IEE Std 522 1

[petronila]

Dear all

I was taking a look of the IEEE Std-522 and in this document the test sequence for a new pre formed winding is showed at pag.7 the summary is:

1. Test all coils prior insertion 2. Test all coils after assembly into the stator with out grouping and with out insulate nothing 3. Once all groups are connected before insulating series connections and before connect the groups to create the phases
4. After all connections are made and insulation applied.

The question is : For a 6 Kv Motor wich are the Voltages levels to use in each numbers of the above procedure.

Can someone indicate a best procedure?

Thanks and regrads

Carlos

 

[RKeener]

The purpose of the IEEE 522 standard is to ensure that motors can withstand high voltage transients. Transients levels often are several times greater than line voltage. The test method which best simulates real world high voltage transients is the fast rise time impulse test (0.1 - 0.2 usec rise time) This type of test equipment is manufactured by Baker SKF and PJ.

As the standard says, the impulse test voltage should be agreed upon by the manufacturer and the end user depending on the anticipated level of transients. Most often, the 3.5 pu test voltage is used for all stages of testing form wound coils and assembled motors. To calculate the 3.5 pu test voltage, multiply the line voltage by 2.86. So a new 6 kV winding should be tested with about a 17 kV peak impulse. This standard states that the maintenance impulse test voltage should be 75% of new or about 13 kV peak.

The IEC 34-15 equivalent standard recommends an even higher test voltage (4V +5 kV). It recommends a test voltage of 29 kV peak on an assembled varnished 6kV motor.

This proof test voltage may seem excessive, but, it is necessary to ensure that a motor will be reliable. Due to the high distributed capacitance of a large motor, a voltage transient stresses primarily the coils close to the motor leads. These coils must be able to withstand an occasional very high turn to turn transient stress.

I have worked around a lot of motor manufacturing plants and find that reputable manufacturers can supply coils which can withstand the IEC recommended test voltages.

 

[petronila]

Hello RKeener

Good information.

Thanks

Carlos

 

[electricpete]

For sure, the customer spec needs to be considered (some customers consider the in-process check as a quality control check for final product, since the individual coils deep in winding can't be tested during the final test when connected/vpi'd).

BUT, the customer and the shop also need to understand the implications.

You should not be using full final test voltage on green coils

6.7 Reduction in test levels for uncured coils

To avoid overstressing unimpregnated (dry taped) coils, the test voltage may be reduced to some fraction
(generally 60–80%) of the value specified in 6.2 (or 6.3), as agreed between the manufacturer and the user.

To avoid overstressing uncured resin rich coils, the test voltage may be reduced to some fraction (generally 40–60%) of the value specified in 6.2 (or 6.3), as agreed between the manufacturer and the user. The test value in this case should not exceed the DC hi-potential test value.

Further, “Effect of Surge Testing on Unimpregnated Ground Insulation of VPI Stator Coils” by Stranges, Snopek, Younsi (*), Dymond states:

IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 38, NO. 5, SEPTEMBER/OCTOBER 2002

The authors recommend using an applied test voltage of 50% of the value given in IEEE 522 (1992) for surge testing of green coils in the stator. Customers and manufacturers should be wary of any increase in this value because of possible damaging effects on the pre-impregnated ground insulation.

* Dr. Younsi occassionaly visits our forum. Hopefully he will visit this thread.


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

 

[electricpete]

The article was written in 2002. The standard I have is 2004. I think the 40-60% in the 2004 standard is a result of incorporating the recommendations of these authors.

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

 

[electricpete]

Yes, that is confirmed. My 1999 version of IEEE5222 states:

6.6. For unimpregnated or green coils, the test voltage may be 60–80% of the value specified in 6.2. or 6.5., as agreed between the manufacturer and the user.

It is reasonable to conclude the current lower value 40-60% is a result of articles similar to the one linked above.

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