of the problems and recommedations very ineresting.

Please start a thread if you continue asking questions. That's what we are here for, asking and answering questions. But the guy who started this thread has a problem that he is trying to solve, and old farts like me have enough problems remembering my own name three out of five days a week. You will get better help, and people who might want to help you will have no idea about your problems/questions if you do not have your own thread.