Cheap multimeter accuracy

[opps]

Hi all, I am back working at a regular customer's home. At the far end of the garden he has a single story summer house with its own consumer unit. He mentioned that the RCB would not reset and asked me to have a look at it. His garden has 8 different lighting circuits (don't ask...).

I eventually traced the problem to one of the cables that runs back to one of the switches in the house. I decided to loosen the light switch plate and use a length of telephone cable to connect it, first to the live at the switch. The other end ran in to the summer house and I connected the telephone cable to one terminal on my multimeter and the other to the live where it enters the summer house. I saw a reading of 004. I guessed it is just a voltage drop issue. Touching the multimeter probe against what would be the switched live and earth showed 1. I then repeated the above for the switched live, again it showed 004, I then touched the probe against the live, it showed 1. When I touched the probe against the earth, the display flashed 1350 momentarily and then eventually 1. I repeated this a number of times, with the same result.

Should I assume that cable has some kind of damage that allows a tiny bit of leakage between the switched live and the earth, and which is not detectable by a cheap meter?

 

[JohnW2]

I eventually traced the problem to one of the cables that runs back to one of the switches in the house. I decided to loosen the light switch plate and use a length of telephone cable to connect it, first to the live at the switch. The other end ran in to the summer house and I connected the telephone cable to one terminal on my multimeter and the other to the live where it enters the summer house.

Perhaps it's just me, but I find that rather hard to follow/understanding.

I saw a reading of 004. I guessed it is just a voltage drop issue. Touching the multimeter probe against what would be the switched live and earth showed 1. I then repeated the above for the switched live, again it showed 004, I then touched the probe against the live, it showed 1. When I touched the probe against the earth, the display flashed 1350 momentarily and then eventually 1. I repeated this a number of times, with the same result.

What are these meter readings (what 'range' was it on?) - volts or what - and can you perhaps be a bit clearer about what the two probes of your meter were connected to when you got these readings?

Kind Regards, John

 

[opps]

Perhaps it's just me, but I find that rather hard to follow/understanding.

What are these meter readings (what 'range' was it on?) - volts or what - and can you perhaps be a bit clearer about what the two probes of your meter were connected to when you got these readings?

Kind Regards, John

Sorry for not being clear.

I wanted to check the cable to see if there was any leakage across live/switched live/earth. I was using the continuity setting on the multimeter. The cable was isolated at the time. The distance from the house to the summer house is about 20 metres. Naturally the length of the leads on the multimeter are not long enough to span that distance, hence I used a length of telephone cable to run form each part of the cable back to the other end of the cable.

I hope that is clearer- I call it a "long line test".

 

[JohnW2]

Sorry for not being clear. .... I wanted to check the cable to see if there was any leakage across live/switched live/earth. I was using the continuity setting on the multimeter. The cable was isolated at the time. The distance from the house to the summer house is about 20 metres. Naturally the length of the leads on the multimeter are not long enough to span that distance, hence I used a length of telephone cable to run form each part of the cable back to the other end of the cable. ... I hope that is clearer- I call it a "long line test".

Thanks. A bit better, but I'm still not completely clear (again, perhaps it is just me!)

Are you saying that you are using your meter (on continuity setting), together with a 'long lead', to determine 'continuity' between the ends of the live feed from house to summer house? If so, are the readings you are quoting in Ohms? If so, they are low enough to indicate that there is 'continuity'.

However, I don't understand you reference to "leakage across live/switched live/earth". You sre very unlikley to measdure 'leakage" (as in low insulation resistance) with a multimeter, and certainly not with it on its 'continuity' range - and as for "leakage to earth", there does not seem to be any 'earth' in the equation of what you have so far told us about, so I'm not sure how you were hoping to detect that..

Kind Regards, John

 

[Deleted member 267285]

How cheap a multimeter are we talking? What make/model?

 

[opps]

The meter is very basic. I set it to the "continuity setting ", it looks like sound waves. TBH, I don't know what scale of resistance it uses. If there is no continuity, it will say 1, if there is continuity it should say 0. There was 20m of 1.5mm 3 core "pond cable" and the 20m of telephone cable, with that set up it was showing 004 (on which ever scale it uses- I guess that it is a percentage) when I check the individual wires..

Continuity from one end of the perm live and back- using the telephone cable as extensions for the meter showed 004. The same was true for the switched live. However, when I repeated the test for and bridged the switch live and earth via the meter, it showed a number that was far higher than one (albeit briefly), it then dropped to 1, suggesting no continuity.

 

[RandomGrinch]

I would prefer to use the resistance mode for this measurement - giving a definitive (if slightly inaccurate) numerical result.
The display in continuity mode needs to be interpreted - apart from the obvious, '0' is 'perfect' continuity, '1' will be open circuit. My multimeter shows a display when measuring continuity, but the manual only mentions a beep when below 70Ohm and OL when open circuit (so basically ignore the display!).
As above what is the model of the meter?
I'm wondering if the brief reading you get when connecting to SL is to do with capacitance in the cable, or some contamination around the terminals?

 

[opps]

How cheap a multimeter are we talking? What make/model?

Erm, about a tenner... I can't remember the make.

It is clearly less accurate than the RCD though, that will trip if I try to use that cable, and yes, I stripped down the whole circuit. In the meantime I have connected a number of the lights to the functioning live/switched wire.

I get that the RCB is better at detecting faults, my question was more about the limitations of multimeters.

 

[opps]

I would prefer to use the resistance mode for this measurement - giving a definitive (if slightly inaccurate) numerical result.
The display in continuity mode needs to be interpreted - apart from the obvious, '0' is 'perfect' continuity, '1' will be open circuit. My multimeter shows a display when measuring continuity, but the manual only mentions a beep when below 70Ohm and OL when open circuit (so basically ignore the display!).
As above what is the model of the meter?
I'm wondering if the brief reading you get when connecting to SL is to do with capacitance in the cable, or some contamination around the terminals?

Good points. BTW mine doesn't actually beep though.

Edit- I think that the cable is a continuous length, but unless I can find where it runs, I can't be sure.

 

[oldbutnotdead]

Your multimeter is using 5 or so volts at microamps to determine resistance. It should detect a gross fault (metal to metal short), it may or may not detect a subtle fault (eg carbonised insulation).
The proper tools for measuring insulation resistance use between 240 and 500 volts. They also cost a bit more than a tenner
EDIT When using a long lead as you are, the first job is to measure the resistance of your long lead. You need to subtract that value from any readings made using it to determine the resistance of the cable you are measuring.

 

[opps]

Your multimeter is using 5 or so volts at microamps to determine resistance. It should detect a gross fault (metal to metal short), it may or may not detect a subtle fault (eg carbonised insulation).
The proper tools for measuring insulation resistance use between 240 and 500 volts. They also cost a bit more than a tenner
EDIT When using a long lead as you are, the first job is to measure the resistance of your long lead. You need to subtract that value from any readings made using it to determine the resistance of the cable you are measuring.

I suspect that you are correct. Tnx

 

[ericmark]

A late friend of mine has a compressor, he had a problem so I got out my clamp on multi-meter, and could not really find a fault, the fault was that bad it ruptured a 13 amp fuse, returning and testing with a problem low ohm meter (must used 200 mA) and insulation tester (uses 500 volt) and the problem was very clear.

Nothing wrong with the multi-meter, but simply not the right tool for the job.

A insulation tester is not that expensive, around £35 quick hunt And the VC60B is cheap enough, but does not do continuity only insulation.

I got one and tested against my Mega, and same results. It's not the price that matters it is the volts and amps.

 

[SUNRAY]

FWIW I've found that all of the cheap (read disposable) meters I've owned have compared very well with my calibrated (used to be until 1994) company supplied meters, to the point I tend to trust any reading taken. However I'll always quote as 'using cheap meter' or something similar just in case i've discovered the reason the meter is cheap.
My main 'goto' at the moment is a Parkside at £10 from Lidl/Aldi (as seen here:

A real moan tonight...

It is one of those 'several people have had a look so can you go in and sort it out' jobs, I popped in to aquaint myself with it with the expectation I'd get the call when required. This is an outbuilding [happens to be an ex telephone exchange] currently used as changing room and pump house...

www.diynot.com

and so far has done everything I've needed it to do. At that price it doesn't matter when it gets dropped from a great height (my record for such a mishap is 30m so far). Generally speaking the nth degree of accuracy isn't required anyway even if these don't compare with a £1K multifunction device.

 

[RandomGrinch]

FWIW I've found that all of the cheap (read disposable) meters I've owned have compared very well with my calibrated (used to be until 1994) company supplied meters,

In the our labs we regularly check our meters against a bench meter that we keep calibrated, and we have very few major discrepancies.
We have hundreds of Flukes, but for a mid range choice, I can recommend Tenna/Uni-Trend meters - they seem relatively student proof!

 

[JohnW2]

FWIW I've found that all of the cheap (read disposable) meters I've owned have compared very well with my calibrated (used to be until 1994) company supplied meters, to the point I tend to trust any reading taken.

For what it's worth, exactly the same for me. Whenever I've compared their readings with expensive calibrated ones (which I've done quite often over the years - 'for reassurance'), the readings have always been very close.

Ironically, in all the years/decades, the only one I've had which started producing 'obviously incorrect' readings was a Fluke - and to give credit to them, Fluke replaced in FOC (I thought it was worth "trying it on"!), even though it was well out of warranty!#

Kind Regards, John