1 ohm resistors-measurement problem

[Diabolical Artificer]

Just bought some 1ohm 50w resistors to make a dummy load.at the same time I bought a new DMM.When I measured the resistance of the new r,s,Im only getting 0.3 reading.However when I measured a 100 ohm resistor,am getting an accurate reading(takeing into account the +/-5 % tolerance)
This considering the meter reads 0.3 ohm when leads are shorted to each other.Getting the same with my old DMM,any ideas? M

[murphyv310]

Hi.
Try the old ohms law way pass one amp through one and see what you get in voltage across the resistor (1V) if not they may be wrongly marked or your DMM is inaccurate at the low end.
Cheers
Trevor

[theoldtrout]

I think the above post is possibly the best way to verify these resistors, but then there are two instruments and possible cumulative errors. I have found that many low cost DMMs tend to be inaccurate at the bottom end; I'm fortunate to have some resistor standards 0.2% to check this.

[Herald1360]

If the resolution of your DMM ohms range is OK, try measuring (say) a known 10 ohm resistor, then the same resistor plus your 1 ohm in series. The difference should be a reasonable indication of its value.

Or if you've got any reasonably stable voltage source, like a bench psu, connect the 10 ohm 1ohm in series across the output, set the psu to approx 1 V, then measure the voltage across the 10 ohm and the voltage across the 1 ohm with the DMM. The voltage ratio will be the same as the resistance ratio and it doesn't matter what the voltage actually is, as long as it's steady.

[Kat Manton]

Hi,

A DMM which can use four wires for resistance measurement would do better. I haven't tried my new bench DMM (which can do 4-wire measurement) on really low value resistors but I do know my old Beckman handheld (2-wire only) was pretty much useless for very low resistance measurement.

Regards, Kat

[Colin]

I had this problem when trying to make up some low resistance meter shunts, became very frustrated and eventually built this device ...

http://www.discovercircuits.com/front/injector.html

It was cheap to build and works really well. It's been a very useful piece of kit to have in the workshop.

[SPCh]

Ahh ! The delights (and pitfalls !) of metrology.

It all comes down, as I've said before, to having something you can believe in.

And you can't, mustn't, blindly believe in anything !

It all comes down to Ohm's Law (which I hope we can all believe in - at least on DC).

V = IR; I = V/R; R = V/I

But can you trust (the I & the R), or (the V & the R), or the (V & the I) ?

The safest way is the traditional Wheatstone Bridge: the absolute V & I are irrelevant
because you're simply balancing R(1) against R(2) with a zero detector:
(and even a DMM can identify "zero" with a fair degree of accuracy !)

If you feed a current - any current - through two resistors in series, the ratio of the voltages across them will give the ratio of the resistances.
If the voltages are EQUAL the accuracy of the voltmeter no longer matters:
the resistances are equal.

But you need an adjustable "reference resistor" to establish the equality -
in effect, the variable resistance arm of the Wheatstone Bridge or a reliable "resistance box".

How reliable is "reliable" ? It's your choice, but the result can never be better than the accuracy of your reference.

For practical purposes - i.e. a fallback approximation to reality - the most accurate resistors you'll find (if you're prepared to believe them) are

(1) between the terminals of a good quality multimeter (AVOs were generally better than 0.3% - but HOW MUCH better ? You see the problem ...!), or

(2) between the terminals of a "good" decade resistance box. "Good" does not necessarily mean expensive; it generally means OLD.

"Classic" resistance boxes (from good 'stables' such as Tinsley, Gambrell, Baird & Tatlock, Philip Harris or Griffin & George), when new, offered better than 0.25% (guaranteed) accuracy and often better than 0.15% (in practice, though never claimed or guaranteed).
With time, abuse and repair these values could only get worse.

For what it's worth (which may not be much) I've spent much of the last five years trying to measure resistances "accurately". It's a time-consuming (and often frustrating) exercise, and one which - even in principle - can never yield more accuracy than your "ultimate act of faith". But with two "lumps of stuff" certified by several (one is not enough !) professional metrology labs at within 0.05% of their nominal values (and that's only 5 ohms in 10kohms !) I'm comfortable, to within 0.1%, of most resistance values I now determine.

But it's neither quick nor simple: you have to wait for - or "engineer" - the correct temperature and humidity (yet another metrological exercise) before you can even begin !

Conclusion: there's no EASY answer. Decide what accuracy you "need", and design your test method appropriately. But you'll always need "quality references" (which never offer accuracy, but only guaranteed levels of inaccuracy) before you begin.

Sorry to sound pessimistic, but metrology is as much a branch of philosophy as it is of engineering; and, as Lehrer said, philosophers are just "people who give helpful advice to people who are happier than they are".

[DangerMan]

Ah... minefield-alert!...
included in what SPCH says [it may be buried in his text, and its context not realised, but it's none the less important!] the resistor may be eventually proven to be 1 ohm at DC, but could be considerably different at any AC frequency, especially at RF. You may need to account for "L".
What sort of dummy load are you making ? [RF or DC?]
If for use at RF, wirewound resistors might just have too much inductance to be properly useable... even specially wound "anti-bifilar" [supposedly non-inductive] ones. Depends on the resistor and the frequency.
If for low voltage DC... do you have several 1 ohm resistors ? If so, what do they measure when wired in series? Do you begin to "escape" from the region where inaccuries due to cable and contact errors are significant compared to one resistor value?
Pete

[russell_w_b]

Originally Posted by magnetic

Just bought some 1ohm 50w resistors to make a dummy load.

Are you wanting precision or just something to dump power into with mininum reflected power?

Why not configure them as you plan, then measure the resistance (assuming you want a higher resistance, of course). They may then be easier to measure with the equipment you have, and you could 'trim' the values in to the required value easier.

For individual measurements, I would use a wheatstone bridge, or a kelvin-type four wire device, as suggested in other posts.

[MichaelR]

Originally Posted by magnetic

This considering the meter reads 0.3 ohm when leads are shorted to each other.Getting the same with my old DMM,any ideas? M

From the original question I fail to see how the DMM can be so far out at measuring 1 ohm resistor , it is showing 0.3ohm the same reading when the DMM is reading a short circuit.

What DMM are you using , does it have a quoted spec ?

All of the advice given so far is excellent but I do not think it answering the original question.

Mike

[Diabolical Artificer]

Hmm,to be honest Im non the wiser,SPCh,s post seemed more like a philisophical discussion,my head hurts!
Im trying to build a dummy load,to test valve amps (AF).Ive got an osc,scope,getting a sig gen from oldtimer,now I need a dummy load.
So have got,3x 1 ohm 50w,1x 1ohm 100w,in series that should give me 4 ohms.Also got 1x4.7ohm 50w +3.3ohm 15w,in series should give me 8 ohm ish.Also got various R,s off old cross overs?old psu,s.Trying for,4 8 16 ohm dummy load,with all R,s mounted on scavenged heat sinks,with a few fans to keep cool,etc.
Remember just starting out.I know ohms law,etc,but knowledge pretty basic.
Am using aTenma 72-7750 DMM,this the meter giving me 0.3 probes shorted.My other meter a Tilka 90030064 reads 0.7 ohms probes shorted.So in a nutshell,how accurate should my DLoad be,and if I cant rely on my DMM,s how can I calculate accuratly? M:all componants CPC/Farnell

[russell_w_b]

Originally Posted by magnetic

'Im trying to build a dummy load,to test valve amps (AF)'...

We load-test our Leak TL/12 Plus valve-amps with a couple of 25W 33R 5% resistors and the arrangement works perfectly satisfactorily.

I wouldn't get too hung up over tolerance precision and just go for the recommended load for your amplifier with the values you have, but make sure they're rated to handle the full RMS power at peak voltage.

Ensure the frequency response of your meters is accurate for your range of test frequencies! You should be able to check any discrepancies by finding the peak voltage from your 'scope and thus obtaining an RMS figure for your power calculations.

[MichaelR]

Originally Posted by magnetic

Am using aTenma 72-7750 DMM,this the meter giving me 0.3 probes shorted.My other meter a Tilka 90030064 reads 0.7 ohms probes shorted.So in a nutshell,how accurate should my DLoad be,and if I cant rely on my DMM,s how can I calculate accuratly? M:all componants CPC/Farnell

You are right to try and reason what is going off with your measurements and readings , if you do not, you will never get an understanding and in particular with using measuring instruments you may well "fool" yourself with the readings.

Can I ask the question? are you not able to measure the resistance of the resistors using either of your DMM or are you worried about the difference in reading between the 2 DMM when their leads are shorted.

Mike

[Station X]

I don't think the problem has been accurately defined. As I read post #1 if the DMM's leads are shorted it reads 0.3 ohms, but still reads 0.3 ohms when measuring a 1 ohm resistor.

What reading do you get if you measure the series combination of 4 times 1 ohm resistors?

[Diabolical Artificer]

Hi,thanks for all the input so far.With the 4 1 0hm r,s connected in series,am getting a reading of 1.2 ohms on my old Hilka,and on my new Tenma well no reading at all.Thats to say it reads 22.5,11.3,.4, etc.IE it wont settle to any reading.However with a 4.7 and a 3.3 in series,am getting a steady reading of 1.2 ohm on the new tenma.
Apart from the totally inaccurate readings,it seems neither DMM likes measureing at low resistance.But surely the resistors Ive got cant be out by that much.
This all started out,when I was trying out me new DMM for the first time.It measures great on caps.I did wonder if the phisycal size of the r,s meant they were suking up too much juice or something from the Dmm,s.
Guess the only way I,ll know for sure,is to put some V,s through,and measure using Amp meter and voltmeter.M

[John M0GLN]

I don't suppose its of much help, and I don't want to turn your thread into a discussion on the merits of various makes of DMMs, however out of interest I tried a 1R0 resistor on my Fluke 73, with the leads shorted it read 0 and when measuring the resistor it read 1.0, so they can or should read low values.

John

[Colin]

Sounds like something's wrong somewhere ... If the DMM will read 0.3R with the leads shorted (which sounds about right for a pair of ordinary test leads) and no reading at all (or it floats) with the four 1R resistors in series then the string is open circuit.

[MichaelR]

Hi Magnetic,

I too think there is something wrong it should measure the resistance how accurate that measurement is another question.

Mike

[Station X]

Are you using the same test leads with both meters?

[KeithsTV]

Are you sure the 1ohm resistors are 1 ohm and not 0.1ohms? What are the markings on them?

I've just found in my power resistor box, an RS 50w metal clad resistor, stock number 158-446, which is marked R1+/-5% and is 0.1ohms.

On some DMMs, such as the Fluke 79, it is possible to "Calibrate" the probes by switching to the continutiy setting, shorting them out and pressing the button. This then zeros the meter and allows low resistance measurements to be made. Without this the reading when the probes are shorted together the reading is 0.3ohms.

Keith