804B Odd Problem

[Larry Olson]

Or maybe not. I just can't seem to find any useful information. I have purchased an 804B as a back up power source as I'm about 10 miles from the Gulf Of America in hurricane prone area.

I assumed the genset was operating correctly as the volts and frequency were correct and the engine starts up first time every time. The problem didn't raise it's ugly head until I put a load on it. My buddy gave me an old 20K air handler that I converted to a load bank that I could use 5K at a time. After watching CallMeColt's video on you tube saying that the only way to really test and MEP generator is to put under a load for 6 hours. I set out to do just that. All volts to N were around 125VAC so I kicked the first 5K on, no problem. Kicked the second 5K on and noticed my voltage had increased a little bit on the panel gauge, put the next 5K on and it went up even more. At this point I was at 110% of output for the 2 phases I was connected to (L1 and L3). I grabbed my multimeter and looked at the voltages under load to N and I had L1-125V, L2-143V, and L3-142V. And also 232V across L1 and L3. Took the load off and all 3 phases to N were back to 125VAC.

Anyone have any idea's what's going on? I went through the regulator troubleshooting steps looking at voltage and the stator ohms. Only thing I found out of spec's was the volt potentiometer specs. Called for 0-20,000 ohms and I had 0-18,000 as I had a TRC regulator. I spent $300 on a green mountain regulator that I probably didn't need and tried that, same thing. Not sure what to look at next, any help would be appreciated.

 

[Scoobyshep]

How do you have this wired? Is it a 3p load?

 

[Larry Olson]

How do you have this wired? Is it a 3p load?

Single Phase L1-L3. No load on L2.

 

[Scoobyshep]

Single Phase L1-L3. No load on L2.

On the output terminal board do you have the bond between L0 and ground installed?

 

[Larry Olson]

On the output terminal board do you have the bond between L0 and ground installed?

Yes

 

[Scoobyshep]

using the neutral at all? What are the amp readings per leg? I cant speak from memory where this set reads voltage for the AVR, it may be sampling a heavily loaded leg and artificially boosting voltage. Youd think they would be close but heavy load vs no load does weird things. (btw 2 legs heavy loaded and 1 unloaded is a unwise idea)

 

[Larry Olson]

I didn't look at the amp readings, I would imagine around 35A across L1 and L3. I am aware of the affects of running these imbalanced. I am currently looking for the info on how to test the stator coils as that is my next step I presume. Thanks for any help.

 

[Scoobyshep]

I didn't look at the amp readings, I would imagine around 35A across L1 and L3. I am aware of the affects of running these imbalanced. I am currently looking for the info on how to test the stator coils as that is my next step I presume. Thanks for any help.

I would get amp readings


Stator testing is easy. you need a megger (be careful it puts out up to 1000v).

I prefer putting out the highest test voltage, down side here is the other things connected so best practice is head only Meaning you need to unwire all 10 leads from the head.

megger basics: theres 2 flavors these days hand crank (old units) and digital (new)

hand cranks: less accurate, but will put voltage out regardless of load as long as you crank

digitals: high accuracy BUT in very low resistance tests will retard the output voltage to protect the meter. so lead to lead tests can show good when they are bad. look in the meters manual and there should be a minimum resistance. adding a resistor of that value in series allows the meter to be used in a winding test application, just subtract that resistor value from the reading.

in a meatball heat test, i prefer hand cranks(we had a manager try to throw my crank megger out once, the crack of a wooden mallet was enough of a deterrent to curb that behavior)


Meg each lead to ground they should all be high.

Then the following point to point: 1-4 2-5 3-6 7-10 8-10 9-10 these should all be low and close to the same value.

now im an overachiever and since its this far apart i test every lead to every other lead. with the exception of the pairs above they should all have a high resistance to each other.

 

[kloppk]

Testing of the Stator windings and more is covered in TM 9-6115-643-24.
Just search for "resistance" in the PDF.

 

[Larry Olson]

Testing of the Stator windings and more is covered in TM 9-6115-643-24.
Just search for "resistance" in the PDF.

I just found the tests for the generator head in the -24 manual. Paragraphs 4-18 through 4-24. I'm also going to look at the diode pack to see if anything is abnormal there. Anyone have any experience with diode failures? Thanks for all the help, I'll get to the bottom of this yet!!! Going to have to borrow a megger from someone tho.......

 

[Scoobyshep]

Diodes in the head are more for the rotor. they fail one of 2 ways. Open: you can loose a couple and never know, this usually leads to more failures because the remaining diodes have to work harder. Shorted: reallly low output voltage (if any).

Understanding why they are there: Spin a magnet in a coil you get power. the strength of the magnet changes the voltage. so the magnet is replaced with an electromagnet (long wire wrapped around an iron bar) now you can change the magnet strength. old sets (and cheap consumer sets) use brushes to pass the magnet power (excitation) to the rotor. problem is they wear out, so they made them brushless. so now in the stator theres a winding to excite the rotor. that unfortunately makes AC and thats not good for an electromagnet. enter diodes that makes the rotor AC into DC the dc makes a magnet and excites the main part of the sator.

 

[Larry Olson]

Thank you for the explanations. Makes sense. Trying to borrow a megger on FB.

 

[Ray70]

Before going too deep into the generator, would it be worth throwing a secondary 120V load on L2 just to verify you get the same result with all 3 legs loaded somewhat equally? If you get the same issue, then dig deeper??
Hate to see you go through a bunch of testing only to find out the load imbalance was goofing up the regulation circuit.

 

[kloppk]

If in the end you find you need a stator I know someone looking to part with one.

 

[Guyfang]

I think its the load imbalance. Its a 15KW. You overloaded it on two phases. 110% if I remember correctly. lets see what @LuckeyD has to say. He has probably seen this situation down range a number of times.

 

[LuckeyD]

Today I hurt all over with the arthritis. If I sound gruff, it is the old me coming thru when I ran several thousand Soldiers around the planet.
WOW. Lets think just a moment. I'm doing the scratchy head trick. Issue shows at 100% load testing on 2 phases, (imbalanced load), and when load removed it acts normal again. Wow. Speculation; Shows with only this load: Hmmm, sounds like a feed back from that load, oh by the way.
Step1: Get information on your gen from the TM section. Read, store info, use it later, especially on how to fix it. That 6 hour load test is a bit overkill. So is that air handler, and a load test is for all phases, not part of the gen. You step up to about 80-90% of full load, test for about 2 hours and step it back, and cool down the main gen at no load and then shut down. This is OEM Not me. 100% and more is a NO GO. Unbalanced is also a NOGO. Sorry, but you do want to keep that Gen I believe.
Step 2: Get 3 each, space heaters. Same size. They sell these and you can find some cheap. Same size. 3ooo watts each works well, but 4000 is better. Connect each 120V heater from each phase to neutral. All loaded the same for each phase and this is mostly a resistive load. Try that. Before you tear it all apart.
3. I am in the middle of the attached. It ain't soup yet. Look it over. It even helps on testing the AC Machine. I'll ask Guy what a Head is because it was either something I scratched a lot or something on board ship i won't go into, it reminds me of a Porta John in any of the war zones.. Your engine seems strong, so
lets look over the components that make up your regulatory system. Check each to the VR info attached. This way you have either a TRC pure regulator system or a green Libby. Something there is not right. You only have 1 excitor stator the VR controls. After testing with all phases and you have the strange results as you started, I would suspect the rotor diode ring or rotating exciter. Please look to see if you have mixed things in the regulator system. Let us know

I can go into how to on anything on that gen. First try loading it right.

 

[Larry Olson]

Before going too deep into the generator, would it be worth throwing a secondary 120V load on L2 just to verify you get the same result with all 3 legs loaded somewhat equally? If you get the same issue, then dig deeper??
Hate to see you go through a bunch of testing only to find out the load imbalance was goofing up the regulation circuit.

I did that with a couple of heat guns and still had high voltage. It wasn't quite an equal load but it didn't seem to matter. I will try an equal 3 phase load next time I run it.

 

[Larry Olson]

Today I hurt all over with the arthritis. If I sound gruff, it is the old me coming thru when I ran several thousand Soldiers around the planet.
WOW. Lets think just a moment. I'm doing the scratchy head trick. Issue shows at 100% load testing on 2 phases, (imbalanced load), and when load removed it acts normal again. Wow. Speculation; Shows with only this load: Hmmm, sounds like a feed back from that load, oh by the way.
Step1: Get information on your gen from the TM section. Read, store info, use it later, especially on how to fix it. That 6 hour load test is a bit overkill. So is that air handler, and a load test is for all phases, not part of the gen. You step up to about 80-90% of full load, test for about 2 hours and step it back, and cool down the main gen at no load and then shut down. This is OEM Not me. 100% and more is a NO GO. Unbalanced is also a NOGO. Sorry, but you do want to keep that Gen I believe.
Step 2: Get 3 each, space heaters. Same size. They sell these and you can find some cheap. Same size. 3ooo watts each works well, but 4000 is better. Connect each 120V heater from each phase to neutral. All loaded the same for each phase and this is mostly a resistive load. Try that. Before you tear it all apart.
3. I am in the middle of the attached. It ain't soup yet. Look it over. It even helps on testing the AC Machine. I'll ask Guy what a Head is because it was either something I scratched a lot or something on board ship i won't go into, it reminds me of a Porta John in any of the war zones.. Your engine seems strong, so
lets look over the components that make up your regulatory system. Check each to the VR info attached. This way you have either a TRC pure regulator system or a green Libby. Something there is not right. You only have 1 excitor stator the VR controls. After testing with all phases and you have the strange results as you started, I would suspect the rotor diode ring or rotating exciter. Please look to see if you have mixed things in the regulator system. Let us know

I can go into how to on anything on that gen. First try loading it right.

I have already liberated the ends of the stator wiring in anticipation of testing each of them. I was also planning to look at the diodes and the rotor however I will hold off that. If the coils are OK than I will re-assemble it and load test it correctly. I'm pretty sure it will still act up.

 

[LuckeyD]

Simple. Please check that the R1, R16, T1 and regulator all belong together. One does not just change a Voltage regulator on these gens. Oh No, they had to make things hard for everyone. Then give it a go. Oh, while you are at it, use a mirror behind the load terminals. See if all is cool there.

 

[Larry Olson]

Simple. Please check that the R1, R16, T1 and regulator all belong together. One does not just change a Voltage regulator on these gens. Oh No, they had to make things hard for everyone. Then give it a go. Oh, while you are at it, use a mirror behind the load terminals. See if all is cool there.

The regulator I have installed now doesn't use R1,R16 or T1. I am however thinking about re-installing the original TRC regulator.