EBME Forums Forums Maintenance Technical Assistance Continuity Testing

When measuring low resistances, such as a piece of wire, the contact resistance of the test leads and their resistance becomes significant. Are you using a true four wire resistance measurement? With two wires passing the current and two wires measuring the resistance of the equipment (earth lead) and not the connection.
I have known many pieces of equipment to fail an EST and then pass once the earth pin was cleaned.
Robert

I'm saying that the negligible changes in the resistance of a resistor can't be compared to resistance change in a copper cable when power is disspated in it and that a significant resistance change should be observed in a copper cable that's heated up so you can feel it getting warm.

If resistance changes of up to 22% in a Copper cable with current flowing through it (i.e. power being dissipated in it) isn't observed under these circumstances then, in my opinion, a resistance tester isn't doing it's job very well.

Temperature coefficients of resistance are real and there's a higher likelihood of inaccuracies at lower test currents. i.e. instruments have errors (this is documented)- they can be "wrong" - it's whether this is acceptable or not.

What I've said, consistently, is that, based on practical considerations RE: measuring instrument design, at lower currents the accuracy and resolution of poorly-designed resistance testers are likely to be poorer thus contribute to significant errors - including causing a "fail" at low test currents when it indicates a "pass" at higher currents. This is precisely what I was getting at, in fact.

But you've said yourself that current needs to be passed through the resistance under test - a fluke doesn't do this so you're argument is invalid. Try measuring resistance in a Copper Cable while you're passing a range of currents and you'll see what sources of error there are in practice (the error being the inability of a resistance tester to be able to see the small changes in resistance due to temperature for example).

I've already established that I think they'll be no significant change in the resistance of a resistor , irrespective of the current passing through it, i.e. the (non-destructive) power dissipated in it.

Resistors don't vary in resistance when they're heated - that's how they're designed - that's the point of my previous posting (you obviously didn't read it or understnad what I was getting at). Copper cables will change their resistance over time, with current.

Bottom line is that we need to understand the measuring devices we're using and sources of errors before we decide whether the resistance values that we're required to record at lower currents, in particular these days, are accurate and repeatable.

When measuring Low Ohms there are sources of error:

At higher temperatures - thermally generated voltages at connections of dissimilar metals within the DUT and at test terminals. Losses in two-wire systems due to voltage-drops with flow of test current. Poor voltage measurement accuracy and/or resolution (especially at lower currents). Poor current source stability (over time/with temperature). Poor contact resistance at the test terminals.

dear Colin

The reason for 100mA is so that continuity of earth screening (functional Earth) can be tested as well as PAT testing on IT equipment. 100mA is not part of a medical standard (as is a 1010 body model) but merely an additional low current option for those interested in using less than 1A.

With regards to wetting contacts and having probelsm with 1A on Fluke products, Rigel / Seaward have presented a paper at the June IPEM seminar which might answer most questions / comments riased in this briefing. Anybody interested in further information, please contact me.

Hi RJ

Actually I do understand what you are saying (at length). I happen to agree with some of it.

Certainly the instruments are not 100% accurate, I merely observe that it strange that they measure resistors accurately but not wires, and that measured resistance changes, if it changes at all, in a matter entirely opposite to what I would have expected. Nothing you have said causes me to doubt the significance of these observations. But then I am a bit thick.

In general I dislike being patronised..." a simple formula", etc. I think I first met that particular formula in 1966 in first year O level physics - yes I am that old. I understood it then and I still understand it, although for how long I do not know...yes I am that old. Did I just repeat myself? How silly.

Actually Grahame it is a simple formula and I wasn't being patronising. I apologise if that's the way it came across. I agree that testers may be able to measure resistance pretty accurately but the whole point of my lengthy posts have been to push the point that not all testers may be able to measure, i.e. resolve, the differences in resistance due to heating, poor contacts, thermal emfs, etc, repeatably and with acceptable accuracy, at lower test currents in Copper earths - That was my point. Isn't the ability to observe small changes in resistance (not noise), whilst manipulating a cable, what we should be looking for if we can't drive 25A through a cable to test its integrity?

Thats a bit of a short reply !!!

Interesting, informative and factual as ever - thanks for that.....

Thanks for the information JB - unfortunately I didn't get the opportunity to attend the IPEM seminar.

Colin,

Buy the Rigel, we use the 233 , but the 277 will do most things you require of it. It tests things most pass. Job done!