Author Topic: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401  (Read 2389 times)

0 Members and 1 Guest are viewing this topic.

Offline jorgemefTopic starter

  • Regular Contributor
  • *
  • Posts: 137
  • Country: pt
My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« on: November 22, 2022, 11:27:32 pm »
Hello,

I bought a BM869s uncalibrated and tried to test its accuracy.
I have a LB02A which I also hadn't tested before against a calibrated source and finally got access to a keithley 2000 and a HP 34401, both calibrated back in may 2019.

I measured LB02A against Brymen and Keyhtly and HP.

Here the chart of deviation against set values on the LB02A of the 3 equipments for Volt and mV settings of LB02A.

What can I conclude? Bryman is accurate within 1 mV in volt range and within 50 uV in mV range, and LB02A has an accurate set value between 2,5 and 3 v and voltage progression given b "V = Set V + 0,0004 x SetV - 0,0013"?

Unfortunately I checked with Brightwin and they informed me the LB02A cannot be calibrated by end user, but even if I would try to calibrate via the internal potentiometer which I will not do without understanding what it does, this seems not a linear error. Maybe some error due to DAC?

BR,
Jorge
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #1 on: November 22, 2022, 11:51:58 pm »
Are you measuring the meters all three at once in parallel or one at a time?  I'd be concerned about the larger than expected discrepancy between the Keithley and the HP. 
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 jorgemefTopic starter

  • Regular Contributor
  • *
  • Posts: 137
  • Country: pt
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #2 on: November 23, 2022, 07:36:06 am »
It was measured one at a time, after warm-up of 15 minutes in an environment of 24ÂșC.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #3 on: November 23, 2022, 03:31:53 pm »
I'd try that again, all 3 in parallel.  One at a time allows your source to drift as you go. Your two 6.5 digit meters should agree to better than 30ppm or so unless you are very unlucky.  Also a 15-minute warmup is very short, in my experience the 34401A would like a 3-hour warmup for highest stability, no idea on the other 3 instruments.
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.
 
The following users thanked this post: jorgemef

Offline jorgemefTopic starter

  • Regular Contributor
  • *
  • Posts: 137
  • Country: pt
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #4 on: November 23, 2022, 05:03:29 pm »
Thank you for the feedback. Will try again in a few days once I get access to the calibrated multimeters as per your suggestion.
 

Offline J-R

  • Super Contributor
  • ***
  • Posts: 1319
  • Country: us
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #5 on: November 23, 2022, 07:43:08 pm »
In my experience, even immediately after power-on, those bench DMMs are not going to be 250-300 counts off, closer to 25-30.

I would interpret the graphs as showing the BM869s is doing quite well, but the LB02A is not in the same class.  I would suggest the PDVS2mini instead as a precision voltage reference:
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-2-mini-detail
 

Offline AVGresponding

  • Super Contributor
  • ***
  • Posts: 4931
  • Country: england
  • Exploring Rabbit Holes Since The 1970s
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #6 on: November 24, 2022, 06:37:39 am »
I'd try that again, all 3 in parallel.  One at a time allows your source to drift as you go. Your two 6.5 digit meters should agree to better than 30ppm or so unless you are very unlucky.  Also a 15-minute warmup is very short, in my experience the 34401A would like a 3-hour warmup for highest stability, no idea on the other 3 instruments.

It depends a little bit on ambient temperature, but 15 minutes is way shorter than I would do. The K2000 needs a good longish time just like the 34401, just like any decent 6.5 digit meter I suspect, barring those that use soft power buttons and keep the reference hot.
nuqDaq yuch Dapol?
Addiction count: Agilent-AVO-BlackStar-Brymen-Chauvin Arnoux-Fluke-GenRad-Hameg-HP-Keithley-IsoTech-Mastech-Megger-Metrix-Micronta-Racal-RFL-Siglent-Solartron-Tektronix-Thurlby-Time Electronics-TTi-UniT
 

Offline J-R

  • Super Contributor
  • ***
  • Posts: 1319
  • Country: us
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #7 on: November 24, 2022, 10:34:08 am »
Just tested my calibrated HP 34401A (cal data @ 72.8F) and un-calibrated but known good Keithley 2015 THD. Ambient was 63.7F.   All tests run within 10 seconds of a completely cold power-on using a fully settled PDVS2mini:

34401A was 48 counts high at 100mV, 11 counts low at 10V.

2015 THD was 32 counts low at 100mV and 14 counts low at 10V.

I've used both of these quite a bit and just have not found them to require extensive time to settle, unless I'm using every last digit to compare against calibrated references in volt-nut mode.  In that case I also break out the Keithley 2010, get the room to 72F and let everything simmer for half the day...

Anyway, this is just one data point, perhaps others can post their results.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #8 on: November 24, 2022, 05:21:57 pm »
Anyway, this is just one data point, perhaps others can post their results.

Yours seem typical.  My 34401A is ~20 counts low right at power on, depending on room temp, improves to ~15 counts within a minute or so and then slowly creeps up to being pretty much dead on after 2-3 hours.  I haven't correlated any of that to room temp, but I'm sure it matters some what.

Fluke 8505A starts out 3 counts low, goes 2-3 counts high during warmup, pretty much dead on at 1 hour.  Fluke 8846 settles pretty much perfectly in ~30 seconds from a cold boot, maybe 2-3 counts additional settling over a few minutes if the temperature is more than 5 degrees away from 72F.

Obviously these amounts don't matter much in the OPs context except that I'd like to see his two calibrated 6.5 digit meters correspond as closely as possible. 
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 jorgemefTopic starter

  • Regular Contributor
  • *
  • Posts: 137
  • Country: pt
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #9 on: December 18, 2023, 10:51:35 am »
Hello,

I hacked my LB02a to fine tune the reference to 2.5 more preciselly and also adjust its output range at 10v.
I noticed 200uV spikes in the 10V output in some readings which I could notice with my kheithly2000 but could not see before with the brymen.
I noticed the readings only some times, and if the sampling rate is high (0,1 samples per second).

Is this normal? I think is allways there and appearing when sampling rate is in sync with the spikes rate and does not results from my hack, but just wondering.
On the keythly sometimes the average reaches 300uv with these spikes. So maybe is normal in these process calibrators but just wondering as if such this is an anoying 300uV noise which will be always there.

BR,
Jorge

« Last Edit: December 18, 2023, 10:56:14 am by jorgemef »
 

Offline jorgemefTopic starter

  • Regular Contributor
  • *
  • Posts: 137
  • Country: pt
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #10 on: December 18, 2023, 07:05:37 pm »
Hi,

With additional measurements I tracked the noise to something introduced through the LM358.
DAC output seems very litle noise (3-4 uv) and feeds OP07C in+ and LM358 IN+b through 10K resistors.
Adding voltage measured on these opamp for the different voltage settings. Noise pops on the LM358 from time to time and vanishes again after some time.
Well, 300uV of noise even at 1v output is still within tolerance range of 0,025%. :)

Cheers,
Jorge
« Last Edit: December 18, 2023, 07:09:22 pm by jorgemef »
 

Offline GnomeZA

  • Regular Contributor
  • *
  • Posts: 125
  • Country: za
Re: My LB02A vs my new Brymen BM869s vs Keithley 2000 vs HP 34401
« Reply #11 on: December 19, 2023, 12:42:39 am »
Just tested my calibrated HP 34401A (cal data @ 72.8F) and un-calibrated but known good Keithley 2015 THD. Ambient was 63.7F.   All tests run within 10 seconds of a completely cold power-on using a fully settled PDVS2mini:

34401A was 48 counts high at 100mV, 11 counts low at 10V.

2015 THD was 32 counts low at 100mV and 14 counts low at 10V.

I've used both of these quite a bit and just have not found them to require extensive time to settle, unless I'm using every last digit to compare against calibrated references in volt-nut mode.  In that case I also break out the Keithley 2010, get the room to 72F and let everything simmer for half the day...

Anyway, this is just one data point, perhaps others can post their results.

My experience mirrors yours.  I have a AD584 that is powered on for a few months now from battery and my K2000 drops about 5 counts from cold to powered on for hours.  Getting this exact same experience now months later.  I've even written down the "cold" and "hot" measurements.

I know this isn't the most extensive test, but it is the best I have to measure drift.

I mention months because what this is in an apartment in the southern hemisphere that has gone from winter to summer and low(er) humidity to high(er) humidity.  So fairly remarkable result as far as the K2000 is concerned really.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf