MEP-803a overload circuit

I was able to run another load test today. After start and warm up I progressively applied a 60% load. It ran fine for 12-15 min and then tripped the overload again. I reset everything and applied 50% load and it ran fine for the remaining 1:40. During the last 30-40 min I added a large burner on low. It would cycle on for 20-30 seconds and then remain off for 2-3 min. When on, it would step the load up to 75%, but wouldn't remain on long enough to trip O/L. This seemed to be a nice test and it handled the shifting loads fine. I'm still concerned that it won't take anything over 50% for a period of time without tripping. It should chug away with 60, 70, 80% or more.

Im feeling that thermal K1 O/L switch may be faulty. I'll have to read the TM to determine how it can be tested.

TNriverjet said:

I was able to run another load test today. After start and warm up I progressively applied a 60% load. It ran fine for 12-15 min and then tripped the overload again. I reset everything and applied 50% load and it ran fine for the remaining 1:40. During the last 30-40 min I added a large burner on low. It would cycle on for 20-30 seconds and then remain off for 2-3 min. When on, it would step the load up to 75%, but wouldn't remain on long enough to trip O/L. This seemed to be a nice test and it handled the shifting loads fine. I'm still concerned that it won't take anything over 50% for a period of time without tripping. It should chug away with 60, 70, 80% or more.

Click to expand...

Shouldn't it "chug away" with up to 100% load, isn't that why it's rated like it is?

K8 is the overload/short circuit relay. Test procedures for K8 are on page 2-58 of the -24 manual. I would also remove and open up K1.

Guyfang said:

K8 is the overload/short circuit relay. Test procedures for K8 are on page 2-58 of the -24 manual. I would also remove and open up K1.

Click to expand...

K8! Thanks Guyfang

Another Ahab said:

Shouldn't it "chug away" with up to 100% load, isn't that why it's rated like it is?

Click to expand...

Yes, my 802a would nearly kill itself trying to stay online with a huge load. I had the resistance heat (emergency heat) try to come online once while the 802 was happily running the rest of the house.

Ok, today I was able to run the test procedure for K8 from the -24 TM on page 2-58 and 2-59. All testing is conducted with the voltage reconnection switch in the 120/208 3P position.

There are essentially 6 tests to run in this procedure. The first 3 test operation/function of the OVERLOAD relay. There is one test per leg and testing is run on each of 3 phases.

The second set of 3 tests are run on the SHORT CIRCUIT relay. Again, each of the 3 phases are tested.

Here are my results:
Overload test one: O/L tripped in 9-10 seconds. (TM states this should take about 1 minute)
Overload test two: O/L tripped in 9 seconds (again 1m standard)
Overload test three: O/L tripped in 7 seconds (1m standard)
Interestingly the Hz readings were not consistent. The meter was reading about 56-57.5 Hz while I had the wires disconnected and tagged for testing

Short Circuit test one: immediate trip as per TM
Short Circuit test two: immediate trip as per TM
Short Circuit test three: immediate trip as per TM
Again, the Hz meter was not consistent when in test mode. Readings 57.5 - 60 Hz

The TM states if any of the above tests fail, the K8 relay needs replaced. Since the time factor is not clearly quantified, for the first series of three tests, I am not sure if this unit "passed" or "failed". Again the TM states, "After approximately one minute, AC circuit interrupter relay should open and OVERLOAD lamp should illuminate on malfunction indicator panel". It seems to me it is tripping too soon for this test, but the procedure did not say it had to endure 60 seconds to pass.

I'm planning to spend the remaining part of the afternoon taking K1 out and apart per Guyfang's recommendations.

Guyfang, did I get this sent out before bedtime Saturday?

You just made the deadline! I was going to turn of the box, and decided to check ONE MORE TIME, and low and behold, you made it. It sounds to me like you may need a K8. Can you borrow one?

You just made the deadline! I was going to turn of the box, and decided to check ONE MORE TIME, and low and behold, you made it. It sounds to me like you may need a K8. Can you borrow one?

Also, have you checked the Burden Resistors? R11, R12 and R13. The test procedure is in the book.

I was just out there testing the burden resistors per the TM, but I can't get readings that make sense. I've read somewhere that they should really be un-soldered and tested for most accurate readings. However, the TM reads that they can be tested in place. To complicate things, I'm using an analog ohm meter that is older me, so I may need to go buy a good new digital amp clamp with test leads to get better readings. I did replace the batteries in my old meter and can now get more consistent readings.

I'll see if I can borrow a K8 from Jamawieb and work through taking apart and cleaning inside K1.

I just pulled K1, and may have found the problem... Shouldn't this be a 60 Hz??? See photo of tag on K1 rated at 400 Hz.


Also, here are photos of the inside...


it's pretty clean. I haven't taken the Allen screws off the top yet, but am not sure I need to if this is a completely wrong part for an 803a. I'm wondering if this belongs in an 813a?

It looks like the coil voltage for that contactor is 28VDC so nevermind the 400Hz spec. With a component like that the most important info is the coil voltage and the current rating. The reason it says 400Hz on the side is most likely because a contactor that was rated for up to 400Hz was needed.

TNriverjet said:

I was just out there testing the burden resistors per the TM, but I can't get readings that make sense. I've read somewhere that they should really be un-soldered and tested for most accurate readings. However, the TM reads that they can be tested in place. To complicate things, I'm using an analog ohm meter that is older me, so I may need to go buy a good new digital amp clamp with test leads to get better readings. I did replace the batteries in my old meter and can now get more consistent readings.
I'll see if I can borrow a K8 from Jamawieb and work through taking apart and cleaning inside K1.

Click to expand...

You'll have to un-solder one side of the burden resistors. The TM does not say to but I can never get an accurate reading without un-soldering.

I may have an extra K8 I could send you for testing. I've never had a K8 fail, it is usually a burden resistor or dirty connection. First time for everything though.

This would have been a good time to exercise your logistical talents. Look up the K1 to see if it has a UOC, Usable On Code, to determine if it is a 60 hertz, or 400 hertz part. You can learn a lot by reading. Haoleb is correct. Control voltage is DC, 28 volts. And the relay is rated for use with 50/60-400 hertz.

Agreed. K8 is normally a very robust part. Have only seen maybe two bad ones. Usual culprit is the resisters. But the K8 test procedures tend to point to the relay. My only problem with the test is that the relay still is in the set, and not disconnected from the burden resistors. Or am I wrong? I assume that the K8 test, requires the K8 to sense an overload, for 60 seconds, before it pops off. That's what happens in real life. It doesn't pop off as soon as it senses an overload. Yours does, after a few seconds.

And as jamaweib said, always un-solder at least one side to test the resisters. You will never get a trustworthy reading any other way. Do not use too much heat when you de-solder. The posts, that hold the resisters in, are plastic. If you overheat them, they tend to break.

Yeh guys. I really enjoy reading and learning from you all. I'm pretty new here and just got my 803 all cleaned up and ready to wire into the house. However, I've read conflicting info about bonding the ground & neutral terminals when using the genset stationary. Your advice and knowledge would be greatly appreciated.

Johncar48 said:

Yeh guys. I really enjoy reading and learning from you all. I'm pretty new here and just got my 803 all cleaned up and ready to wire into the house. However, I've read conflicting info about bonding the ground & neutral terminals when using the genset stationary. Your advice and knowledge would be greatly appreciated.

Click to expand...

This has been discussed and debated a bunch. If you're using 4 wire hookup to the house with an independent ground I would just unbound them. To be honest, I've never un-bonded the unit I set up at my parents house because I forgot when I first installed and never went back to redo.

Thanks very much for the reply. Think I will keep them bonded. Did you work on these units in the military?

Johncar48 said:

Thanks very much for the reply. Think I will keep them bonded. Did you work on these units in the military?

Click to expand...

Howdy,
In the middle of a field... bonded strap in place., and ground is connected to ground rod in the dirt.


Your house electrical panel, you have a proper ground in place. You are using the ground which was put in when the house was built.
Generator is connected with a 4 wire connection. Bond strap removed. All 4 wires connected, 2 hots, neutral, and ground. Again, you are using your house panel ground.

Thanks Mike. Running 3 wires to house panel, 2 hots & a neutral. Panel is grounded. House is new but still under construction for I am building it myself. 2 a/c units (2 and 3 1/2 ton units) and lights produce about a 40% load factor but when demand hot water kicks in generator takes a dive south. Water heater rated at 4800 watts. Live in rural area so may need to switch to mep 003a which I just bought. 803 unit has 5200 hrs on it but is clean & runs well. 003 only has 1200 hrs on it but needs work. Have no idea when it was last run for fuel tank is missing, all filters have been removed & all fluids have been well drained. So it's a project. Appreciate the info & thanks for response.

Johncar48 said:

Thanks Mike. Running 3 wires to house panel, 2 hots & a neutral. Panel is grounded. House is new but still under construction for I am building it myself. 2 a/c units (2 and 3 1/2 ton units) and lights produce about a 40% load factor but when demand hot water kicks in generator takes a dive south. Water heater rated at 4800 watts. Live in rural area so may need to switch to mep 003a which I just bought. 803 unit has 5200 hrs on it but is clean & runs well. 003 only has 1200 hrs on it but needs work. Have no idea when it was last run for fuel tank is missing, all filters have been removed & all fluids have been well drained. So it's a project. Appreciate the info & thanks for response.

Click to expand...

Howdy,
You are missing the point.
4 wires connected between generator and panel. bonding strap removed.

Since you are still building...
Generator connection choices

Maybe your MEP-803A is not working at its peak performance.
Working MEP-803A 3kw to 16kw loading

Ok, I took apart K1 today and cleaned the insides with deoxit. It really wasn't too bad and seemed to be making good contact. There were a few minor burns spots on the contacts, but were not charred. Here are photos:


They look a lot darker in the photos than they did in person...

I will also be testing burden resistors per TM page 2-61 paragraph 2-43.2. I've ordered a newer more accurate "amp clamp" style tester that should arrive later this week, so I'll be getting these unsoldered next weekend for the test.

Does R15 or CR2 look a little charred or is that just burnt insulation from soldering?
Does it matter which end of each resistor is unsoldered?