I have a choice as an end user on how I test cables. I’d like to avoid unbolting the LV terminations. Since I can’t do insulation resistance testing phase to phase with the delta winding connected, I wonder how much values there is in doing all that work. Can phase to ground tests accomplish most of my goal?
How often is insulation deterioration noticed phase to phase but not phase to ground?
When conducting insulation testing to ground with the cables connected, you will only measure the value of the weakest insulation to ground because everything will essentially become one large piece of copper.
Under this scenario, everything will read the same value. If a low insulation resistance is obtained, you will need to investigate what is causing the low readings, which means isolating the cables anyway.
If two phases were shorted, you wouldn’t know until it fails in service because everything is charged at the same DC potential for the test. You might get a good reading to ground, but when testing to other phases, you could find it’s unacceptable.
To obtain the most accurate values and determine whether a cable is deteriorating, you should test it individually with all other phases grounded.
How often do you find phase to phase shorts/insulation breakdown of cables without any indication of degradation on the phase to ground in a system that has been operating undisturbed?
My guess was almost never. All my cables peg out at 2.2GOhm.
Not often. Testing everything to ground you will get the value of the weakest chain in the link. If that value is unacceptable, you will need to test individually to find the issue. In this case being low voltage, concern would be raised in the 100M or less range. You are miles above that at 2G.