Is it acceptable to enter low IR on MWC

[FreeLec]

When doing minor works on a circuit, which may take a very short time to complete and it requires a MWC the IR test between L-N and E may be zero due to other electronic items.
You can end up spending 10 times longer looking for and disconnecting them just to get a good IR reading.

The company I work for has a policy of entering the low reading with an explanation of why and then arranging extra time at another time and charging to diagnose and rectify, even if it just means tracing electronic items, disconnecting and re-testing.

This is acceptable to the establishments they are working for.

However, it does not seem correct to enter what is effectively a fault on a certificate that is supposed to show the circuit as tested and okay, just as with an EIC. It's not a EICR so I would have thought this wrong.

However, what other way of doing it is there short of saying to the customer that before any work is done the circuit has to be tested as to fitness and any faults rectified but it's done the other way round with this lot ending up with a fault on the MWC. Only when the fault is cleared (or found that it is just electronic items) is the MWC changed.
The one with the fault is still given to the client until the updated one is done.

There are times, though, when no one has gone back to re-test as the client has given no further instruction, so the client has ended up with a MWC with an IR fault.

Any thoughts?

 

[Octopus]

Define low reading.......

 

[FreeLec]

Anything under 1 meg but usually 0.01 or 0.00 as even one piece of equipment can cause that

 

[Andy78]

Are the correct procedures for testing of IR with electronic equipment connected followed ?

 

[FreeLec]

L-N connected together. N removed from N bar. 250V, not 500V.

 

[Lucien Nunes]

L-N connected together. N removed from N bar. 250V, not 500V.

and

0.01 or 0.00 as even one piece of equipment can cause that

...do not tally. I don't know of any normal piece of equipment that has just tens of kΩ or lower IR to earth. What sort of equipment are we talking about here? The only time I see this kind of result on a 250V test is where US-spec 120V equipment containing surge arrestors designed for 120V supplies is supplied via an auto-transformer from 230V. Because the transformer passes the DC test voltage direct to the load, the VDRs can start to conduct and throw the reading off.

What specifics can you think of on these sites that would be causing this?

 

[FreeLec]

Have to go out. I will respond in about an hour or two.

 

[Octopus]

Do your employers charge for the 2nd visit?

 

[FreeLec]

Yes.
They charge for the time taken to investigate whatever the problem may be. If it's a matter or taking something out of circuit to re-test or a fault that needs rectifying it is chargeable. If the fault is major then that may even involve arranging a further visit to allow time to rectify. All chargeable.
This is something that is arranged with the clients as they are educational establishments where disruption can be an issue.

 

[Pete999]

When doing minor works on a circuit, which may take a very short time to complete and it requires a MWC the IR test between L-N and E may be zero due to other electronic items.
You can end up spending 10 times longer looking for and disconnecting them just to get a good IR reading.

The company I work for has a policy of entering the low reading with an explanation of why and then arranging extra time at another time and charging to diagnose and rectify, even if it just means tracing electronic items, disconnecting and re-testing.

This is acceptable to the establishments they are working for.

However, it does not seem correct to enter what is effectively a fault on a certificate that is supposed to show the circuit as tested and okay, just as with an EIC. It's not a EICR so I would have thought this wrong.

However, what other way of doing it is there short of saying to the customer that before any work is done the circuit has to be tested as to fitness and any faults rectified but it's done the other way round with this lot ending up with a fault on the MWC. Only when the fault is cleared (or found that it is just electronic items) is the MWC changed.
The one with the fault is still given to the client until the updated one is done.

There are times, though, when no one has gone back to re-test as the client has given no further instruction, so the client has ended up with a MWC with an IR fault.

Any thoughts?

Link L&N and test IR to Earth and note on the cert, sorry forgot set test voltages to 250v

 

[FreeLec]

Pete999

Look at comments 4 and 5 above.

 

[Pete999]

Pete999

Look at comments 4 and 5 above.

Point taken must read the whole thread

 

[FreeLec]

LUCIEN NUNES

Sorry! Completely forgot to look at your comment when I got back as someone else left one and I got distracted by that.

I don't understand why you would think that the two comments would not tally. Some electronic equipment have functional earths so there will be a current flow, all be it very small, between L-N and earth. One well known example right at the start of the circuit is an RCBO with an earth tail. Even the simplest circuit with nothing connected, e.g: radial to one socket outlet with nothing plugged in, will show 0.01 Mohm between L-N and E unless you either disconnect the earth tail from the earth bar or you remove both L and N from the RCBO.

 

[davesparks]

Why is this only becoming apparent at the testing stage?
Any faults in the existing should be identified before you start work, not after.

 

[Octopus]

Hum.... so you quote for a job, don't isolate everything, issue a cert stating extra investigation is required, then go back for a 2nd chargeable visit.

Sounds like a dodgy money making exercise to me.

 

[FreeLec]

davesparks
Agreed. However, in reality I would never get anything done.
A spark I work with said he thinks it only refers to anything that has been added.
I agree with you, if a circuit has a fault by rights it should not have any modifications until the fault is rectified. I test for Zs as this is quick. If there is a problem I don't start work on it, report it, move on to the next thing.
However, IR can involve a lot longer process and I would never get anything done if I ended up fully testing the circuit with disconnecting every piece of electronic item that has functional earths etc, just to make sure the circuit IR is OK before I start anything.

Also, you may be modifying in a way that you are adding nothing but removing and re-routing.
Same issue.

Murdoch
NO! That's not what I said.
Re read the response I gave to you (no 9)
Read the last sentence.


Remember, I work for a company, not myself. I already said that I'm not comfortable with this and this is not the only work being done. Other work does not have these issues.

 

[FreeLec]

Something's wrong with this. I keep loosing the message I post.

 

[FreeLec]

Something's wrong with this. I keep loosing the message I post.

 

[Lucien Nunes]

I don't understand why you would think that the two comments would not tally

Because I design, build and test a wide variety of electronic equipment and find that very few devices in general usage today deliberately have such a low resistance L+N->E. Functional earthing yes, capacitive leakage current yes, all understood, but 10kΩ DC resistance? Typically the lowest I encounter for an SMPSU is around 7MΩ. We know about RCBOs but I didn't count them as connected equipment and you are in a position to disconnect them.

OK, I understand that you might be coming across a lot of faulty circuits and your original question still applies to those, but I would be keen to know what current-using devices, other than faulty ones, are complicating your testing to the extent that it requires callbacks.

 

[FreeLec]

Regarding circuits that have faults, until I am 100% certain that all equipment is disconnected I cannot say for certain that there is a fault in the circuit.
Many is the time that I have tested a ring or radial and have got low IR until I find something hidden out of the way or not expected to be on the circuit being tested. Unplug it and finally the reading will go up to >999. This can also include active RCD sockets.
Lighting can have something somewhere in the dropped ceiling that can take an age to find. Controllers for DALI, DSI, 1 to 10V etc.
As to exactly what the equipment is; I don't keep a note so I cannot tell you what exactly they are.
My intention now is to directly test various equipment disconnected from the supply, between L-N and E on the supply lead and see what type of equipment cause the problem.
When you are testing you don't take notice of exactly what the item is. You see something and disconnect it or unplug it.
I don't know how long this will take until I get a decent list of items but when I do I'll post it under an obvious title for you and others to see.
However, I don't understand why you think that there should not be an issue. You accept that RCBOs have that issue so why not other types of equipment?

Here are links to another post I put up. Look at the response from others who replied with an explanation showing that there is an issue. I was trying to understand what it is that causes this. I have a basic understanding of electronics but when it gets to deep I get lost in the explanation.

IET Forums - Electronic Equipment and IR testing

Electronic Equipment and IR testing. What component causes low reading to earth? | Screwfix Community Forum

The callbacks are agreed with the client in the sense that we have been told not to spend time doing something that has not been factored into the original quote. Account has not been taken of any problems. No one is being conned. The client wants it this way as there are sometimes limited time slots to get the work done and has to be tightly scheduled. Drives me mad. It's frustrating to work that way.