Strange Mode of Generator Stator Failure?

[JohnMcNutt]

This could be a long explanation so I will try to keep it simple. 10 kWe genset. Single phase 120/240 output with auxiliary winding for AVR. Brush type field.

Unit intermittently produced power until it didn't. Potted AVR assembly found burned. Tried several spare universal (line powered) AVRs laying around. About 2 VAC output.

Rotor ohms about 20, good. Fixed excitation amp draw as expected. Applying 12vdc to field gives specified voltage across all windings when unloaded. (125VAC across L1-L2, roughly half voltage.) Here is where it gets strange. Connecting an AVR across the line will bring the line voltage down to 24VAC or so in this condition.

Installed another AVR specifically made for this unit. Now we are getting somewhere. AVR is supplying 225 VDC at 10.8 amps to field. Spec is 60VDC max. Output voltage is now about 18 volts AC across both lines. Neglected to measure auxiliary winding voltage but suspect it is very high because exciter output voltage is able to reach that power and current level.

The engine does not grunt or load down during any of these tests. Stator ohms are as specified.

It seems to me like the collapsing voltage might be due to a high resistance in the stator power windings but the ohm reading does not bear that out. Also, the low voltage seems evenly divided across the power windings, so one particular high resistance spot seems to be unlikely.

About 1/3 of the visible winding seem a different color that is a dark gray-green opposed to the normal reddish-gold but I can't tell if that is from heating or just a different color varnish.

Ideas?

 

[pwrengrds]

My thought is there is something wrong with the field. The 20 ohms seems high for field resistance. The field voltage and amp draw agrees with your ohms measurement. The field voltage is higher than expected for a small generator. From the information it seems the running field voltage should about 25-40 volts and 10 - 15 amps which makes the expected field resistance something less than 5 ohms. It takes 11 amps field current to get the stator voltage to nominal but that should be less than 40 volts not 225 volts.

 

[waross]

Check the speed of the unit.
Check the multimeter that you are using.
Check that the slip rings are not glazed and producing high resistance.
Check the bearings for play. I have seen small brush type sets with sloppy bearings that caused the brushes to bounce with strange results.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[JohnMcNutt]

Field resistance spec is 17.2 ohms. It seems typical for this style machine. Field current is smooth and constant with the machine running, not jumping around as I would expect with a brush or other continuity problem. Brushes and slip rings are visually good. As has been pointed out, field current and voltage agree with field resistance when supplied while running both 12vdc and regulator voltage which happens to be 225 vdc. I surmise from the regulator going full field trying to get the voltage up.

From these tests I did not believe there was really any possible rotor problem.

 

[waross]

Installed another AVR specifically made for this unit. Now we are getting somewhere. AVR is supplying 225 VDC at 10.8 amps to field. Spec is 60VDC max. Output voltage is now about 18 volts AC across both lines. Neglected to measure auxiliary winding voltage but suspect it is very high because exciter output voltage is able to reach that power and current level.

Something wrong.
Possibly a meter reading error.
Possibly a bad meter.
Probably a bad AVR. It is my experience that AVRs "laying around" are often laying around because they have problems.
10.8 Amps at 225 Volts is over 2400 Watts. That is excessive for a 10 KVA generator. If the readings are correct, the rotor will soon be emulating toaster.
50% voltage at 12 Volts applied is often a good reading.
I suspect a bunch of failed AVRs and metering issues.
Another possibility is a hidden break in the AVR wiring to the brushes, but that would not explain a valid 10.8 Amp reading.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[IBRCAN]

Waross-gen output voltage is only 18 VAC with 225 VDC/10.8A field, so I think there's something more than an AVR issue.

Could the field be grounded? All of that power must be going somewhere. Do you have access to a megger tester?

John, have you verified with a known good meter? When I am troubleshooting, I tend not to even look at any meter but my Fluke.

 

[ScottyUK]

Can you arrange an open-loop test with a suitable load - eg load bank or an array of tungsten lights - then bring the field up using a variable DC supply? You should be able to achieve rated output voltage from the set with load applied.

What is the machine configuration? Slip rings to directly excited main field?

 

[JohnMcNutt]

Meter readings taken in multiple times and multiple ways with my trusty Fluke 115 and 337 which I use every day and have used before and since.

I agree that the AVR is putting out 2400 watts. That is probably why the first one fried, because it was supplying so much power until it went poof. Thankfully that happened before the rotor went poof I guess.

These readings were taken with a spare AVR and we did not leave it run long enough to burn itself up also. But with 2400 watts going to the rotor and no output from the main power leads, something is amiss. However, we also know that the auxiliary winding is capable of supplying the 2400+ watts necessary to power the AVR.

 

[waross]

How about the sense wiring? That is the wiring from the main field to the sense terminals of the AVR. An open or high resistance there would give wide open excitation.
As Scotty says. directly excite the field and check the output voltage. Any load is better than no load. If you don't have a variable DC supply handy, use two or three 12 Volt batteries. A reverse diode across the field to handle the discharge is a good idea.
You may have some insulation breaking down as the output voltage rises.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[JohnMcNutt]

Wide open excitation is definitely occurring but the sense wiring is internally connected in parallel with the output wiring and the output is only 17 volts which is why the excitation is going so high.

 

[ScottyUK]

Sense wiring connected in parallel sounds odd. Normally the auxiliary winding is independent of the main output. A sketch of how the windings and interconnecting wiring is connected at present would be helpful.

Have you checked the output under load with manually controlled excitation? What do the voltages on the main and auxiliary windings do, and what is the excitation voltage and current? If you cross-reference your measurments to the sketch that would help us understand what you are seeing.