Core hot-spots - repair or reinsulate/restack [URGENT] 2

[electricpete]

Details on the motor in question
4,000 HP, 1200 rpm 13.2kv vertical squirrel cage induction motor.
stator Core weight 4300 pounds

Motor was built with C3 core insulation (organic).

Original windings were B-stage, very loose in slots.

Motor was rewound in 1998. No burnout was required. Core loss watts per pound was 1.5.

The rewound insulation system had some problems (not related to core), so we have elected to rewind again.

Now in 2006 we find core loss at 2.5 watts per pound before burnout and 3.4 watts per pound after winding burnout.

The core has circumferential bands of hotspots mostly in the axial center.
Pre-burnout hot-spots:
20F above ambient at 10 minutes; 25F above ambient at 20 minutes 35F above ambient at 30 minues. Core average temp rose to approx 12F above ambient at 30 minutes.

Post-burnout hotspots: 55F above ambient at 20 minutes. Core average temp was 14F above ambient at that time.

If the hot-spots could be successfully repaired with "cracking" (mechanical agitation), would that be sufficient to assure reliable operation of the motor. Or is a core restack with reinsulated or new laminations required?

Additional info: This motor and it's 14 total sister motors run 180F max winding temperature year-round. At least 4 of the sister motors have had similar conditions noted during previous rewinds which were resolved/retested by cracking type repairs between 1998 and 2004. None of those showed any increase in winding temperature. We periodically check all 6 stator RTD's in each motor to verify the hottest one is monitored by plant computer.

PS - this is urgent only because we are trying to make a decision within the next few hours. If you have any input, please let me know now rather than later. Thanks.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

By the way, two burnouts occurred between the before/after burnout results above. First burnout was at 650F max and second burntout was at 675F max.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

These motors were orgiinally built around 1982. And have been in service since 1986-1989 with some repair periods (only one previous repair period as mentioned in 1998 for this particular motor).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

I use 10 deg C (50 deg F) limit between the hot spot and average temperature during a 60 minute test. Core separation will work if the no. of hot spots not very high. If large areas of hot spots are present, I revarnish or replace the core.

Since the burn-off process invariably degrades the core varnish, I use manual method of removing the coils. This is a time consuming work in VPI windings but at least the core is saved.

Relying on winding RTD's to detect core hot spots is not enough since the RTD's may never be near the hot spots. Winding RTD's are for winding only.

 

[electricpete]

Thanks kumar. If the core can be repaired to get under 10C rise without restack, is that good enough?

The main pattern of the hotspot is a circumferential band around the axial center of the generator. It seems to me that should go by the location of the RTD's. Also the winding (rtd) should be hotter than the core since the winding transfers its heat to the core, right?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

pete

10 deg C between hot spot and average is fine.

Hot spots tend to have very localized heating. Unless, you have the winding RTD's right at the hot spots, you will not see major differences in the winding temperatures.

Full flux loop test (I don't trust ELCID) on empty core is the only way to get a complete thermal map of the core.

A circumferential band at the axial centre indicates a rotor rub. Do you see any such rub marks on the rotor too ?

 

[aolalde]

Hi Electricpete.

I am sorry that my comments are late for your decision.

It is hard to ensure satisfactory results of a core repair. Test of the core loss (Watt/Lb) and hot spots temperature rise at 110% operating back iron flux density is the way to ensure satisfactory results.
Note that most of the core tests are made at 85,000 L/sqin, you need actual winding data to calculate the specific core density for the voltage and winding of your machine, then check the performance as if 10% over-voltage were on your line.
As you stated, "cracking" the laminations could clear the shorts. After that we test and if the result is satisfactory, we impregnate under VPI the bare core stack to fill the interlaminar air spaces with resin (oven cured) to keep a steady condition of the stator core. Then we retest before the winding insertion.