Generator with low IR

When faced with a low insulation resistance (often below 1 Meg Ohm) I calculate the losses and make a judgement call as to the severity of the losses and the associated heating.

Phase to phase = 15 MΩ.
15 MΩ at 400 Volts = 0.027 mA
0.027 mA at 400 Volts = 0.0107 Watts per phase.
Total heating = 0.0107 x 3 = 0.032 Watts losses.
I would not be concerned.

Phase to phase IR? Is it 6 leads generator? If not, you should be getting zero megohms between phases.

For low voltage equipment I generally see the pass/fail at 1 megohm, with I think comes from a NETA (inter-National electrical testing association) testing guideline. I don’t remember the section, if you’re interested I can go back and find it and get you the reference.

Casey

edison123 said:

Is it 6 leads generator?

Click to expand...

In North America, quite possibly a 12 lead generator.

wcaseyharman said:

For low voltage equipment I generally see the pass/fail at 1 megohm,

Click to expand...

I never worried much unless the resistance was below 1 megOhm.

wcaseyharman said:

For low voltage equipment I generally see the pass/fail at 1 megohm, with I think comes from a NETA (inter-National electrical testing association) testing guideline. I don’t remember the section, if you’re interested I can go back and find it and get you the reference.

Casey

Click to expand...

Would you please do that for me please

Neta MTS-2011, page 135, section 7.15.1 has the following guidance:
- 1megohm per kV on armature windings made before 1970 and field windings

-100 megohms for ac and dc armature windings made past 1970 (form wound)

-5 megohms for random wound coils and form wound coils below 1000V.

This corrected to 40 deg C



Casey

Emad Shaaban said:

We are currently assessing the condition of a three-phase 2.5 kVA, 400V emergency generator, which is used for black start only. During our routine checks, we observed the following insulation resistance values:
• Initial readings (before cleaning):
• Phase to ground (IR): as low as 404 kΩ
• Phase to phase: approximately 1.2 MΩ

We then dismantled the rotor and performed the following maintenance steps:
• Cleaned the windings with isopropyl alcohol
• Coated the sleeve with electrical varnish
• Dried the unit using an air dryer

Post-maintenance insulation readings:
• Phase to ground: improved significantly to 1 GΩ
• Phase to phase: now reads 15 MΩ

My question is:
Are these 15 MΩ phase-to-phase readings acceptable for such a small generator used only for emergency/black start purposes? Or should further action be considered to improve these values

Click to expand...

It is interesting that the resistance of the phase windings to ground is less than the phase-to-phase resistance.
Generator is small and I assume that is good reason why insulation resistance value after 15 seconds testing is not noticeably lower than value after 60 seconds. This ratio can sometimes give a better insight into the state of insulation than the measured value itself

Please understand what the test is actually telling you.
An insulation resistance measurement is determining the resulting dielectric capability of an insulation system as a result of absorbed moisture INSIDE the insulating material, AND the presence of conductive SURFACE contaminant. Both of these conditions tend to degrade (or bypass) the dielectric properties of the insulation.

The OP found some low initial values (404k phase-ground, 1.2M phase-phase). After cleanup, both increased but not by anywhere near the same amount.

Why did the phase-ground value increase more than phase-phase? Because in the cleaning, the phase-ground tracking path along the surface was broken up and it no longer conducts. In contrast, the air dryer most likely did not get the paper insulation between phases to a temperature high enough to really drive out all the absorbed moisture. To do so would require raising the conductor temperature to something like 115 C and holding it there for an extended period (several hours) and then letting it slowly cool so it doesn't draw moisture back in.

It is also entirely possible that the phase paper insulation has been physically damaged to some extent (rips, tears, broken off pieces, etc.) such that it no longer provides enough of a barrier between adjacent phase conductors. Remember that these materials become very brittle after having the initial (and any subsequent) varnish treatment.

Are the last reported values (1 G-ohm phase-ground, 15 M-ohm phase-phase) acceptable? Regardless of the coil design, the 1 G-ohm value is fine. If the coil is a mush-wind design (e.g. round conductor with enamel coating), the 15 M-ohm is okay. But if the coil is a form-wound type (e.g. rectangular conductor with enamel and/or mica tape), it really should be significantly higher - up to around 100 M-ohms. I would look at anywhere the leads are in close proximity - like where they pass from inside to outside of the frame. Check for abrasion and/or other cracks and tears in the insulation that may be trapping moisture or other contaminant.

Emad Shaaban said:

We are currently assessing the condition of a three-phase 2.5 kVA, 400V emergency generator, which is used for black start only.

Click to expand...

The conditions of use (used for black start only.) are more concerning to me.

A spot check of insulation resistance may be immaterial.

Anecdote Alert:
True story, I was there.
Hurricane Mitch and the island of Guanaja.
Hurricane Mitch hit the small island of Guanaja three times in three days.
It ripped the tin roof off of our generator plant, about 200 feet from the shore, and dumped salt water on our generators.
I was the part time system engineer.
As soon as the wind stopped I was on a plane out to the Island.
The cables on our 5 diesel generators rose vertically from the generator junction boxes and thence by way of overhead trays to the switch room.
The water ran down the cables and past the old and loose cable glands.
When the cover was removed from one set, there was a white, salty, high water mark at the level of the bottom of the opening into the interior of the generator.

Talking with the major stock holder and president of the Utility;
"What do you suggest that you/we do?"
Me;
"Let me see what I can do with the first set, and based on what I find, we will decide what to do with the other sets."
"That sounds like a good idea. Go ahead."

First step, megger check. Zero Ohms to ground.
Second step. Try a voltmeter on a hunch. Reading a small voltage from the windings to ground.
The copper windings and the iron core, with the salt impregnated windings acting as an electrolyte were acting as a primitive battery.
The repair.
First step, we pulled the rotor.
Second step, using an air driven parts cleaner and rain water instead of solvent, we washed the rotor and the stator for hours.
We reassembled the set and shorted the leads. We excited the set with a battery charger and drove about 200 Amps through a winding rated at about 720 Amps.
With the cooling air intakes and exhausts blocked that was enough to build up a little heat in the windings.
Every couple of hours we would remove the blockages and let the moisture laden air circulate out, and then replace the blockages and let the warmth build up again.
We did that for at lest two days, maybe three days. It was a long time ago.

The result: I found a place where someone at some unknown time had done a megger test on the set and recorded 1.8 Meg-Ohms.
We got the set up to 7 Meg-Ohms.
That was the first set online and that held the load for a number of months.
Another meeting:
"Sir. based on the condition of that set, I recommend that the other sets be sent to a shop in Miami to be professionally cleaned, dipped and baked. As soon as we get a set back we should send this set up to Miami as well.
This set will probably hold the load indefinitely, but if it is ever shut down, do not start it up again before it goes to Miami."

A set came back, refurbished and was put into service. The original set was shut down, but it was not immediately sent up to Miami. (Did I mention that we were in Central America and shipping to Miami was not trivial.)(Why was it not sent? That is a crazy story for another post.)
Some months later, the load was increasing and more capacity was needed. Up until now, when extra capacity was needed, the plant operator would call a seafood plant and they would go onto their own generator, relieving some of the load.
This time, the plant operator was thinking;
"This set was working well. Why can't I use it now?" (What does the Gringo know??)
He started up the original machine and that was the last day in the life of that machine.
[/AA]

The point. The megger reading today may have little relationship to a megger reading several months from now when a black start is needed.
My recommendation:
Heat the windings. In this instance I would use stator current rather than space heaters.
I would arrange for a low voltage to automatically be connected to circulate in the windings whenever the set is off line.
I would include a current monitor alarm to warn if the heating current is interrupted.
You want the windings and the core to always be several degrees above the dew point.
Remember, the atmospheric dew point may rise rapidly at times. you want your windings and core to have enough delta T above ambient to anticipate a rapid rise in atmospheric dew point.