Multimeter, testing and reliable readings.

[Harry Bloomfield]

Eh?

Kind Regards, John

Not in what I read..

 

[JohnW2]

Not in what I read..

Yes, I realised that's what you meant, but (sorry!) it doesn't alter the fact that I am a bit lost ...

... If measurements gradually change and then 'settle down', despite your not 'touching anything', that can be due to capacitance/.capacitors associated with whatever is being measured.

Also true if there are capacitors in circuit, charging up and/or long cables, when measuring high resistance values/insulation.

What 'also' were you saying that I hadn't?

Kind Regards, John

 

[Harry Bloomfield]

Yes, I realised that's what you meant, but (sorry!) it doesn't alter the fact that I am a bit lost ...

What 'also' were you saying that I hadn't?

Kind Regards, John

Fact is I speed read and had not spotted your mention of capacitance. Is it really such a big issue if we both mention it?

 

[JohnW2]

Fact is I speed read and had not spotted your mention of capacitance. Is it really such a big issue if we both mention it?

Not at all - I was merely wondering whether you had perhaps thought of (but forgotten to mention) an 'also' which I had overlooked, given that the only bit of my post which you quoted (when you wrote "also ... capacitance") was that relating to capacitance! In any event, I hadn't noticed that your comment related to measurements of high resistance, so we are 'quits' in terms of not have read carefully/fully enough

In passing, in either case (low or high resistance measurement) it would really have to be 'a capacitor', rather than the capacitance of 'long leads'- since, for any credible (at least, for electricians) length of cable the 'charging up' of the capacitance of the cable would probably be over in microseconds.

Kind Regards, John

 

[conny]

I was wondering if moving the commutator could be inducing some kind of interference.

I may remove it from the housing and see if it reduces the variance.

Just to point out, I was not looking for accurate measurements per say. I was following a fault finding guide.

The guide suggested that while the values did not matter in themselves, a damaged bar would show an inconsistent result compared to the general value the others gave.

I was confused as I was seemingly getting wildly different values all the time.

I can actually see which bar is damaged, so I was more trying to see how that may manifest itself in a reading.

If you want to test the armature properly then you need to remove it from the influence of the field coils/magnets.
Set the armature on two wooden blocks with a vee cut in each block to rest the shaft ends in. Lightly polish the comm with fine glass paper, (NOT emery cloth!), and blow away any dust. If there is carbon dust between the comm segs this too has to be removed by lightly undercutting with a comm saw. You can make a passable comm saw with a fine hacksaw blade, (24tpi). Run the blade lightly across a grindstone to remove the kerf edges and grind the front of the top edge into a gentle curve so the front teeth can get right to the back of the comm slot. Once the bars are polished and the micas' undercut you can begin your testing. Depending on how the armature is wound the results may be different. A round wire, hand wound, lap winding will give totally different readings to a coil wound wave winding.
Lets assume your armature is something like a drill armature or washing machine so will be lap wound. Put a pencil dot at the front edge of 2 adjacent comm bars. We will refer to these as numbers 1 and 2. Subsequent bars will be 3, 4, 5 ...etc until you come back to number 1. This is your start point. Press your probes firmly between bars 1 and 2 and note the reading. Next place the probes between 2 and 3 and note the reading. Repeat this procedure between 3 and 4, 4 and 5 etc etc until you come back to the last bar and number 1 The readings will differ between each set or each couple of sets of bars. If you plotted them on a graph you will see they rise and fall. The lowest reading will be where the first coil was wound on to the armature and the highest will be the last coil to be wound on. You should find the lowest reading is next to the highest in your list of readings. If there is a wildly differing reading between and 2 bars, or even across a small section, then it seems to settle down to expected readings then you have a problem in that section and a more detailed test needs to be carried out. The test you have done is a static resistance test. A more accurate test is a 'live' volt drop test but involves applying a small voltage across a small section of bars and doing the above test again. This will give more accurate readings and a truer test of the condition of your armature.

 

[plugwash]

The saving grace is that, provided it is correctly calibrated, what none of these meters can do is produce a reading which is less than the true resistance,

Unfortunately a bad connection while zeroing out the leads can very much result in under-reported readings during the real test.

 

[JohnW2]

The saving grace is that, provided it is correctly calibrated, what none of these meters can do is produce a reading which is less than the true resistance...

Unfortunately a bad connection while zeroing out the leads can very much result in under-reported readings during the real test.

That's very true, and I suppose that I was thinking of 'zeroing' as part of "correctly calibrated".

Mind you, I don't think that any of my multimeters (even the more expensive ones) actually have a 'zeroing' facility, so I don't know how common it is. I must confess that, even with the MFT, I don't usually bother unless I'm using 'long leads' (which is very rare for me\), since the resistance of standard test leads is pretty negligible.

Kind Regards, John

 

[plugwash]

I've certainly seen a zeroing facility on a bench multimeter, not sure if any handhelds offer it.

 

[davelx]

I've certainly seen a zeroing facility on a bench multimeter, not sure if any handhelds offer it.

My analogue ones do

 

[Harry Bloomfield]

My analogue ones do

I binned my analogue ones years ago, all except my small AVO collection.

 

[bernardgreen]

An accurate ( but inconvenient way to measure resistance is the Wheatstone bridge method.

https://www.schoolphysics.co.uk/age...electricity/text/Wheatstone_bridge/index.html

This device for measuring the resistance of a telephone line,

 

[JohnW2]

My analogue ones do

So do mine, in a sense, but it's certainly not intended as a means for compensating for test lead resistance when measuring small resistances. Since it's just a mechanical adjustment of the pointer, I imaginee that if one tried to use it to compensate for lead resistance, that would go wrong because of the non-linearity of the scale, wouldn't it (even if one had an analogue meter that could sensibly measure very low resistances)?

Kind Regards, John

 

[JohnW2]

An accurate ( but inconvenient way to measure resistance is the Wheatstone bridge method.

That would not really address the issue being discussed here, would it? No matter what equipment or methodology one uses to measure a small resistance, the accuracy of the reading will be at the mercy of the quality (resistance) of the connections between the measuring equipment and whatever is being measured.

Kind Regards, John

 

[davelx]

So do mine, in a sense, but it's certainly not intended as a means for compensating for test lead resistance when measuring small resistances. Since it's just a mechanical adjustment of the pointer, I imaginee that if one tried to use it to compensate for lead resistance, that would go wrong because of the non-linearity of the scale, wouldn't it (even if one had an analogue meter that could sensibly measure very low resistances)?

Kind Regards, John

It might, I've not investigated ( I might - I have some low-value resistors in the component store). I tend to use my Avo-8 if I want an analogue meter to measure resistance, and that does have proper zeroing adjustments and a defined sequence for doing it.

 

[BS3036]

That would not really address the issue being discussed here, would it? No matter what equipment or methodology one uses to measure a small resistance, the accuracy of the reading will be at the mercy of the quality (resistance) of the connections between the measuring equipment and whatever is being measured.

Kind Regards, John

Not for a 4-wire measurement.