MEP803a 240V to Ground

[aleigh]

I have a MEP-803A and for various reasons need to produce "european" style 240V, in my case, 240V single phase, to ground. I have reviewed the TMs, etc, and my understanding is that if I put the MEP in 120/240 single-phase mode, remove the neutral bond on N if it is there, and then bond L3 to ground through approved means, I should result in 240V across L1 and L3, 240V to ground, and neutral will become something strange. I do not need 120V, so I would just leave the MEP neutral disconnected and L3 would become my new neutral resulting in a 2-wire system.

I am posting just for 2nd opinions and in case people with more experience with these generators know a reason this would be a bad idea for the generator. Since it's strange, it's not really covered in the manuals.

 

[Jeepadict]

@Guyfang this sounds like your wheelhouse

Sent from my SM-G970U using Tapatalk

 

[Scoobyshep]

Easy peasy, leave the bond alone. Wire between l1 and l3, the load will see 240 even though its 2 legs of 120.

Sent from my SM-G973U using Tapatalk

 

[aleigh]

That results in an ungrounded system - I'll produce 240V, but in reference to nothing. I require 240V to ground, hot/neutral, 2-wire.

Edit: Or rather, I'll produce -120 to +120 if N is bonded to ground. EIther way, I wind up with hot/hot not hot/neutral.

 

[Scoobyshep]

You will be using l3 as your "neutral" (neutral as the load sees it) your ground is only there for safety (it shouldn't see current unless there is a problem) the only difference is your faked neutral is not coupled to chassis.


The only other thing you can do (without a transformer) would be to make a modified grounded system (similar to corner grounded delta).

Sent from my SM-G973U using Tapatalk

 

[aleigh]

It matters in this application, because I am switching with another service which is 240V to ground, bonded, hot/neutral. When I transfer I don't want phase to ground potential to change. Hence my desire to just bond L3 to ground.

 

[Scoobyshep]

It matters in this application, because I am switching with another service which is 240V to ground, bonded, hot/neutral. When I transfer I don't want phase to ground potential to change. Hence my desire to just bond L3 to ground.

Then its a transformer or corner grounded.

Disclaimer: I have not personally performed this on this model of generator, I have done this on transformers and 1 commercial set. Attempt this at your own risk and use extreme care.

If the reference to ground is completely isolated, (you need to be 1000% sure there isnt another in the head/set) and bond L3 to the ground. Done correctly it will create a single phase 240 to ground source.



Sent from my SM-G973U using Tapatalk

 

[aleigh]

These concerns are exactly why I am asking.

 

[Scoobyshep]

These concerns are exactly why I am asking.

If it were me, I would start by breaking the neutral ground bond and checking the resistance between L0 and chassis. It should show very high resistance. Then I would ground L3 to chassis through a fuse (5 amp maybe, intent there is to establish L3 as neutral, fuse is there for testing that way if there is another path to L0 the fuse will prevent a bigger fireball)

Again at your own risk


The other option is a 1 to 1 transformer (240 in 240 out ) then establish the ground there.

Sent from my SM-G973U using Tapatalk

 

[aleigh]

Thanks, that's helpful. Isolation transformer of this size is costly, obviously, hence my desire to avoid it unless it's necessary. If it is required, then that is what has to happen, but I don't want to put one in for no reason.

 

[Guyfang]

Scoobyshep, I could not have explained it better.

Aleigh, I would go with a transformer. We had this same problem, but in reverse. Our Hawk Missile sites used 416 volts, 400 hertz electricity. For the big loads, radar system, we had frequency converters. For the small "stuff", read that single phase, we had transformers. It, (the freq converters) still didnt work right. But thats another story.

 

[aleigh]

240/240 15kva isolation transformer is ~$1300. And serves no purpose at all if the L1/L2/L3/N studs are fully isolated on the MEP803a. But are they? That's the mystery.

 

[Ray70]

I could very well be missing something here, but I'm not sure connecting L3 to chassis ground will be a good idea. I would imagine it will cause problems with all the DC negative controls and gages, not to mention making the entire machine HOT to ground / neutral by 120V??
Am I misunderstanding something??

 

[Scoobyshep]

I could very well be missing something here, but I'm not sure connecting L3 to chassis ground will be a good idea. I would imagine it will cause problems with all the DC negative controls and gages, not to mention making the entire machine HOT to ground / neutral by 120V??
Am I misunderstanding something??

It seems that way, but what that does is establish l3 as the ground reference point (same how L0 is the normal reference point). Trick is not having another refrence point. Look up a corner grounded delta transformer, That will help shed some light on the subject.

 

[aleigh]

I could very well be missing something here, but I'm not sure connecting L3 to chassis ground will be a good idea. I would imagine it will cause problems with all the DC negative controls and gages, not to mention making the entire machine HOT to ground / neutral by 120V??
Am I misunderstanding something??

It is as Scoobyshep says, about the bond. Remember neutral was a hot until you bonded it to something. That is what makes it neutral, being bonded. It's arbitrary what you bond. In terms of the onboard electrics, my understanding is that they run off the battery / DC alternator, so the control system electrics are isolated from the gen head outputs. Also just to be pedantic, grounding the chassis and grounding L0 or L3 is not exactly the same, but for the sake of your question only, it is a distinction without a difference, assuming the chassis has a valid ground rod or alternative arrangement. I actually would establish the neutral bond elsewhere in my application, not at the generator load board.

But of course as Scoobyshep has cautioned about and I am cautious about, if these assumptions about the onboard or the isolation on the head outputs are not true, there will be bad problems.

 

[aleigh]

So, the breakers seem to be rated for straight 240V (not 120/240V split phase), which is good. The part that is bothering me now is the voltage regulator (AVR). I'm concerned there will be problems if it depends on measuring phase to ground during operation. I notice it has 120V sense leads on it, but those could be disconnected. What I have not been successful at so far is finding a schematic for its internals in the TMs. Another thing that crops up is that in 1-phase mode, the AM-VM transfer switch is going to read across L3/L0 for voltage, which would probably peg (since 0-240V not -120 - +120V), although it would read-right at the 3-phase L1/L3 position.

As the uncertainties mount an isolation transformer to create a separately derived system is looking better and better.

 

[Guyfang]

So, the breakers seem to be rated for straight 240V (not 120/240V split phase), which is good. The part that is bothering me now is the voltage regulator (AVR). I'm concerned there will be problems if it depends on measuring phase to ground during operation. I notice it has 120V sense leads on it, but those could be disconnected. What I have not been successful at so far is finding a schematic for its internals in the TMs. Another thing that crops up is that in 1-phase mode, the AM-VM transfer switch is going to read across L3/L0 for voltage, which would probably peg (since 0-240V not -120 - +120V), although it would read-right at the 3-phase L1/L3 position.

As the uncertainties mount an isolation transformer to create a separately derived system is looking better and better.

The A1 is NOT a part to be repaired. It is not to be opened,or repaired at any level of maintenance in the military. It is a "Pluck and Chuck part". So you are never going to find a schematic for it in an Army TM.

Open to see my comments.

 

[aleigh]

So, this is what I think I am going to do when I am at the site next. I'll remove the neutral bonding connector on/behind the load board and check for continuity between N/L1/L2/L3 and ground. Finding none and with the generator floated, I'll run it and see how the voltages come out to ground. If Neutral floats (there is some potential from neutral to ground), then it seems certain the head is isolated. And then I'll see how I feel about moving up to Scoobyshep's fuse test.

It occurs to me that if the generator actually cared which leg had the ground reference, then an isolated generator on an ungrounded system which had a phase to ground fault would have some dire problem, but I've never heard that reported. Just the opposite, the conventional wisdom first ground fault in an ungrounded system goes undiscovered until the second one occurs.

 

[Scoobyshep]

You will have continuity from L1 2 or 3 to L0. Big one is L0 to ground

Sent from my SM-G973U using Tapatalk

 

[aleigh]

Thanks for the clarification. L1/L2/L3/N to Ground is what I meant to write. If I pull the bonding, I would expect them all to be discontinuous relative to ground, but not each other.