Author Topic: HP 3458A ; A/D board spare part availibilty and cost  (Read 45404 times)

0 Members and 2 Guests are viewing this topic.

Offline maat

  • Regular Contributor
  • *
  • Posts: 144
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #25 on: April 06, 2020, 07:13:05 am »
I did have a sick A3 board as well, which gave itself away by its awful 0.85ppm/K tempco. At first I assumed, that it was just luck and the board simply had a slightly higher tempco. To get an idea about the tempco I started logging the internal temperature. Since the the TEMP? Command has a crappy resolution I opted to use the higher resolution ACAL DCV + CAL? 175 command, but the resolution is still no god. Ifinally placed a temperature sensor inside the case, that gave similar results and response time. Anyway, since I had the CAL? 72 values now, I plotted them and realised they drifted.

The attached plots are post processed. I did a fit with two parameters (time and temperature). The drift plot has the temperature component removed and the tempco plot has the drift component removed.

I can also provide the python scripts to record and plot/fit the data, All I need to do is brush them up a little.

End of the story: 3 Months later the DMM died with a
114,"SYSTEM ERROR -- balanced rundown convergence" during further testing and
202,"HARDWARE FAILURE -- "SLAVE TEST: CONVERGENCE"  after self test.

Lucky me, I had opted for a warranty extension, which KS offered during the last calibration  :-DD
 
The following users thanked this post: TiN, felixd, MiDi

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #26 on: April 06, 2020, 07:32:17 am »
Since the the TEMP? Command has a crappy resolution I opted to use the higher resolution ACAL DCV + CAL? 175 command, but the resolution is still no god. Ifinally placed a temperature sensor inside the case, that gave similar results and response time. Anyway, since I had the CAL? 72 values now, I plotted them and realised they drifted.

The attached plots are post processed. I did a fit with two parameters (time and temperature). The drift plot has the temperature component removed and the tempco plot has the drift component removed.




Hello,
have you placed the temperature sensor directly on A3, or on U180?

Frank
 

Offline maat

  • Regular Contributor
  • *
  • Posts: 144
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #27 on: April 06, 2020, 08:08:28 am »

Hello,
have you placed the temperature sensor directly on A3, or on U180?

Frank

Unfortunately no, because I was trying to mimic behaviour of the "TEMP?" command. I did that, because I assumed, since the tempco was still in spec, that I had to live with it and that I could correct for it using the TEMP? command, once I had a good understanding of how bad it was. Also due to the warranty extension, I did not want to remove the stickers at that point.

The sensor (Fluke 5611T, PTFE coated 10k thermistor) was placed on the the inguard metal shield (side) and fastened using Kapton tape. This was as close as I could get with a pair of tweezers and gave near identical results to the TEMP? command, albeit better resolution.
 

Offline voltampereTopic starter

  • Regular Contributor
  • *
  • Posts: 58
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #28 on: April 06, 2020, 09:32:38 am »
voltampere
I don't see much drift in your hp3458z plot. Just normal settling for warm-up time.  :-// So don't understand where you get conclusion that your A3 is bad.

TIN, the meter was running already for at least two weeks or so, if i remember correctly.
Still warmup drift???

Of course i can not be 100 % sure that the board is out of spec.
Maybe i was too impatient. Maybe i have to let it run for half a year and see where it ends.
Or try to accelerate the drift process with some baking in the oven?

Regards

Joerg
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14080
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #29 on: April 06, 2020, 09:52:27 am »
No need to stress the board, as we do not know where the acceptable limit is for the higher temperature. It would be more like watch the ADC for a few weeks. The 2.5 ppm in 4 days is still boarder-line, just a little over the limit.  The typical bad ADCs seem to be way higher drift.
I think the 0.43 ppm/day limit is more like a value that all good ADCs still meet this limit. There can be temperature and maybe humidity effects so that it may just need more time to get a reliably value for the drift. So maybe better read it as 6 ppm in less than 2 weeks.

The ADC gain drift is only one symptom of a drifty U180.  The actual real problem is more with other resistor ratios. The CAL72 is just the one that is easiest to measure. So a drifty A3 board can also show higher DNL errors, more TC or similar effects.
 
The following users thanked this post: e61_phil, MiDi

Offline maat

  • Regular Contributor
  • *
  • Posts: 144
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #30 on: April 06, 2020, 12:13:35 pm »
In my opinion, the 0.43 ppm/day drift rate is a rather weak criterion, because a good ADC definitely drifts less 0.1 ppm/day. The SN-18 unfortunately is a bit superficial on this issue. As was KS, when I nagged them about the issue.

The problem is, that in order to properly determine a drift you need
  • a failry stable lab, so you don't see any 5 or 10 K excursions, have humidity going on a rampage when the weather turns and so on
  • to to have to instrument turned on 24/7 for at least a month or two
  • to continuously monitor both temperature and the calibration constants
  • apply a 2d fit to determine both tempco and drift
Using the method suggested by HP/Agilent to determine the drift reduces the problem to taking two samples 7 days apart, without proper temperature monitoring, therefore the error margins are fairly high. A 1 K drift is typicall deemed acceptable, and T_cal ± 5K is the normal working range. On the other hand continuously logging will take the instrument out of service, which might not be acceptable. From the style of writing I would also assume, that HP did not anticipate, that all users are capable or willing to properly extract the drift of their instrument. So my guess would be, that 0.43 ppm/d drift is a dying DMM slapping in your face and not a marginal situation, but it can be easily measured, hence not many false positives.
 
The following users thanked this post: e61_phil, MiDi

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #31 on: April 06, 2020, 01:03:44 pm »
In my opinion, the 0.43 ppm/day drift rate is a rather weak criterion, because a good ADC definitely drifts less 0.1 ppm/day. The SN-18 unfortunately is a bit superficial on this issue. As was KS, when I nagged them about the issue.

The problem is, that in order to properly determine a drift you need
  • a failry stable lab, so you don't see any 5 or 10 K excursions, have humidity going on a rampage when the weather turns and so on
  • to to have to instrument turned on 24/7 for at least a month or two
  • to continuously monitor both temperature and the calibration constants
  • apply a 2d fit to determine both tempco and drift
Using the method suggested by HP/Agilent to determine the drift reduces the problem to taking two samples 7 days apart, without proper temperature monitoring, therefore the error margins are fairly high. A 1 K drift is typicall deemed acceptable, and T_cal ± 5K is the normal working range. On the other hand continuously logging will take the instrument out of service, which might not be acceptable. From the style of writing I would also assume, that HP did not anticipate, that all users are capable or willing to properly extract the drift of their instrument. So my guess would be, that 0.43 ppm/d drift is a dying DMM slapping in your face and not a marginal situation, but it can be easily measured, hence not many false positives.

This 0.43ppm/d criterion is just the interpolation of the timely drift specification of about 8ppm/year, down to 24h.

If you apply a square root function for the ageing of the LTZ1000A reference: dU/Uref ~ 8.2ppm * SQRT(t[h] / 8760 hr), then for 24h you get 0.43ppm drift, 90days would be 4.1ppm, and 2 years 11.6ppm, about the values given in the specification.

There's no official specification for the short term stability of this instrument, neither for the T.C. of the A/D converter w/o ACAL.
Therefore no wonder that KS is close-lipped about these (internal quality) criterion.

'Normal' or 'healthy' values are a vanishing timely drift for CAL?72, practically meaning you can't barely distinguish it from fluctuations, or T.C. effects.
And a T.C. of about +/- 0.35 ppm/K for U180, and +/- 0.5ppm/K in total, i.e. ~ +/- 0.15ppm/K for the LTZ1000A, regarding these T.C.s, see specification. 

My 3458A behaves exactly like that, therefore an absolutely stable room temperature dT << 1K is required.

Frank   
 

Offline maat

  • Regular Contributor
  • *
  • Posts: 144
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #32 on: April 06, 2020, 03:32:19 pm »
This 0.43ppm/d criterion is just the interpolation of the timely drift specification of about 8ppm/year, down to 24h.

Absolutely true, which is, why I called it "weak", because from the two datapoints, that one is supposed to take, a correct interpretation is difficult.

Regarding my old A3 board, I monitored it over the course of a month or so and the drift was linear with few wiggles due to the temperature compensation not being 100% accurate. All I wanted to say is, that one should monitor the drift over some time. Typically after about two days the statistics were good enough and they fit converged nicely.

The new (I don't know if it is one of the all new, or just refurbished. I haven't had a look at it yet) A3 board  hardly drifts and the tempco is excellent. I had a chance to measure it while I was measuring a Vishay 300144ZT divider in ratio mode and the air con went offline and the the emergency mode kicked in pulling the temperature down. The absolute ratio was 10 V * 17/(20+17) = 4.59 V and the divider is not responsible for the drop, because it was in the isothermal chamber during that time.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #33 on: April 06, 2020, 11:51:45 pm »
Quote
TIN, the meter was running already for at least two weeks or so, if i remember correctly.
Still warmup drift???
I would consider 3458A coming from cold essentially in infancy first 3 month. Somewhat useful data can be gathered after monitoring these 3 month in 24/7 operation. When you trying to split sub-ppm deviations on voltage, look in wrong direction may upset the poor instrument :).

Quote
Maybe i was too impatient. Maybe i have to let it run for half a year and see where it ends.
Just continue monitoring in stable temperature conditions and see where you end up after 4 weeks.

Quote
Or try to accelerate the drift process with some baking in the oven?
Absolutely not. Why ruin possibly good just unwarmed board with overstress? Baking anything in oven is great way to certainly kill what is not dead yet.
Don't take example from my experiments with freezing and heating, because I was 100% sure that unit is very dead, as it would not even sample anything due to integration errors. So unlike you, I had nothing to loose.
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 
The following users thanked this post: niner_007

Offline voltampereTopic starter

  • Regular Contributor
  • *
  • Posts: 58
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #34 on: April 07, 2020, 10:13:42 am »
Well, if i speak of "baking" it would mean that i put in in the oven at 60-70 degrees celsius!
Not like 100 deg or above. Just to help it with accelerated drying.

But you guys convinced me!
Before i start playing desperate experiments i will give it another shot in a marathon warmup watch.
Though i really doubt the success. I bought this meter about a year ago.
And i am pretty sure that i had it on for extended periods a couple of times already.
Anyway, it sure does not hurt, and costs only electricity ( maybe 15-20 € ).

So i should be back here in a bit more than a month and tell about the outcome.

Thank you all so far..

Regards

Joerg
 

Offline niner_007

  • Frequent Contributor
  • **
  • Posts: 256
  • Country: us
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #35 on: May 18, 2020, 07:11:31 am »
I don’t know how it’s in Europe, but in US, absolutely smooth, I had the part number, they explained all other part numbers, and options, and they could even order from stock outside US, I had ordered 4 IIRC. It costed $1700 - $2000 brand new, and $700 for the refurbished one.
« Last Edit: May 18, 2020, 07:24:04 am by niner_007 »
 

Offline ap

  • Frequent Contributor
  • **
  • Posts: 282
  • Country: de
    • ab-precision
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #36 on: May 18, 2020, 10:31:47 am »
Was that recently? I bet not. It has changed a few months ago (at least here), they stopped selling these outside of service orders. Would be interesting if thats not the case in the US at this time.
Metrology and test gear and other stuff: www.ab-precision.com
 

Offline CalMachine

  • Frequent Contributor
  • **
  • Posts: 477
  • Country: us
  • Metrology Nut
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #37 on: May 18, 2020, 07:26:16 pm »
I don’t know how it’s in Europe, but in US, absolutely smooth, I had the part number, they explained all other part numbers, and options, and they could even order from stock outside US, I had ordered 4 IIRC. It costed $1700 - $2000 brand new, and $700 for the refurbished one.

I just tried to order the 3 ADC boards and they told me none of them are available.  This was, literally, 10 minutes ago
All your volts are belong to me
 

Offline niner_007

  • Frequent Contributor
  • **
  • Posts: 256
  • Country: us
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #38 on: May 19, 2020, 01:35:42 am »
oh wow, good that I have my stash then :)
 

Offline niner_007

  • Frequent Contributor
  • **
  • Posts: 256
  • Country: us
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #39 on: May 28, 2020, 08:53:46 pm »
I can confirm this as well :-[ I have a 3458A that reports ADC connection errors, I'm not going to use an ADC from my stash for this, I started a contract with Keysight. I guess overall, it does not seem that expensive and it's reasonable. I can only guess what this would be with Fluke  :--
 

Offline e61_phil

  • Frequent Contributor
  • **
  • Posts: 962
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #40 on: May 28, 2020, 09:01:36 pm »
I can confirm this as well :-[ I have a 3458A that reports ADC connection errors, I'm not going to use an ADC from my stash for this, I started a contract with Keysight. I guess overall, it does not seem that expensive and it's reasonable. I can only guess what this would be with Fluke  :--

Hard to judge. Maybe we will know that better sometime. As soon as Fluke also build ADCs in their instruments which dies like flies.
 

Offline niner_007

  • Frequent Contributor
  • **
  • Posts: 256
  • Country: us
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #41 on: May 29, 2020, 05:46:50 am »
I can confirm this as well :-[ I have a 3458A that reports ADC connection errors, I'm not going to use an ADC from my stash for this, I started a contract with Keysight. I guess overall, it does not seem that expensive and it's reasonable. I can only guess what this would be with Fluke  :--

Hard to judge. Maybe we will know that better sometime. As soon as Fluke also build ADCs in their instruments which dies like flies.
Mine is not drifting, it is dead  :-//
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14080
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #42 on: May 29, 2020, 08:27:51 am »
If the meter report a connection error, this may be a different failure. Best case just a loose fiber optic connector. There were also reports about those fiber optic parts to fail. So this could be a repairable failure that does not involve the custom U180 chip.

AFAIK the failing U180 first causes drift (and possibly some INL errors - so compensation with more frequent ACAL is not a real option) and later the convergence error, when the error gets too large.

Having a new ADC board installed at Keysight and not done by the user is not so bad, if the extra costs are not outrageous. After such a repair one would need a new calibration anyway, so in most cases the meter would have to be send out anyway.
 

Offline IanJ

  • Supporter
  • ****
  • Posts: 1580
  • Country: scotland
  • Full time EE biz & Youtuber
    • IanJohnston.com
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #43 on: May 30, 2020, 12:15:46 pm »
This 0.43ppm/d criterion is just the interpolation of the timely drift specification of about 8ppm/year, down to 24h.
.
.
.
My 3458A behaves exactly like that, therefore an absolutely stable room temperature dT << 1K is required.
Frank

I have been logging CAL? 72 to a spreadsheet on mine since Oct. 2018 and which outputs Drift PPM (referenced to Day 1). I have 30 samples across 562 days. The conclusion was that........there was no conclusion!
With my workshop temp varying and the resultant Temp? ranging from 32.6 to 38.3 DegC it's really hard to make anything of the results. The only thing I have been able to read from the results is pairing up similar/identical Temp? readings from different days and comparing the Drift PPM figures, and if they are pretty close then I assume my board is ok.

I often wondered if there was a way to make a big enclosure/box for the 3458A to sit in that is perfectly temperature controlled.........but probably too much trouble.

Ian.
Ian Johnston - Manufacturer of the PDVS2mini & author of the free WinGPIB app.
Website & Online Shop: www.ianjohnston.com
YT Channel (electronics repairs & projects): www.youtube.com/user/IanScottJohnston, Twitter (X): https://twitter.com/IanSJohnston
 
The following users thanked this post: Grandchuck

Offline MiDi

  • Frequent Contributor
  • **
  • Posts: 599
  • Country: ua
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #44 on: May 30, 2020, 02:18:25 pm »
If the drift is low and nearly constant you should be able to calculate and compensate for TC.
I did that in my ACAL measurements since Dec 2018 and included CAL 175 (high-res Cal-Temperature).
Now I know my unit has ~ -0.33ppm/°C TC and settled to drift rate ~ -0.03ppm/day for ADC (CAL 72) - if the filter is cleaned regularly :palm: - see attached chart.

Edit: This is a follow up to my post 2 months earlier - probably I do not bore you with it for longer period anymore ;)
« Last Edit: May 30, 2020, 02:32:08 pm by MiDi »
 
The following users thanked this post: chuckb, wolfy007, Grandchuck

Offline niner_007

  • Frequent Contributor
  • **
  • Posts: 256
  • Country: us
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #45 on: July 19, 2020, 11:46:00 am »
My repaired/calibrated 3458A is coming back from Keysight on Monday, exciting! :)
 

Online dietert1

  • Super Contributor
  • ***
  • Posts: 2023
  • Country: br
    • CADT Homepage
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #46 on: July 20, 2020, 08:46:00 am »
@ Midi: Since i am not familiar with 3458A, could you write a sentence what the measurement is? As far as i understand this is not a comparison between two references, but a cross check where the ADC measures its own reference. Is that correct? The second curve with larger fluctuations, is that the temperature sensor?

Regards, Dieter
« Last Edit: July 20, 2020, 08:47:41 am by dietert1 »
 

Offline MiDi

  • Frequent Contributor
  • **
  • Posts: 599
  • Country: ua
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #47 on: July 20, 2020, 09:43:48 am »
The diagramm shows the cal values stored from instruments calram for last (A)CAL, here it is executed every hour.
Cal 72 is gain correction, which afaik is calculated from measurement of the internal voltage reference - light grey line in diagramm as deviation in ppm.
Cal 175 gives internal temperature - blue line.
The black line is temperature corrected Cal 72.


Would be interesting to have a comparison of ADC TC for couple of older and newer A3 boards - have the newer lower TC?
 
The following users thanked this post: dietert1

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14080
  • Country: de
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #48 on: July 20, 2020, 03:05:33 pm »
The average TC should be close to zero, just the scattering may be slightly different. Besides temperature there can also be an effect of humidity / board stress so it is not just temperature. However it takes lots of samples (meters) to judge on scattering.  The special chip is still the same - so likely not much change - maybe with batches.
 

Online dietert1

  • Super Contributor
  • ***
  • Posts: 2023
  • Country: br
    • CADT Homepage
Re: HP 3458A ; A/D board spare part availibilty and cost
« Reply #49 on: July 20, 2020, 05:01:41 pm »
What i see there are surprisingly large total shifts of several ppm and an exponential decay of drift with a time constant of 2 years or more. I have seen both with thin film resistors, so i would guess that ADC is a hybrid and the resistor arrays in it cause those shifts. How old is that ADC assembly?
I can also see that by cleaning the fan/filter and applying simple numerical TC and drift compensation, the ADC performs to a standard deviation of less than 0.05 ppm (residual in straight parts of the black line). Do you have similar numbers for the behavior of the reference in that instrument?

Regards, Dieter
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf