Author Topic: What is the internal resistance of the Uni-T UT139C (on the mV range)?  (Read 3908 times)

0 Members and 1 Guest are viewing this topic.

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
What is the internal resistance of the Uni-T UT139C when used as a DC voltmeter in the 600mV range?
The manual does not specify - it just says 10 megaohm for all ranges. But this is not the case, since the UT139C is capable of reading the correct VBE in a transistor with disconnected base. (My UT61E cannot do it because the mV range stops at 200mV).

I can measure the resistance in the volt-autorange setting (it comes out as 13.5 megaohm, in the right ballpark) but if I try to measure the current in the 600mV range I can only read 0.00 microamps. Someone with serious instruments (easy to find in here) and a UT139C (which might be in a different disjoint set :-) ) has ever measured it with a decent accuracy?

It is probably like the input resistance of the UT61E which is > 3 gigaohm, but... why on earth it is not stated in the manual is beyond me.
I mean, it's a good feature so why not advertise it?

« Last Edit: March 07, 2023, 10:49:33 pm by Sredni »
All instruments lie. Usually on the bench.
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 8995
  • Country: us
  • Retired, now restoring antique test equipment
I do not understand your conclusion about the Vbe (open) measurement.
If a voltmeter has roughly 10 megohm input resistance and some finite input current (> 60 nA) which is positive (which it may be) and you place it across a diode (B-E) in the forward direction, you should see somewhat less than 600 mV (since the diode current is very low).
To measure the input resistance, wire a 20 megohm (or greater) resistor in series with a 1.5 V cell (alkaline or other) to the 600 mV input.
To get an accurate result, you can measure the cell with the voltmeter on an appropriate range.
« Last Edit: March 08, 2023, 12:13:23 am by TimFox »
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Tim,

it's a transistor with a floating base. I applied 8.2 volts between collector and emitter with the base disconnected. Simulation (and theory) says that due to leakage currents the VBE will about 230-240 mV at 27°C but I could not measure it inside my house because of the capacitive coupling with the mains lines (I get a negative value of a few volts for VBE) or for too small voltmeter resistance (it acts as a shunt).

I went out in my garden, far away from any buried mains cable and with the UT139C in mV range I was able to have a reliable measure. It was in the middle of the night and it was cold; at 0°C the simulated value is 380mV (I measured about 400 mV stable). Redid it tonight at 7°C and read some 320-330 mV (I did not bother to jot it down).

10meg input resistance is not enough to get the correct reading.

As for the negative multi-volt VBE I get in my house, I simulated capacitive coupling with AC mains and I see the VBE is a multi-volt sinuosoid dangling from the 250-300 mV level. The BE diode and the coupling capacitor seem to behave as a clamping circuit.

All instruments lie. Usually on the bench.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
I have the similar UT133A. A quick test with a 1 µF capacitor and switching from the Ohm to 600 mV range shows a moderately fast discharge curve. From the time it takes to get from 600 mV to 300 V, I would guess some 10 M ohm. The test is easy to repeat.


For the measurement with the diode, I am afraid one could still pick up some mains hum. Another factor is light hitting the diode, that can produce a tiny current and finally even of the input is high impedance there usually is a bias current from the meter.  Even with the good bench DMMs it is common to see some 10-50 pA in high impedance voltage mode.

Just a diode in equilibrium and no external power source will show 0 V  as there is always some residual parallel resistance and with no power there is no voltage.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
I have yet to try the voltage divider method Tim suggested because I have to find first a multimegaohm resistor in my pile of junk. I was aware of the method but I thought that with the tolerances of my instruments I could not get a reasonably accurate value. I might get a fairly good indication though, and surely it will tell me right away if the internal resistance is 10 meg or much higher. I will do it as soon as I find a suitable resistor.

I am convinced that the 600 mV range of the UT139C has much more than 10 megs because I am getting reliable measures of the BE voltage that I do not get with other meters with 10-15 megs resistance. The diode is not unpowered. There is a leakage current and there is a voltage across it (predicted by both theory and simulation). The voltage I measure (in the middle of the night) far away from any mains cable (I mean farther than 15 meters from ANY cable) is not only consistent with the theoretical value, but it also follows the temperature dependance predicted by theory. This is not a random reading. (And I have to say, I was surprised I could measure it so consistently).

EDIT: the values I measured are slightly higher than what theory predicts and this could be the result of the current flowing in the meter's probes. I have to say, tho, that up to now I have only tested one transistor. I will try with different BJTs, maybe the one I got is cooked and gives by mere chance values that appear to agree with theory?
« Last Edit: March 08, 2023, 04:53:16 pm by Sredni »
All instruments lie. Usually on the bench.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
The diode voltage reading is not a good test, as it is sensitive to AC interference and it is by no means clear what the correct reading is. Even if there is no mains hum, there could be interference from the meter itself and possibly from a radio station, posisbly amplified by resonance in a cable.
With just the diode/ transitor and no power source the prediction of the theory is zero voltage - that is very basic / fundamental thermodynamics, nothing fancy.


The test with the capacitor (e.g. 1 µF range film type) is easy. A smaller capacitor could also check for very small bias currents.

For a first test with the resistor just 1 MOhm should also be OK.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
This is LTSpice's simulation.


https://i.postimg.cc/Yqsn1PDV/screenshot.png

https://i.postimg.cc/T1SVWz2J/screenshot-2.png

(the reason for the tran plot is that I also simulated the effects of capacitive coupling with the mains line, to explain the negative measure obtained by userxxxx in this post on SE:  https://electronics.stackexchange.com/questions/655771/what-will-happen-if-we-give-voltage-to-a-transistor-bjt-across-emitter-and-col)

I am usually diffident of simulations with floating terminals, but in this case it appears Mike Engelhart (always be praised) got it right. (Microcap provides a similar result, even if with a smaller VBE). I am not sure what you are referring to when you say that the voltage across the base and emitter terminals should be zero. This is not what theory predicts. At room temperature it should be some 230-250 mV and thermodynamics will be very happy with that. Loosely speaking, the voltage comes from ICEO.
(As I wrote in my comment on SE, you can imagine two diodes back to back: the lower diode will be biased by the leakage current and will show a voltage across it; the transistor is more complicated because of the shared base but the need for a voltage there is clear).

EDIT: oh well, and now I do not see the images even on my smartphone. Ah, the evolution of the web: taking it all away one piece at the time... I added the direct links.
« Last Edit: March 08, 2023, 06:50:23 pm by Sredni »
All instruments lie. Usually on the bench.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
I am usually diffident of simulations with floating terminals,

As you should be if you expect your simulation to reflect any actual reality.  I tried your circuit with two high-impedance meters, on selectable for 10M or >10G, the other >1G.  I used a 1N3904 since I didn't have 2N2222 handy.  The result was that the >1G meter showed a very small voltage, a few millivolts (often negative) and a bit of noise.  Interestingly it behaved differently if I measured the 8V first and then the base--it then would measure -1.6V.  The other meter in 10M mode showed nothing but some noise but in the >10G mode, it showed varying voltages of 10s and 100s of millivolts, almost always negative.  So clearly all of these are dominated by the meter characteristics and external noise that is being rectified.  Your circuit is essentially an untuned receiver and I think you'll have a pretty hard time finding a place where the external noise is low enough to see the effects you expect.  You might add to your simulation by putting very high value resistors across the base and emitter to account for the voltmeter so you can tell us what currents are expected in each situation.  The meters will have input currents in the single-digit nA or less by spec, most likely a lot less in reality. 

Just to be complete, I measured the leakage current from C to E @8.1V with the base open and got less than 1nA.  I didn't break out the picoammeter, but do you expect <<1nA to bias the B-E 'diode' to 380mV?  Whatever your theoretical expectation, this looks like something that would be difficult to measure with consumer-grade handheld instruments.  I didn't try warming the transistor up to increase the leakage, perhaps that would yield a more measurable result.

To answer your actual question, I've no reason to think that the mV ranges on the UT139C are anything other than 10M.  Sorry, approximately 10M.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
The voltage is there with exactly zero base current. The current drawn or supplied by the meter can only alter the measurement.

You should repeat the measurements with handheld instrumentation only and in some large garden, far from any buried ac cables. In the middle of a field, maybe.
As for the RF, I live nearby one very powerful radio station and it seems not to have effect on my measurements which, I repeat, are clean, consistent, and also follow the temperature dependence expected from the BE junction.

How far were you from the closest AC cables?
All instruments lie. Usually on the bench.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
The open base voltage is pretty poorly defined. It depend on the leakage current in the transistor, that can vary quite a bit between units (e.g. how clean is the case) and AC signal the circuit does pick up.
With a good small signal transistor the leakage currents can be in the low pA range and lower.
To get a somewhat useful result one would need to have the whole circuit in a shielded box and have a really high impedance meter - more like > 1 T ohm. So not a Multimeter, but more a electrometer with sub pA range input current. One could get way with a low bias OP-amp as a buffer and than the DMM.  One may get away with a bit more leaky meter with a larger transistor, like 2N3055 or 2N3773.

Even than I would not expect the result to match the spice simulation well, as the models may very well not be accurate in such a low relevance situation.  The quality of the model varies, especially with less important properties like leakage currents.

There are a few handheld meters that have a high impedance low voltage range - internally this may be part of the resistance function.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #10 on: March 08, 2023, 08:07:36 pm »
Ok, I've found a 10meg resistor. Tried the volt autorange and the divider halves the supply voltage, confirming 10 meg or so for the input resistance of the UT139C. But when I switch to the mV range I read almost all the voltage (496mV with a 0.5V on my inaccurate supply display) confirming that the UT139C has a gigaohm sort of input resistance on the millivolt range.

I will update this post with detailed measurements of the value of the resistor, the actual supply voltage, and a reasonable estimate for Rmeter).
All instruments lie. Usually on the bench.
 
The following users thanked this post: paulbt

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #11 on: March 08, 2023, 08:27:13 pm »
Ok, I've found a 10meg resistor. Tried the volt autorange and the divider halves the supply voltage, confirming 10 meg or so for the input resistance of the UT139C. But when I switch to the mV range I read almost all the voltage (496mV with a 0.5V on my inaccurate supply display) confirming that the UT139C has a gigaohm sort of input resistance on the millivolt range.

That's interesting, it seems to not be in the documentation.  Or I missed it, it is poor documentation at that.

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #12 on: March 08, 2023, 08:42:37 pm »
The voltage is there with exactly zero base current. The current drawn or supplied by the meter can only alter the measurement.

I understand, I would just like to characterize it with an approximate source impededance, which is of course not truly the model but will give an idea of how much the meter might affect it.  Then you'd have to characterize your meter for input impedance and input bias current.  In the simulation, can you tell me what the C-E current is with the base open, then with say a 5G resistor from B to E?
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #13 on: March 08, 2023, 10:48:59 pm »
The emitter and collector current in the simulation is 1.4 nA and yes, if I simulate with a shunt resistor the voltage drops to a few millivolts with a 10 meg resistor and even with a 3 Gohm resistor. I had it simulated before my first measurement in the cold of the night and I went out there just to prove I would only read a few millivolts (and not the minus 0.6 V of userxxxx on SE). And I was surprised to read hundreds of millivolts, in the ballpark of the simulation with an open base.

I also ran a simulation stepping Rshunt and only when it reaches several teraohms the simulation gives a VBE of hundred of millivolts. This is a voltage that should require an electrometer, and that is why I was surprised of being able to read it with bottom of the barrel instrumentation (the UT139C is a 50 bucks DMM!)
I am still left with the doubt I cooked the transistor because - ahem - before connecting 8.2 V between collector and emitter I placed 8.2 V between emitter and collector and the little guy got lukewarm.
In a few hours I will go out and risk my life in the abysmally low temperature of + 10 °C and try to measure a couple of untouched transistors.
All instruments lie. Usually on the bench.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #14 on: March 08, 2023, 11:43:34 pm »
The emitter and collector current in the simulation is 1.4 nA and yes, if I simulate with a shunt resistor the voltage drops to a few millivolts with a 10 meg resistor and even with a 3 Gohm resistor. I had it simulated before my first measurement in the cold of the night and I went out there just to prove I would only read a few millivolts (and not the minus 0.6 V of userxxxx on SE). And I was surprised to read hundreds of millivolts, in the ballpark of the simulation with an open base.

I also ran a simulation stepping Rshunt and only when it reaches several teraohms the simulation gives a VBE of hundred of millivolts. This is a voltage that should require an electrometer, and that is why I was surprised of being able to read it with bottom of the barrel instrumentation (the UT139C is a 50 bucks DMM!)

Well, don't get sucked into the trap of believing an erroneous measurement just because it looks like what you expect.  I'm not saying that is or isn't happening here, but since you haven't accounted for the input bias current of the meter there's more to be considered.  I'm also a bit dubious that you could take these measurements in any reasonable amount of time because the currents are so low and there is input capacitance to deal with.  But with no experience, I'm open to any possibility and I'll let you explain what you think are the relevant theories.

I started a quick experiment myself and have dragged out the picoammeter for the next steps.  I'm trying analyze it a piece at a time and also sidestep the issues of noise and meter bias that come up with a direct measurement.
 The first step was to look at just the B-E junction with a very low current.  My 8842A has a bias current of approximately -38pA so I connected it to the 2N3904 base and emitter so that it was forward biased.  After some time, the voltage settled at ~229mV.  The thing is, 'some time' was about 90 minutes.  I can lower the voltage a tiny bit by warming the transistor with my fingers. 

Next I'm going to measure the C-E leakage current with the base open.  I'm expecting much less than 1.4nA.  In your simulations with the 10M and 3G resistors in place, what is the leakage current then?  Still 1.4nA or does removing that tiny bit of current from the base reduce it?
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #15 on: March 09, 2023, 12:27:06 am »
OK, 8V C-E gives me unmeasurably low (<2pA) current, 10V yields 2-3pA and 20V is 5pA.  My analog picoammeter only goes down to a 100pA full-range scale, so I'm at the limits of precision.  Putting a 1G resistor from E to B increases the current but the reading is very unstable (40-250pA) and I'm using a really messy long-leaded layin-on-the-bench sort of setup so I suspect this may be a noise issue.  Still the first measurement is way less than 1.4nA.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #16 on: March 09, 2023, 12:54:10 am »
Yes, I am starting to think I got a black swan here. Not one, but two consistent measurement that mimicked the expected voltage.
It is raining now, so my next measurements outside will have to wait (I am made out of salt and I will dissolve myself if I catch some rain). So, I tried to see what I get inside, in the middle of the interference zone, just for the sake of curiosity.

I read about 240 mV at 19.5 °C. (which I was not able to read last time - mind you this is the opposite side of the table)
LTSpice says 328 mV, while Microcap says 215 mV (yes, I know there are variation just like in beta - I use simulations to get an idea of what I should see in the real world - few millivolts, tens of millivolts, hundreds of millivolts...).
Now I'm wondering: is this another coincidence, the third one?

I moved the voltmeter and the dut around, it reads 309 mV.
Move it again, and it's 307 mV
Move it again and it's 270 mV
Move it again and it's 275 mV at 19.4°C
Turn it off and on and it's 275 mv at 19.2°C
Rotate voltmeter and repeat, it's 258.2 mV at 19.1°C
(it should get higher, not lower with lower temps)

And yes, the measurements takes several minutes - about five to ten minutes (I did not check) and the approach is most certainly exponential. I too thought about charging the base capacitance, but is it charging it to the value I was expecting to read and imposed by VCE?

Funny thing, the measures I obtained in the middle of the garden in the dark of the night were almost instantaneous. What was I seeing? I was using normal probes, while now I used clips and no hands...

EDIT: regarding the instrument's bias current, I figured that a few picoamps would not change VBE that much if the 'internally generated' value of hundreds of millivolts was to be seen. It's the impedance of the instrument that changes the configuration, altering the measure.


« Last Edit: March 09, 2023, 01:13:22 am by Sredni »
All instruments lie. Usually on the bench.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #17 on: March 09, 2023, 01:07:31 am »
(it should get higher, not lower with lower temps)

If the leakage current were irrelevant (or even zero if the power supply was off) you would still have the meter input bias current across the B-E diode.  And since I get ~229mV with my meter with just a B-E connection and that voltage drops if you warm the transistor (as it should in that case), perhaps you are seeing the same thing only with a different meter, different bias current and a different transistor.  Try reversing the multimeter leads.  If the effect were not dominated by the meter, then you should see similar results with just a polarity change indicated on the meter. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #18 on: March 09, 2023, 01:35:44 am »
In your simulations with the 10M and 3G resistors in place, what is the leakage current then?  Still 1.4nA or does removing that tiny bit of current from the base reduce it?

Without base resistor, 1.359 nA for IE and IC, with IB = 0A
And VBE = 383 mV at 0°C

With 10 meg base resistor to gnd
IB = -7.9998 pA        it's getting out of the base, into the resistor
IC = 8.0024 pA         but the current in the RC resistor, which shoud be equal is 7.9936 pA
IE = 0.1974 fA
With these values VBE is 79 uV. Microvolts!
It's basically skipping the BE diode. And there are numerical errors here and there.

(with 3 gigaohm VBE becomes 24 mV)

With 70 gigaohm resistor to gnd
IB = -5.072 pA        out of the base, into the resistor
IC = 422.196 pA
IE = 417.119 pA
With these values VBE is 355 mV, almost there to the open base value.
« Last Edit: March 09, 2023, 02:01:16 am by Sredni »
All instruments lie. Usually on the bench.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #19 on: March 09, 2023, 02:01:25 am »
EDIT: regarding the instrument's bias current, I figured that a few picoamps would not change VBE that much if the 'internally generated' value of hundreds of millivolts was to be seen. It's the impedance of the instrument that changes the configuration, altering the measure.

I wouldn't assume anything about the impedance and bias current unless you can characterize them.  I've no idea how high the UNI-T bias current could be and the input 'resistance' is often not really ohmic and for meters with a spec of >10G that can mean >>10G. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #20 on: March 09, 2023, 02:06:47 am »
Try reversing the multimeter leads.  If the effect were not dominated by the meter, then you should see similar results with just a polarity change indicated on the meter.

256 mV at 18.9°C
invert terminals and... it does not change sign!!! Ah no, it's decreasing and needs to go through zero first.
-293 mV at 18.8°C
invert again and...
+251 mV at 18.8°C

Well, not bad for a sub-50bucks multimeter...

Quote
I wouldn't assume anything about the impedance and bias current unless you can characterize them.  I've no idea how high the UNI-T bias current could be and the input 'resistance' is often not really ohmic and for meters with a spec of >10G that can mean >>10G. 

I agree, and that is why I asked if someone with decent instruments had measured it. I tried the divider method but when I measured the voltage at the source I get the same voltage at the meter under test - actually a mV less, since my meters are not accurate enough to allow me to get enough figures to compute Rmeter reliably. It sure is big and, as you say, it is not a resistance but it is probably some FET frontend.
« Last Edit: March 09, 2023, 02:14:00 am by Sredni »
All instruments lie. Usually on the bench.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #21 on: March 09, 2023, 02:42:54 am »
Well, not bad for a sub-50bucks multimeter...

 It sure is big and, as you say, it is not a resistance but it is probably some FET frontend.

It occurred to me that the supply voltages for the front end might be a lot lower for an instrument with 600mV range than one with 20V, so the leakage actually could be pretty low.  Very high end DMMs get down in the single-digit pA range, but those are typically on input ranges of 12-20V.

On the 8842A, following your lead I tried a ~5V supply with a 1G resistor in series.  Shorting the resistor I get 5.0614V, with the resistor in series I get 5.0560V.  That works out to the meter having a 'resistance' at this level of about 937G. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #22 on: March 09, 2023, 09:08:56 am »
Well, not bad for a sub-50bucks multimeter...

 It sure is big and, as you say, it is not a resistance but it is probably some FET frontend.

It occurred to me that the supply voltages for the front end might be a lot lower for an instrument with 600mV range than one with 20V, so the leakage actually could be pretty low.  Very high end DMMs get down in the single-digit pA range, but those are typically on input ranges of 12-20V.

On the 8842A, following your lead I tried a ~5V supply with a 1G resistor in series.  Shorting the resistor I get 5.0614V, with the resistor in series I get 5.0560V.  That works out to the meter having a 'resistance' at this level of about 937G.
The test with just the resistor and shortening it is measuring the input current. With these meters the input current is not directly propoertional to the voltage, there is quite some bias.
The input resistance is more descring how the input current changes with input voltage. So one would have to do the test also with a 2nd voltage (e.g. -5 V, maye 0 V).
 

Offline aristarchus

  • Regular Contributor
  • *
  • Posts: 107
  • Country: 00
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #23 on: March 09, 2023, 10:05:02 am »
Not sure if this be of your help, just letting you know my experience with my UT139C.
I remember that it had difficulties making proper readings in low values (cant recall mV or mA), it was ok otherwise but in low readings it was giving improper results.
Then I found somewhere some pcb photos and compared the part values.
To my surprise one resistor was not the same value on pcb with what was on the photo, the pcb one was 1K IIRC (been a few years and forgot what but I think somone in R33, R4, R5).
When I changed it, the multimeter started working ok.
Could be a bad batch, most likely.

Either way I do not know if this could be your case, just saying my experience with this multimeter and placing the pcb photos I found at the time that helped me.
 

Offline SredniTopic starter

  • Frequent Contributor
  • **
  • Posts: 746
  • Country: aq
Re: What is the internal resistance of the Uni-T UT139C (on the mV range)?
« Reply #24 on: March 10, 2023, 10:04:17 pm »
Thanks for the heads up, but I don't think the meter is the problem here. The (absence of a proper) measurement procedure most surely is. I don't think the meter is giving erroneous values on its own. I measured 0.7 mV both on the UT61E and the UT139C and they both agreed.

Right now I am mostly curious about how on earth I was able to get those two series of fast, clean and repeatable measures the first and second night. It is clear that the meter needs some 10 minutes to stabilize the reading, but I am absolutely certain the measures I did in 0°C and 4 (or was it 7)°C were very fast, but they shouldn't have been.
I tried again, same place, same time of the night (but different day and temperature, but most importantly different probes) and I get the same slow rising to the equilibrium value I get inside the house. I wonder if the relative humidity is playing a dominant role in this. Since the measurements are so dependent on small amounts of charges on the base, an environment that does not allow for charge to accumulate or deposit (if not driven) might explain what I saw, but ...

To add to the list of coincidences, here is a plot showing how - in numberland - the simulated BJT VBE with an open base changes with VCE, while stepping the shunt resistor from 1gigaohm to 71gigaohm in 5gigaohm steps. The first curve on the left is the simulation for actually open base at 18°C.


https://i.postimg.cc/jS5kcmz6/screenshot-10.png

The dots are the measurements I took with 2.9V(*), 4.36V, 8.3V, 16.3V at 19.5°C. Of course there are errors associated with each measure and the actual VBE, like beta, would be different in simulation and actual transistor but... I am getting a trend with VCE that mimicks what should be expected.
(It's true that all exponentials look alike and this might be a different exponential process, but... another coincidence?)

(*) the 2.9 VCE measurement was not exactly reproducible. I was able to get this trend starting from 0 and then going up in voltage, but when I got back (by switching batteries by hand, therefore messing with charges in the worst possible ways) reducing the voltage, the higher voltages were close to the ones shown, but the 2.9V VBE was much higher than that shown.

I plan to redo this measurement with appropriate shielding, clean probes, shorting all terminals before each measurement and in an electrically quiet environment to see if I can get a reproducible curve that fits the expectations. I would appreciate if someone in this forum, with better instruments and setups could try to see if this is a feasible measurement or not.
All instruments lie. Usually on the bench.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf