Megohm testing

[zog]

IR testing is all about trend analysis, what if you test something and it is greater than 2000M but you record "Infinity" on the report, someone elese tests the same equipment 3 years later and it is 2200M? It came all the way down from "Infinity to only 2200M. This would appear to be a negative trend when in fact it may not be.

 

[mivey]

Zog, do you ever give 110%

 

[zog]

No, thats impossible. I can only give 100%, and that is in short spurts, usually I give about 80%, the other 20% go towards life support functions.

 

[brian john]

I hope. How well can anyone determine anything on a "spot" test

In trouble shooting I can tell loads from spot readings when chasing ground faults.


I recently had a situation where busway that we had meggered at (average readings) 40,000 megohms, second company came in months later and had readings of 4,000. The factory engineers ripped me a new one, in writing to the end user, to me and to the county. This was brought on because a building engineer and electrician were burnt installing a busway fusible switch into an energized busway, this occured after the two seperate readings.

I had to point out several things.

Different instruments, different scales, and it was coincidental that my readings were 10 times the second contractors.

Different times and who knows what happens in between readings.

The second contractor readings were full scale readings, mine weren't,

The engineers said they had never heard of a megger that read above 10,000 megohms and I had to send them literature to prove I was no IDIOT.

We had to prove to our customer* before and after replacement of the busway that we KNEW what we were doing, the readings were close to the original readings we had recorded.

6 weeks later the busway blows and we look like real buffons* till we found the leaky roof 1.75 inches of rain in less than 24 hours.


*One of the most difficult persons I have ever had the pleasure of working with.

 

[76nemo]

In trouble shooting I can tell loads from spot readings when chasing ground faults.


I recently had a situation where busway that we had meggered at (average readings) 40,000 megohms, second company came in months later and had readings of 4,000. The factory engineers ripped me a new one, in writing to the end user, to me and to the county. This was brought on because a building engineer and electrician were burnt installing a busway fusible switch into an energized busway, this occured after the two seperate readings.

I had to point out several things.

Different instruments, different scales, and it was coincidental that my readings were 10 times the second contractors.

Different times and who knows what happens in between readings.

The second contractor readings were full scale readings, mine weren't,

The engineers said they had never heard of a megger that read above 10,000 megohms and I had to send them literature to prove I was no IDIOT.

We had to prove to our customer* before and after replacement of the busway that we KNEW what we were doing, the readings were close to the original readings we had recorded.

6 weeks later the busway blows and we look like real buffons* till we found the leaky roof 1.75 inches of rain in less than 24 hours.


*One of the most difficult persons I have ever had the pleasure of working with.

I gotcha, actually I was referring more to motor and tranny trending really. Going back and re-reading, that post didn't make much sense at all:grin:

PWHAI,......guilty

 

[76nemo]

In trouble shooting I can tell loads from spot readings when chasing ground faults.


I recently had a situation where busway that we had meggered at (average readings) 40,000 megohms, second company came in months later and had readings of 4,000. The factory engineers ripped me a new one, in writing to the end user, to me and to the county. This was brought on because a building engineer and electrician were burnt installing a busway fusible switch into an energized busway, this occured after the two seperate readings.

I had to point out several things.

Different instruments, different scales, and it was coincidental that my readings were 10 times the second contractors.

Different times and who knows what happens in between readings.

The second contractor readings were full scale readings, mine weren't,

The engineers said they had never heard of a megger that read above 10,000 megohms and I had to send them literature to prove I was no IDIOT.

We had to prove to our customer* before and after replacement of the busway that we KNEW what we were doing, the readings were close to the original readings we had recorded.

6 weeks later the busway blows and we look like real buffons* till we found the leaky roof 1.75 inches of rain in less than 24 hours.


*One of the most difficult persons I have ever had the pleasure of working with.

What are you using Brian that reads over 10G's????? Mine stops at 10

 

[mdshunk]

IR testing is all about trend analysis, ...

Maybe for you, but I use it mostly for troubleshooting. It is quite helpful to, for me, to compare a one-time reading to historical norms for similar equipment. Certainly a "bad" measurement will often stick out like a sore thumb. That's what I'm looking for. I'm not so worried about how big infinity might or might not be.

 

[jameselectric]

I was always taught that wire that was installed for a while (like motor/fan p.m.'s) were to have at least 1 megohm per 1000volts. So if a person did a 1000 volt meg on a 480 vo wire and got just over 1 meg ohm then its suppose to be ok. of course this NOT new wire. So you guys are saying this older wire would not be ok??

 

[brian john]

was always taught that wire that was installed for a while (like motor/fan p.m.'s) were to have at least 1 megohm per 1000volts. So if a person did a 1000 volt meg on a 480 vo wire and got just over 1 meg ohm then its suppose to be ok. of course this NOT new wire. So you guys are saying this older wire would not be ok??

Can you say NO WAY JOSE.


This is an old standard that has be rehashed in several post. Check the NETA website for current standards for different equipment.

http://www.netaworld.org/

 

[zog]

Maybe for you, but I use it mostly for troubleshooting. It is quite helpful to, for me, to compare a one-time reading to historical norms for similar equipment. Certainly a "bad" measurement will often stick out like a sore thumb. That's what I'm looking for. I'm not so worried about how big infinity might or might not be.

Well this is the CEU forum, so I figured I would post the correct info for those who want to learn instead of what is useful to you.

 

[wptski]

What about meggering NM-B, old installed NM or older black/grey type Romex? What voltage and readings to expect? Newer NM-B from Southwire is marked 600V but older NM isn't marked.

Did a Google and come up with what I believe NETA specs, 600V cable use 1000V and should be >100M. Some older installed NM gets a bit stressed at repeated 1000V tests and resistance declines but seems to recover with time.

You could very well be testing cable that's older than you are!

 

[electricalperson]

What about meggering NM-B, old installed NM or older black/grey type Romex? What voltage and readings to expect? Newer NM-B from Southwire is marked 600V but older NM isn't marked.

Did a Google and come up with what I believe NETA specs, 600V cable use 1000V and should be >100M. Some older installed NM gets a bit stressed at repeated 1000V tests and resistance declines but seems to recover with time.

You could very well be testing cable that's older than you are!

well if im testing old cable thats not up to my megger reading standards its getting ripped out. that one megaohm rule does not go for cable thats mainly for machinery. i dont know that much about meggering motors and stuff like that but someday ill know more about it

 

[weressl]

I am currently working on a job that requires a insulation test and report.Testing all of the shorter runs of 200 feet or less, I get a reading of infinity. On one circuit that is fed underground 600 feet with 350 mcm copper wire, I get a reading of approximately 40 Megohms. Is this normal. The wire pull went well and it has been energized. What is a normal reading.I have never had to megger anything before, where can I get a test report form.

"MEGGER": Insulation Resistance Test

Low voltage, non-shielded cables can be tested using a battery powered ohmmeter or a device called a "megger". Hand held ohmmeters generally have outputs from several volts to 24 volts. They are excellent for detecting direct "shorts" such as bolted faults and low resistant measurements in the kilohm range. A "megger" measures resistance in the megohm range using higher voltages than an ohmmeter. Manual or motor-driven meggers are available for a range of fixed dc voltages. Typical fixed dc voltages are 500, 1000, 2500 and 5000 volts.

2 to 50 Megohm Rule
Applied dc potential may be 500 or 1000 volts dc with the insulation resistance reading taken at one minute. A megohm meter reading of less than 50 megohms may be used for deciding when to investigate the cable installation, readings less than two megohms will most likely indicate damaged insulation. Readings of 2 to 50 megohms are usually associated with long circuit lengths, moisture, or contamination. In most cases a 2 to 50 megohm reading does not indicate the insulation quality, therefore 2 to 50 megohms should not be specified as a pass/fail value. Insulation resistance readings should be made and interpreted by experienced testing specialists to determine the condition of the cable's insulation.
Note that it is difficult to obtain accurate insulation resistance measurements in the field. Factory tests are done in a controlled environment with the cable submerged in water to provide an electrical ground. Using factory test values are not recommended.

 

[mdshunk]

What about meggering NM-B, old installed NM or older black/grey type Romex? What voltage and readings to expect? Newer NM-B from Southwire is marked 600V but older NM isn't marked.

Did a Google and come up with what I believe NETA specs, 600V cable use 1000V and should be >100M. Some older installed NM gets a bit stressed at repeated 1000V tests and resistance declines but seems to recover with time.

You could very well be testing cable that's older than you are!

Old BX will almost always "fail" the test. I swear that stuff must be hygroscopic.

 

[wptski]

Old BX will almost always "fail" the test. I swear that stuff must be hygroscopic.

So what values are real bad out of the bad?

 

[wptski]

Old BX will almost always "fail" the test. I swear that stuff must be hygroscopic.

The reason that I asked is because I just got some very low readings on a branch circuit with old Romex. Only 300K@50V! There is a GFCI at the end of the line, so I wondered if that would cause a low reading. I powered back to trip it, powered down and recheck again. Same thing again.

I have a portable GFCI called a Shock-Buster. Plugged into another handy circuit and got zero ohms at 50V, actually it's 4.62K in the resistance fuction of my Fluke 1507. I plugged into another circuit to trip it and plugged it back into the same circuit again.

Here's where it gets very odd! At 50V, it RESET the GFCI! I took a cord with terminals and wired it to Leviton GFCI. It reads low resistance also but doesn't RESET as the Shock-Buster does.

I'm not how a GFCI would react if wired in the middle of a series of receptacles but it appears that they create a problem when using a megger.

 

[brother]

1 megohm rule

1 megohm rule

well heres the link to the '1 meghom rule' of thumb. I remember hearing about that too when testing cable just installed. Also its what the 'biddlemegger' people have in their book.


http://forums.mikeholt.com/showthread.php?t=96120

 

[gar]

080704-1450 EST

wptski:

Visit this National Semiconductor site and look for the typical circuit just below the heading "Typical Application", about p6 of 8.
http://cache.national.com/ds/LM/LM1851.pdf

The electronics requires current and thus looks like a fairly low resistance compared to wire insulation.

.

 

[mdshunk]

I'm not how a GFCI would react if wired in the middle of a series of receptacles but it appears that they create a problem when using a megger.

Oh, absolutely. You cannot megger a circuit with a gfci in it. You need to remove it and wirenut the cables line to load. The GFCI will not only be ruined by the test, it will give you false readings.

 

[wptski]

Oh, absolutely. You cannot megger a circuit with a gfci in it. You need to remove it and wirenut the cables line to load. The GFCI will not only be ruined by the test, it will give you false readings.

It didn't bother the GFCIs one bit! All three still trip with the test button and RESET.