Testing burden voltage on my multimeters and one very old lab voltmeter

[txescientist]

Saw Dave's video on burden voltage long time ago and wanted to share my results. Also, I got one big and old laboratory Goerz voltmeter  to show you . I chose to measure 100mA and 300mA. These are common in hobby electronics.

-Iskra mA/A meter (made in my home country - former Yugoslavia)
-Mastech analog meter 7050 - my first meter 
-Iskra multimeter, rebranded Goerz, bought from military surplus.
-Mastech 8226T true RMS meter, the best buy for RMS meter here
-Mastech MY68 - cheaper, not true RMS, not so good at current measurement, but good for everyday low voltage hobby usage
-My favorite Won Hung Lo DT9208A, good for toolbox and not much else, but can be handy for voltage measurement when you don't have enough money for 4 better multimeters for measuring input/output current/voltage on the PSU.

The best is Iskra mA/A (~130mV) and the worst is MY68 (~2.7V).

-at the last image you can see that my MS8226T is almost bang on according to the Goerz voltmeter.

Thank you for reading this post  , and sorry for not so perfect image quality, I have old Kodak P850 with 5MP.

[txescientist]

Iskra Multimeter and MS8226T

[txescientist]

MY68 and DT9205A

[txescientist]

Old lab Goerz voltmeter. Sorry, no tear down, didn't have hearth to damage factory seal.

[bson]

It's not surprising that the moving dial has the lowest burden since it's effectively an inductor.  Its DC resistance is basically just a length of wire.  This is why moving dials are still great for nulling!

[wild]

Nice tests. I am also interested in this matter right now, so I find your post helpful.

It's also nice that from you do the measurement for two current value, that way I can actually see that the burden voltage has a linear behaviour for each meter as long as the range switch is not changed. At first I was confused why the Ikra mA/A and Ikra multimeter doesn't show a linear behaviour, but then I realize in the picture that you've changed the range switch for each measurement because they're not an autoranging meter.

I have a case with a meter that shows a 2.9V burden voltage while measuring 1 mA on its uA range. So that means the burden voltage for its uA range can be specified as 0.0029V per uA right? Does it means the multimeter has a shunt resistor of 2.9k Ohm?
So if I measure 600 uA in my circuit it will cause a 1.74V voltage drop which seems to be quite significant. The question is, does this voltage drop also affect the current readings of my meter?

Let's say a simple circuit with a single 15k Ohm resistor runs on 9V battery and it has 600 uA current through it without any measurements activity. Now I connect the multimeter I mentioned above to measure the current, what will happen?

• Does my meter accurately reads 600 uA but simply cause the voltage in the resistor drop to 7.26V? or,
• Does my meter add 2.9k Ohm in series with the 15k resistor causing the total resistance of the system raise to 17.9k Ohm and the current drops to 502.8 uA so that the reading on my meter would be 502.8 uA? or,
• None of the above, and I got it totally wrong.

Sorry for the newbie question, I'm struggling with the basic principle here. I'm considering to buy the meter I mentioned above, but high burden voltage is a known problem to that meter. I just want to know more about the effect of burden voltage.

[txescientist]

If you measure current, you will add series resistance to your circuit. Similarly, if you measure voltage, you will add parallel resistance to your circuit. That is why you want voltmeter with as high as you can get resistance (impedance) and ampermeter with low resistance. Multimeters have significant voltage drop on current ranges. Lower the range, higher is the resistance. If you follow EEVBlog you saw uCurrent that Dave made. This is an excellent way to significantly reduce burden voltage. Also helps if you chose right range for measurement. Measuring 1mA on uA range will get higher burden voltage than on mA range, but on mA range you will lose one digit. So to answer your question, you got it right at 2. your meter will add 2.9k in series, and because you have a constant voltage circuit, your current will drop. Why manufacturers do not use higher gain amps and smaller vaule shunt resistors I really don't know and it would be great if someone told us here. You can make your own uCurrent or you can order it from Dave. I made mine, just because it is too expensive for me - shipping and all...but it is not as nice as Dave's.

[txescientist]

bson: Nice avatar! Good old Baltazar. If you compare MS7050 with Iskra, you will see that regardless that both are analog (moving coil) MS have several times higher burden voltage. This is because moving coil in Iskra is far more sensitive than in MS.
You could get lower burden voltages if you could use galvanometer with light scale  as a voltmeter to a low resistance shunt, or you could use electrical amplifier (x100 like in uCurrent) and a regular meter  . It is all about the sensitivity of measuring circuit.

[joeqsmith]

Some nice old meters. 

It seems like if you wanted to compare meters, you need to keep a constant test setup.  The relative numbers would work out.

I attempted something similar with my Brymen BM869s.   The data sheet shows a burden voltage of 3.3mV/mA for the 50 and 500mA range.   Like most of these meter data sheets, I really don't know what that means.   Is it typical, worst case, who knows.   I was expecting to the resistance to be about 3.3 ohms.   Measuring it directly I got about 1.763ohms.   

With 100.00mA and measuring directly across the banana jacks of the meter, the drop was 0.17531 or 1.7531 ohms.
With 300.11mA and measuring directly across the banana jacks of the meter, the drop was 0.53625 or 1.7868 ohms.
An error of 33.7 mohms

The Brymen uses a Bussmann DMM-B-44/100-R fuse.  Making a 4-wire measurement on this fuse:
With 100.00mA and measuring directly across the fuse the drop was 70.026mV or 0.70026 ohms. 
With 300.11mA and measuring directly across the fuse the drop was 219.58mV or 0.73167 ohms. 

Subtracting the effects of the fuse:
At 100.00mA, 1.05284 ohms
At 300.11mA, 1.05518 ohms
Or an error of 2.3 mohms

The fuse contributes a fair amount of error but it's still part of the system and would need to be considered.

[zlymex]

Nice test.
It's rare to find analog meters with less than 100mV burden voltage at full Amp scale.
There are many analog meters that the head unit has only 30mV burden or less at full range, but the burden voltage of current range is much larger, I don't know why they don't make full use of the small burden voltages already there, probably because the very small resistors needed are difficult to make, or thermal EMF considerations?
One exception is the Simens mV-MULTIZET, only 9mV norminal burden voltage at full scale.  The full scale of the meter head is only 2.64mV(176uA times 15 Ohm), and there is a 3mV voltage range.

[joeqsmith]

Looking at the UNI-T UT181A, I did not see where the data sheet shows a burden voltage.   Measuring it directly I got about 1.185 ohms.   

With 100.00mA and measuring directly across the banana jacks of the meter, the drop was 0.11425 or 1.1425 ohms.
With 300.00mA and measuring directly across the banana jacks of the meter, the drop was 0.34781 or 1.1594 ohms.
An error of 16.9 mohms

The fuse used in the 181A is a much smaller package.   I would guess both the Brymen and UNI-T are using a 1 ohm shunt in this range.

[bson]

bson: Nice avatar! Good old Baltazar.

Indeed!