Maintenance DC Hi-pot of 460vac motors 2

[electricpete]

EPRI NMAC 7502 recommends DC Hi-pot (step test) as a periodic maintenance test for form-wound motors. But they don't recommend it for 460-volt motors (they don't provide any explanation).

It appears to me that NEMA requires AC hipot on new motors, including 460 volt. Why wouldn't dc hi-pot be a good maintenance test for 480-volt motors. Calculate max voltage per IEEE95 (~(1.5-2)x(1.7)x460vdc).. alhtough IEEE 95 scope states it's limited to testing at 5kv and above.. with foot-note that the "methods" have been found applicable to smaller machines.

Does anyone do periodic dc hi-pot for 480-volt motors? Is there some reason this test would be less useful or perhaps more destructive for 480v motors (as compared to 4kv or 13.2kv motor), assuming IEEE95 voltage is used?

 

[rhatcher]

A few thoughts on your post.

First, terminology can be an issue as 'DC insulation resistance testing' is also referred to as 'meggar' or 'hipot' testing, with the difference being the type of instrument used based on test voltage and the numerical value of the result. Specifically, meggar scales read in megaohms and hipot scales read in microamps. Next, meggars usually operate at discrete (fixed) voltages selectable to values of up to 2.5kV for old units and 5kV for modern units. Hipots are often variable voltage and can be operated at any voltage but are not usually called for by name unless the test voltage is greater than 5kV.

My opinion with respect to routine testing may not be from the perspective you seek as I do not represent an end user for motors. However, as a service provider for those that are end users, my company routinely tests all 480VAC motors received in our shop with a surge test and a hipot test at 2000VDC. Those that fail are rewound and those that pass are reconditioned.

Finally, (if I may ask), although I have been away for a while, the last two post of yours that I have seen (there may be more I haven't seen yet) are based on questions from IEEE standards. How did you aquire access to the standards? Is it more effective to become a member or did you select specific standards to purchase individually. Of course, you may be lucky enough to have access to a company sponsored library...either way I am curious.

 

[electricpete]

Thanks for the info rhatcher. One thing you highlighted to me is that the range of voltages identified by IEEE95 needs a little massaging to work for 460v motors. If I use IEEE95 recommendation of (1.5->2)*1.7*VLL, I get a max of 2*1.7*460=~1550V. But you mentioned 2kv... and when I think about it a little more it makes sense.

I think the basis for the IEEE95 numbers is approx 60-75% of dc-equivalent of the ac factory hi-pot. ac factory hi-pot is (2*VLL+1). dc equivalent is 1.7*(2*VLL+1). If you plug in 4kv or 13.2kv, the 1kv is pretty small compared to 2*VLL, and you get approx 2*(60->75%)*1.7. If I multiply out the factor of 2 I get (1.2->1.5)*1.7... which is almost exactly what IEEE95 recommended. But that ieee95 approach of neglecting the 1kv doesn't work at all for 460v motors.
Perhaps the 460v range should be calculated as (2*0.46+1)*1.7*(0.6->0.75).. gives approx a 2->2.5kvDC range for maintenance hi-pots.

Yes, I am one of the guys who is lucky enough to have company-sponsored access to standards. We get it through IHS co. which makes the standards available in pdf format on-line. Also can get ansi, nema, ul through the same service, prepaid by the company, unlimited access. One of the few advantages of working in a big bureaucratic utility environment.

 

[rhatcher]

I am glad to help. You did a better job of justifying my answer (2000VDC) than I could have. As well, you are fortunate to have access to such a library of references and you are obviously putting them to good use...keep it up as we all benefit from your curiosity.

 

[flint]

According to the manual for the Baker surge tester the formula for testing voltages is 2 X supply voltage + 1000 X (1.2 - 1.7). We use 1.4 as a standard.
E.G. 2 x 400 + 1000 X 1.4 = 2520 (2.5Kv)

 

[electricpete]

Thanks flint. Is that for surge testing or for dc hi-pot testing.

In the case of dc hi-pot testing
(2 X supply voltage + 1000) X (1.7)
would be a factory-level hi-pot test for new windings only. I wouldn't want to apply that as a maintenance test.

By the way (separate subject), I have never understood the terminology used by Baker for their surge tests magnitude. I have heard people mention magintudes like 75% (maintenance factor) times (2*VLL+1) when describing Baker surge test magnitudes, ... but that doesn't seem to match the magnitudes ion IEEE Guide for Turn Insulation Testing (IEEE522). Perhaps Baker uses a different waveform (shorter rise time)?

 

[rhatcher]

Flint's formula is familiar as a rule of thumb standard. I have not read the manual for our Baker surge tester in a while so I will not address what is presented there. Looking to his formula though I have seen (or at least interpreted) it differently:

2 X supply voltage + 1000 X (1.2 - 1.7)

The amount 'supply voltage + 1000 ' is obviously set by the machine. The factor '(1.2-1.7)', I always take to be 1.7 as this is necessary so that the applied DC voltage equals to the peak AC value. The factor '2' is what is (by me) considered to be the variable.

Normally I would take '2' as a maximum value, although I believe that I have seen some Navy specified tests that exceed that value. Often, the actual test voltage is in reality less and is decided by the equipment owner.