Author Topic: Suggestions for a millivolt DMM  (Read 18776 times)

0 Members and 1 Guest are viewing this topic.

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2205
  • Country: mx
Re: Suggestions for a millivolt DMM
« Reply #25 on: October 22, 2019, 05:18:19 pm »



Hey! The scale goes up to 11.
My apologies to all "Spinal Tap" fans.
 

Offline 001

  • Super Contributor
  • ***
  • Posts: 1170
  • Country: aq
Re: Suggestions for a millivolt DMM
« Reply #26 on: October 22, 2019, 05:46:43 pm »
I try my AN8009.
I feed a 50Hz sine wave so that the AN8009 shows about 50 µV.
Than i double the input and the AN8009 shows about 75 µV
My AN8008 shows with the above settings about 70 µV and 90 µV.

WHY?  :-//
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Suggestions for a millivolt DMM
« Reply #27 on: October 22, 2019, 06:27:55 pm »
WHY?  :-//

DIFFERENT NOISE FLOORS


P.S. Please stop shouting in your posts, alcohol fuelled or not.
« Last Edit: October 22, 2019, 06:42:44 pm by Gyro »
Best Regards, Chris
 
The following users thanked this post: 001

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Suggestions for a millivolt DMM
« Reply #28 on: October 22, 2019, 06:30:19 pm »
The venerable Datron 1030A True RMS voltmeter has a noise floor of ~80uV wideband (1MHz) and <30uV with the 10kHz band limit filter engaged. They're hard to find though.
Best Regards, Chris
 

Offline macboy

  • Super Contributor
  • ***
  • Posts: 2252
  • Country: ca
Re: Suggestions for a millivolt DMM
« Reply #29 on: October 22, 2019, 06:31:06 pm »
I try my AN8009.
I feed a 50Hz sine wave so that the AN8009 shows about 50 µV.
Than i double the input and the AN8009 shows about 75 µV
My AN8008 shows with the above settings about 70 µV and 90 µV.

WHY?  :-//
Because: Noise

The reading is a summation of the inherent noise of the meter plus the input signal. These add as root-sum-of-squares.
If the input signal was about 32.2 mV initially, then double that (64.4 mV) for the second test, and the inherent noise floor of the meter is 38.2 mV, then the readings make sense.
test 1: sqrt( 32.22 + 38.22 ) = 49.96 ~= 50 mV
test 2: sqrt( 64.42 + 38.22 ) = 74.87 ~= 75 mV

If the signal is at least 5x as high as the noise, then the noise makes little difference:
sqrt( 2002 + 38.22 ) = 203.6 mV
i.e. only 1.8% higher than the input of 200 mV. Note that although the meter contributes 38.2 mV of noise in addition to the signal, the reading is only 3.6 mV higher, not 38 mV higher. At a signal level of 400 mV, the reading is 401.8 mV, which is only 1.8 mV high, or less than 0.5% error.
 
The following users thanked this post: 001, Sultanpepper123

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Suggestions for a millivolt DMM
« Reply #30 on: October 22, 2019, 07:20:43 pm »
Noise is only part of the problem. To a certain limit the offset from noise could be compensated in software. With many RMS implementation another big problem is the precision rectification that does not work reliably anymore if the amplitude is too small. This also includes most (if not all) analog true RMS chips. A first consequence of lower amplitude is slower detection of the sign and thus reduced BW at low amplitude. So the simple noise calculation does no longer work with a real RMS converter.

There are ways around this with a square law detector like used in RF power sensors or digital RMS. However I am still looking for a cheap meter with digital RMS. The square law detectors tend to be limited accuracy and also have some weakness at low amplitude.

The main point for the small voltages is having suitable amplification / ranges.

For a simple test, I would reduce the amplitude from the generator with a divider and this way get a low voltage test signal. Even a crude divider tends to be better than comparing 2 meters way outside the specified range.
 
The following users thanked this post: Sultanpepper123

Online Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Suggestions for a millivolt DMM
« Reply #31 on: October 22, 2019, 08:18:29 pm »
Hello,

I made a AC linearity comparison on a K2000, a HP34401A in 100mV AC range and my scope.
The clear winner is the scope (except for the noise floor at zero).

with best regards

Andreas
 
The following users thanked this post: egonotto, thm_w, Henrik_V, Sultanpepper123

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Suggestions for a millivolt DMM
« Reply #32 on: October 22, 2019, 08:35:05 pm »
« Last Edit: October 22, 2019, 08:38:06 pm by Gyro »
Best Regards, Chris
 
The following users thanked this post: Sultanpepper123

Offline Terry01

  • Frequent Contributor
  • **
  • Posts: 907
  • Country: scotland
Re: Suggestions for a millivolt DMM
« Reply #33 on: October 22, 2019, 08:36:36 pm »
Brymen BM867S or BM869s ?
Sparks and Smoke means i'm nearly there!
 

Online BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Suggestions for a millivolt DMM
« Reply #34 on: October 23, 2019, 03:19:31 am »
Hey! The scale goes up to 11.

Anticipating 10% overshoot ?

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Suggestions for a millivolt DMM
« Reply #35 on: October 23, 2019, 06:52:40 am »
Brymen BM867S or BM869s ?

I thought of those as well but the lowest range is 500mV with 0.3% accuracy, 10uV resolution.

Granted they seem to beat their rated specs by a wide margin but somebody would need to try it. I suspect that if they don't go lower than 500mV then it's for a reason...

 

Offline Terry01

  • Frequent Contributor
  • **
  • Posts: 907
  • Country: scotland
Re: Suggestions for a millivolt DMM
« Reply #36 on: October 23, 2019, 07:10:50 am »
Hello everyone, I'm having a hard time in selecting a suitable multi meter that should measure AC voltages as low as 100 micro volts @ 50 Hz with steps of 50 microvolts. I've searched for quite some time and here they seem to have a problem in that range i suppose. My research shows that i can use Fluke's 8808A, 8845A, 287 or 289 , I may be betting on too high specs or vice versa. I would appreciate if someone sheds some light on some suggestions.

EDIT: This might be useful to some who tried to reach the reference but found out the link is dead, a screenshot from the deadlink of Gellar Labs link which illustrates the dead zone of the agilent 34401A DMM can be found here

Regards

Quote from: Terry01 on Yesterday at 09:36:36 pm

Brymen BM867S or BM869s ?


I thought of those as well but the lowest range is 500mV with 0.3% accuracy, 10uV resolution.

Granted they seem to beat their rated specs by a wide margin but somebody would need to try it. I suspect that if they don't go lower than 500mV then it's for a reason...

I have the 867s but don't have anything that goes down to uV to use as a test  :-\
Sparks and Smoke means i'm nearly there!
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Suggestions for a millivolt DMM
« Reply #37 on: October 23, 2019, 07:54:44 am »
I have the 867s but don't have anything that goes down to uV to use as a test  :-\

Me too.

(actually 857s but it's the same specs)

 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6446
  • Country: hr
Re: Suggestions for a millivolt DMM
« Reply #38 on: October 23, 2019, 08:10:14 am »
At 50 Hz attenuator is just 3 plain resistors...
 

Offline Terry01

  • Frequent Contributor
  • **
  • Posts: 907
  • Country: scotland
Re: Suggestions for a millivolt DMM
« Reply #39 on: October 23, 2019, 08:11:34 am »
I have the 867s but don't have anything that goes down to uV to use as a test  :-\

Me too.

(actually 857s but it's the same specs)

I'm sure one of the guys here will have the meter and the necessary equipment to test uV and they can let us know if it's up to the task or not. If the OP only needs the said resolution for 1 project or whatever the Brymen has to be one of, if not the best $$$'s solution out there?

I will still love mine even if it's not up to the task! The Brymen's are a quality meter.   ;D
Sparks and Smoke means i'm nearly there!
 

Offline Terry01

  • Frequent Contributor
  • **
  • Posts: 907
  • Country: scotland
Re: Suggestions for a millivolt DMM
« Reply #40 on: October 23, 2019, 08:18:45 am »
At 50 Hz attenuator is just 3 plain resistors...

I have a sig gen that can put out 50Hz but 1 mV is as low as it goes and I doubt that could be trusted as it's just 1 of those cheap feeltech things  :-[
Sparks and Smoke means i'm nearly there!
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Suggestions for a millivolt DMM
« Reply #41 on: October 23, 2019, 08:20:54 am »



Hey! The scale goes up to 11.
My apologies to all "Spinal Tap" fans.

I need one now!
 

Online BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Suggestions for a millivolt DMM
« Reply #42 on: October 23, 2019, 08:32:27 am »



Hey! The scale goes up to 11.
My apologies to all "Spinal Tap" fans.

I need one now!

Quick grab them while they are still reasonable priced, before Dave makes a new video about this.  :palm:

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Suggestions for a millivolt DMM
« Reply #43 on: October 23, 2019, 08:40:50 am »
I have a sig gen that can put out 50Hz but 1 mV is as low as it goes and I doubt that could be trusted as it's just 1 of those cheap feeltech things  :-[

Grab a potentiometer and hook it up. See what happens.

How low can you go before it visibly/obviously goes wrong?
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Suggestions for a millivolt DMM
« Reply #44 on: October 23, 2019, 09:11:42 am »
Here is a test with the Brymen BM869s and an Leader audio tester.
The setting on the audio tester is 0 dB = 1 V at 600 Ohm and 50 Hz. The attenuation is activated with 80 dB. The display is fullscale 1.5 mV. 100 µV are displayed.

 :)

I see you still have some unpushed buttons, can you increase the attenuation to 86dB and see if it says 50uV?

What does it read if you use the standard multimeter probes instead of fancy connectors?  :popcorn:
 
The following users thanked this post: Sultanpepper123

Offline evava

  • Regular Contributor
  • *
  • Posts: 172
  • Country: cz
Re: Suggestions for a millivolt DMM
« Reply #45 on: October 23, 2019, 09:31:31 am »
Here is a test with the Brymen BM869s and an Leader audio tester.
The setting on the audio tester is 0 dB = 1 V at 600 Ohm and 50 Hz. The attenuation is activated with 80 dB. The display is fullscale 1.5 mV. 100 µV are displayed.

Peter

Is it possible? Reading RMS value at 0.02% of range?
When manufacturer states in manual accuracy from 5% to 10% of range  is 0.3%+100 counts?
What for are manuals then?
 

Offline Terry01

  • Frequent Contributor
  • **
  • Posts: 907
  • Country: scotland
Re: Suggestions for a millivolt DMM
« Reply #46 on: October 23, 2019, 09:37:32 am »
Thanks Peter.

Your set up is way better than my "feeltech special"  :)
Sparks and Smoke means i'm nearly there!
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Suggestions for a millivolt DMM
« Reply #47 on: October 23, 2019, 09:44:42 am »



Hey! The scale goes up to 11.
My apologies to all "Spinal Tap" fans.

I need one now!

Quick grab them while they are still reasonable priced, before Dave makes a new video about this.  :palm:
You can buy these new. Lowest price (180 euro) I could find is from TME:
https://www.tme.eu/be/nl/details/gvt-417b/analoge-multimeters/gw-instek/
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Suggestions for a millivolt DMM
« Reply #48 on: October 23, 2019, 11:10:43 am »
Is it possible? Reading RMS value at 0.02% of range? When manufacturer states in manual accuracy from 5% to 10% of range  is 0.3%+100 counts?

Yes.

Those numbers aren't "garanteed inaccuracy", they're the outer limits of the bell curve on the production line.

Those bell curves also allow for extremes of temperature, and temperature makes a big difference to handheld meters with this many digits. If you're measuring at the same temperature as the factory's calibration room then you'll get much better accuracy than if you're measuring at freezing (or hot) temperatures, which is where those outer limits of accuracy will come into play.

Bottom line: In practice, and at ordinary temperatures, these Brymens will do much better than those specs
. I've compared mine to a fancy 6.5 digit bench multimeter and it matched it perfectly.

What for are manuals then?

To know when there's something seriously wrong with the meter.

(and when the manufacturer will accept a return under warranty).
« Last Edit: October 23, 2019, 12:11:58 pm by Fungus »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16560
  • Country: 00
Re: Suggestions for a millivolt DMM
« Reply #49 on: October 23, 2019, 11:17:57 am »



Hey! The scale goes up to 11.
My apologies to all "Spinal Tap" fans.

I need one now!

Quick grab them while they are still reasonable priced, before Dave makes a new video about this.  :palm:
You can buy these new. Lowest price (180 euro) I could find is from TME:
https://www.tme.eu/be/nl/details/gvt-417b/analoge-multimeters/gw-instek/

Jeez, that's massive. It looks like a little handheld meter in the photo - not so!

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf