Hi Potential Test on motor

[philly]

I am wanting to perfrom a step voltage test of a 5kV motor using a Hi-Pot machine. I will step voltages between 1000-10000V DC.

How would I connect the 3-leads of the Hi-Pot equipment to the motor?

I am thinking that I would hook the ground lead to the known system ground. I would then hook the two test leads, one to the motor lead under test, and the other to the motor frame? Is this correct?

 

[bth0mas20]

I believe that you should be connecting the lead that applys the voltage to the motor lead on any phase (because the windings are connected internal to the motor). Then you have a load return wire that is connected to the motor frame. Then the other is probally a ground that should be connected to the plant grounding system for discharging after the test is complete.

 

[zog]

I am wanting to perfrom a step voltage test of a 5kV motor using a Hi-Pot machine. I will step voltages between 1000-10000V DC.

How would I connect the 3-leads of the Hi-Pot equipment to the motor?

I am thinking that I would hook the ground lead to the known system ground. I would then hook the two test leads, one to the motor lead under test, and the other to the motor frame? Is this correct?

Hipot test can be very destructive on motors, what type of motor is this? What were your IR readings? What was the PI? (These need to be done before to ensure a hipot test will not damage your insulation).

Usually a DC hipot test is not done on motors, a Pf or DF overpotential test is done per IEEE/ANSI std 95. Where did you get your test voltages from? What values do you expect to see?

I highly recommend you get a certified testing company to do these tests for you, not knowing how to hook up the test equipment is a big red flag.

 

[philly]

Hipot test can be very destructive on motors, what type of motor is this? What were your IR readings? What was the PI? (These need to be done before to ensure a hipot test will not damage your insulation).

Usually a DC hipot test is not done on motors, a Pf or DF overpotential test is done per IEEE/ANSI std 95. Where did you get your test voltages from? What values do you expect to see?

I highly recommend you get a certified testing company to do these tests for you, not knowing how to hook up the test equipment is a big red flag.

We found acceptable readings on the IR test which I believe were around 100megaohm.

We have not yet done an PI test but were planning on doing that as a next step. Sounds like it should be done before Hi-Potting as you mentioned.

What are the Pf and DF tests which you referenced above, I am not familiar with them?

The test voltages I came up with are based upon a standard (dont recall which one) which specified a max test voltage of 2E+100V where E was the rated voltage. For a 4.16kV motor this would equal about 9320V or aprox 10kV. I would start the test at 1000V DC and increment in steps of 1000V holding each voltage level for 1min. I would expect the corrosponding leakage uA readings to have a linear relationship and be fairly straight and nothing that looked exponential in nature.

 

[zog]

We found acceptable readings on the IR test which I believe were around 100megaohm.

Good start

We have not yet done an PI test but were planning on doing that as a next step. Sounds like it should be done before Hi-Potting as you mentioned.

Yep

What are the Pf and DF tests which you referenced above, I am not familiar with them?

Power Factor or Dissipation factor, tests done in lieu of hipot

I would expect the corrosponding leakage uA readings to have a linear relationship and be fairly straight and nothing that looked exponential in nature.

No, you will see an exponential decay at each step. I highly recommend you have a certified test tech do these tests for you. You can find one in your area at www.netaworld.org

 

[philly]

Zog after hearing your respones and thinking it over more, I have decided that you are right and it probably best to go with an outside certified testing company

No, you will see an exponential decay at each step. I highly recommend you have a certified test tech do these tests for you. You can find one in your area at www.netaworld.org

I do not understand this statement. As the voltage is raised in each step I would expect the leakage current to rise on each step as well based upon ohms law. I would not expect it to increase very much which is why I said I should see a fairly linear straight responese.

Figure 1 in the following article shows this response that I mentioned and compares it to a bad response. The good response shows an increaseing linear relationship.

http://www.mt-online.com/article/0307_step_voltage_testing

Why are you saying that there will be an exponential decay with each step?

 

[zog]

Obviously leakage (Total measured) current will increase as you increase voltage to each step but after you get to your final test voltage leakage current should decay as your absorbsion current and capacitive charging currents decrease, you are only left with leakage current, which should be linear. The chart on your link dosent show that part. If you see an "upward bend" (Or non linear) in current during the step increases (As the link is showing) your insulation is bad and the test should be stopped.

 

[philly]

Obviously leakage (Total measured) current will increase as you increase voltage to each step but after you get to your final test voltage leakage current should decay as your absorbsion current and capacitive charging currents decrease, you are only left with leakage current, which should be linear. The chart on your link dosent show that part. If you see an "upward bend" (Or non linear) in current during the step increases (As the link is showing) your insulation is bad and the test should be stopped.

I agree that each time the voltage is raised the charging current and absorbtion current will increase and then decay exponentially with time. That is why I would hold each test point for 1 minute in order to allow these values to decay before taking a reading. Thats why I would plot the readings at the end of each minute rather than at the begining of each step in order to let these decays take place. After having a linear plot I would expect the final value to holt the same leakage current in a horizontal line if left for some time.

In the little bit of research I have done on the power factor/dissipation factor testing it appears that it is also referred to as a Tan-Delta testing. Basically this is a test that measures the insulation losses in a winding. These losses are a result of capacitance between gaps of other locations, and the associated losses can cause heat and be damaging to insulation.

The theory behind the test is that a voltage is applied between the winding and ground similar to any other test and the leakage current is measured. In this particular test however the leakage current is broken up in to resistive and reactive parts and a pf is determined similar to a regular pf calculation. The resistive part comes from the losses mentioned above which are in the form of resistance, and the reactive part comes from an ideal cable or winding insulation which should be totally capacitive. This relationship can tell how much losses are occur in a winding. Am I understanding correctly?

 

[zog]

I agree that each time the voltage is raised the charging current and absorbtion current will increase and then decay exponentially with time. That is why I would hold each test point for 1 minute in order to allow these values to decay before taking a reading. Thats why I would plot the readings at the end of each minute rather than at the begining of each step in order to let these decays take place. After having a linear plot I would expect the final value to holt the same leakage current in a horizontal line if left for some time.

In the little bit of research I have done on the power factor/dissipation factor testing it appears that it is also referred to as a Tan-Delta testing. Basically this is a test that measures the insulation losses in a winding. These losses are a result of capacitance between gaps of other locations, and the associated losses can cause heat and be damaging to insulation.

The theory behind the test is that a voltage is applied between the winding and ground similar to any other test and the leakage current is measured. In this particular test however the leakage current is broken up in to resistive and reactive parts and a pf is determined similar to a regular pf calculation. The resistive part comes from the losses mentioned above which are in the form of resistance, and the reactive part comes from an ideal cable or winding insulation which should be totally capacitive. This relationship can tell how much losses are occur in a winding. Am I understanding correctly?

You pretty much have it right, Tan Delta is a little different than Pf, different measurements, different test equipment but similar theory. You can learn a lot on PF testing at the Doble website, we use the M4000 PF/DF test set.