Curiosity - Oddball megger readings of 4kv outdoor motor 2

[electricpete]

Motor Description; 4kv, 1250hp, 1800rpm, WP2 outdoor vertical motor, with three leads available externallly.

General Background info / Chronology:
2002-2003 - REPAIRSHOPNAME rewound this motor
2003 – 2006 – stored in warehouse
2006 – Installed into PUMP22 position.

Spring 2011 – Removed from PUMP22 due to bearing fault. Refurbished by REPAIRSHOPNAME. Post refurbishment electrical tests sat, including megger, dc hi-pot to 10KVdc, surge test.

Spring 2011 - Installed into PUMP23 position. Plant Insulation resistance test sat (6 Gig-ohms, PI = 4).

Fall 2011 – Air Filters changed

Review of trends (temperature, vibration) from Spring 2011 until Spring 2012 shows no anomalies

Approx 4/10/12 – To support pump work, motor oil drained, motor determinated, motor moved south of pit on outdoor structure (temporary storage location). Hooked up space heaters. Changed air filters.

Severe rainstorm occurred sometime between 4/10/12 and 4/12/12. This should not be a problem since motor is WP2 and normally outdoors.

4/24/12 - Performed megger. Shows short by megger test. Shows approx 0.4 ohms to ground by Fluke all three phases. Verified motor bottom was warm to touch (space heaters working).

4/24/12 approx 1700 - put cover over motor with additional lights for heat under the cover.

4/25/12 approx 0900 - Verified space heaters drawing approximate correct current - 13.2Amps at 120 volts (gives approx 1600 watts, nameplate states 1800 watts).

4/25/12 - Performed bridge tests phase to ground and phase to phase. Results as follows:

Phase to ground tests:
Lead Black Orange Red Average
Phase to ground resistance 0.306 0.304 0.305
Lead resistance 0.06 0.06 0.06
"Phase-to-ground resistance, corrected for leads" 0.246 0.244 0.245

Phase to phase tests:
Leads B-O O-R R-B Average
Phase to phase resistance 0.413 0.413 0.412
Lead resistance 0.06 0.06 0.06
"Phase-to-phase resistance, corrrected for leads" 0.353 0.353 0.352 0.353


Avg Ratio of corrected phase-to-phase and phase-to-ground = 0.353 / 0.245 = 1.439

Obviously, the motor needs repair shop attention. However, I am curious to explain what can cause this pattern. If ratio of phase-to-phase to phase-to-ground was 2.0 (it’s not), we would suspect a hard ground on the neutral. Could this represent severe uniform moisture-related degradation of insulation resistance on all three phases?


=====================================
(2B)+(2B)' ?

 

[waross]

My first check would be the space heaters. Is it possible that the wrong voltage was applied and the windings were cooked by the space heaters?
More likely you are expecting 0.353 Ohms / 2 = 0.1765 Ohms to ground but have 0.245 Ohms.
0.245 Ohms minus 0.1765 Ohms = 0.0685 Ohms. That is a fairly low resistance connection.
Thanks for sharing Pete.
Yours
Bill.


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

 

[electricpete]

More likely you are expecting 0.353 Ohms / 2 = 0.1765 Ohms to ground but have 0.245 Ohms.
0.245 Ohms minus 0.1765 Ohms = 0.0685 Ohms. That is a fairly low resistance connection.

Thanks Bill - I agree with that math, provided the 0.0685 ohms fault to ground occurs precisely at the neutral (to preserve the phase balance).

We'll have a chance to see before long.

Any other ideas?


=====================================
(2B)+(2B)' ?

 

[waross]

Waiting to see, Pete.
Yours
Bill

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

 

[electricpete]

Regarding the space heaters, the space heater current and voltage for this motor is approximately the same as we measure when installed in it's normal position, and roughly matches the nameplate (1800watts).

The space heaters in this family are sized generously. The winding temperature by RTD is 20F - 30F above ambient with the motor off.

It is the standard design for this motor. We have never seen any adverse effects from it during refurbishment inspections.

The space heaters are located a safe distance from the winding in the lower bracket (see attached).


=====================================
(2B)+(2B)' ?

 

[electricpete]

We have a spare motor, so not a big time crunch. But we're trying to decide how much effort we should expend to see if drying will correct the problem (before giving up and sending motor to repair shop). As input to that decision, we have to answer the following question:

Let's say we dried (using welding machine or similar) and got insulation resistance up around 500 Meg-ohms or more and PI satisfactory.....would you then consider the motor ready for installation (with expectation of long-term reliable operation), or would you still want to have it steam-cleaned/baked/tested in a shop to be sure there were no adverse effects ?


=====================================
(2B)+(2B)' ?

 

[electricpete]

I should add, there is a question of what extent the T-lead penetration area was covered up. The terminal box cover stayed with the cable (not the motor). IF additional temporary covering was not placed over the T-lead entry area of the motor, that would provide a possible point for rainwater entry as shown in attached photo(particularly since there are 3 unused T-lead penetrations between the used T-lead penetrations). I don't yet have all the facts regarding whether this area was temporarily sealed or not.

=====================================
(2B)+(2B)' ?

 

[waross]

Been there done that Pete.
After the hurricane, with three 350 KW and two 600 KW diesel generators filled with brackish water we decided to try to clean and dry one 350 KW and then decide how to proceed with the others.
The set we chose was so bad that we couldn't megger it. Insulation resistance was measured with a multimeter and was very low. The reading changed when we reversed the meter leads. Went to low volts and found that the windings were so contaminated that the copper, iron and the brackish water were forming a primitive battery and we could read a part of a volt between the windings and the frame.
I had the boys working on it for about a week. We dismantled the alternator and used an air powered parts washer with fresh rain water to flush out the rotor and stator. Then we reassembled the set and ran the engine for several days. First with the leads shorted to air dry it and then for a day or so with a battery charger on the field to force about 50% of FLA through the shorted windings.
Sometime before the stator had been meggered at about 3 meg-Ohm or 4 meg-Ohm. We got it up to 7 meg-Ohms.
That was the first machine back on-line after the hurricane.
BUT we shipped the other alternators up to the USA for cleaning, dipping and baking in a motor shop.
I warned the operators that once the original machine was allowed to cool down, it must be sent up to the USA before being used again.
What do I know?
We had the 600 KW sets repaired stateside and returned to service. Someone decided to restart the 350 KW after it had sat for some time and had not yet been properly repaired and that was the last day of it's life.
My advice:
I suspect that the insulation may be permanently damaged and even though you may get the insulation value up, it may (probably will) drop when the machine cools off.
I would get the machine to the shop as soon as possible before the moisture does serious corrosion damage.
I would proceed with onsite drying to prevent corrosion damage, but send the motor to the shop ASAP. A good megger reading on a warm motor may not be as good an indication as you think. A little humidity and it may fail.
But I understand that you may be in the position that I often find myself in where someone with less knowledge but more authority chooses to waste money by going cheap at the wrong time.
Yours
Bill

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

 

[electricpete]

Thanks Bill, that's a good data point to keep in mind as we make our decision.

You're also right, the decision in this particular case is not completely up to me. The spare motor is being installed and since the removed motor is not critical to production at the moment the decision about what to do with it is proceeding painfully slowly. We have a meeting scheduled to decide what to do about this removed motor next Tuesday. (Gotta love those meetings).

=====================================
(2B)+(2B)' ?

 

[electricpete]

Update.

We placed the motor indoors and kept the space heaters energized and extra external heat for almost a month - no change in readings.

Sent the motor to a repair shop. Upon disassembly, they determined there is mechanical contact between the upper bracket and a winding jumpers on the top/connection end of the winding. Some signs of abrasion. Apparently it was not a great rewind job previously and over time it wore through. Further inspection is in progress and I have only phone report so far.

Why it failed exactly this time, I can only guess that lifting the motor with a crane may have created stress that pushed it over the edge to final failure.

For curiosity, I'll see if I can find out if this failure was near the neutral or not. Will post pictures if I get any.

=====================================
(2B)+(2B)' ?

 

[waross]

Thank you for the update Pete.
Yours
Bill

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

 

[Skogsgurra]

Loved this diakogue. And learned from it.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

Attached are photos taken at the shop.
Slide 1- connection end (top) - phase jumper damaged by contact with metallic baffle
Slide 2- opposite-connection end (bottom) - coil knuckle damaged by contact with metallic baffle
Slide 3 - Photo of rub marks on baffle, including insulation debris.

The sequence of events at the shop:
Received motor and verified insulation resistance indicated shorted
Disassembled. (no inspection or testing performed at this time).
Put into the bake oven
Tested/Inspected after removal from bake oven. Inspection results: as shown in attached. Test results: several gigaohms insulation resistance. (the damaged insulation is not near any ground plane).

The theory I posted a few days ago was that the water had nothing to do with it - there was mechanical contact where the baffle contacted the coil. However the shop says inspection of the defect does not show conductor, so it should not have been a dead short unless the insulation was damaged.

The shop believes that this coil damage allowed moisture to enter the coil. They base their conclusion in part on their belief that the insulation in the coil interior has a dry powdery texture, not what is expected from VPI. So presumably it would easily soak up moisture.

(If we had checked insulation resistance after motor was disassembled but before it was dried, that may have shed some light on these possibilities, but that opportunity was missed)

I would presume some damage was present from the moment the motor was reassembled after last rewind in 2002-2003 (that is when the baffles were forced into the coils because the coils extensions and endwinding of the 2002/2003 rewind were not dimensioned properly to avoid contact).

It raises some questions: why didn't we see this damage when the motor was disassembled/reassembled for bearing replacement in Spring 2011 (?).

Perhaps the damage was worsened when re-assembled in Spring 2011.

Also note all the electrical testing that was done and showed no problem:
In 2002/2003 after rewind:ac hi-pot after rewind in 2002-2003 (at shop)
In Spring 2011 after bearing replacement: DC hi-pot to 10kv and surge test (at shop)
In Spring 2011 after bearing replacement and installation into plant: megger (at plant) - passed with flying colors.

Clearly there was quite a bit of mechanical damage at least in the last two sets of tests.
Perhaps that's one datapoint to suggest we shouldn't place to much faith in our elecrtrical tests.
Also our specification requires hi-pot but I don't believe we specifically identify that they must be performed AFTER final re-assembly. Considering that this damage presumably occurred during reassembly (as well as the fact that endwinding has closer ground plane after re-assembly), I think it's probably worthwhile that specifications should specifically require that hi-pots are performed AFTER final re-assembly.

=====================================
(2B)+(2B)' ?

 

[waross]

Thanks Pete.
As one of the active Gurus here, you know well how much we appreciate this type of feed back.
Yours
Bill

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