Low Cost PCB's Low Cost Components

Author Topic: Repair : HP 3456A - fixes and improvements  (Read 4488 times)

0 Members and 1 Guest are viewing this topic.

Offline TiN

  • Super Contributor
  • ***
  • Posts: 2890
  • Country: tw
  • Volt-bolt
    • xDevs.com
Repair : HP 3456A - fixes and improvements
« on: October 29, 2015, 09:10:26 PM »
Hello-hello, dear old-school volt-nuts. One of enthusiasts, member TEKTRON, from another russian-speaking forum prepared nice and detailed article about HP 3456A restoration and tweaks.
Since he is not so well with english, we made best of both of us and I translated into english and hosted it ready for EEVBlog audience. Author also wanted to get it available for broader audience, so posting this thread with his pemission.

Sick HP 3456A bought for 50$ from fleabay was cured by some clever non-standard misfit of ROM and RAM chips, as well as usual fixes here and there, common with old units.



Original LM199 voltage reference was improved, and later completely replaced with DIY triple-LM199 board to achieve lowest noise and best stability, surpassing HP 34401.



Performance comparison with 34401 and 3457A was also done (linearity near zero, noise performance near zero).



Article linked on my site in full.



If you like it or have any questions, please let me know. :) Hope you like this experiment.
xDevs.com | Have test gear documentation to share? Upload here! No size limits, firmware dumps and teardown photos welcome.
 

Offline crispy_tofu

  • Super Contributor
  • ***
  • Posts: 1124
  • Country: au
Re: Repair : HP 3456A - fixes and improvements
« Reply #1 on: October 29, 2015, 09:32:34 PM »
Fantastic! I love it!  :-+ :popcorn:
 

Offline nidlaX

  • Frequent Contributor
  • **
  • Posts: 522
  • Country: us
Re: Repair : HP 3456A - fixes and improvements
« Reply #2 on: October 29, 2015, 10:21:20 PM »
Wow, crazy Russians at it again! 8) That triple LM199 is just awesome.
 

Offline lukier

  • Supporter
  • ****
  • Posts: 495
  • Country: gb
    • Homepage
Re: Repair : HP 3456A - fixes and improvements
« Reply #3 on: October 29, 2015, 11:11:42 PM »
Very interesting article.

I scored 3457A and it is on its way to me and I wondered about doing some modifications as well. 3456A feels a bit too vintage for me, especially the trimpot calibration :)

From what I've read 3457A ages really well, it's spec on 24h accuracy (3V range) is ~5ppm, so better than some 7.5 digit multimeters (3457A is mostly marketed as 6.5 digit). I suspect after 20 year aging it might be even better.

How much would I gain by replacing original well aged LM299 board with 3x LM399 or maybe building a LTZ1000 replacement board? I suspect some noise reduction surely, but even LTZ1000 awesome tempco would be meaningless because frontend tempco is much worse (that's why 8.5 digits have more precise parts or ACAL) and ADC linearity is specified to 1-2ppm AFAIR anyway.

If I decide to do some DIY I would need to age the new reference board and then send the entire multimeter for calibration. Without DIY I might be lucky that the meter is bang on and doesn't need calibration (I might send it anyway as it will be my most accurate meter and I could use it to recalibrate my Keithley 2015 myself, with not the best TUR, but I don't want to loose a fortune on calibrations).

BTW Noob question on ADC linearity measurements. Doesn't the function generator must be more precise than the ADC? Either in ramp or sine + histogram method? For 6.5 digits it would mean 140 dB THD? The same way why one would need 0.1ppm Kelvin-Varley divider. That's why 3458A had to be verified against JJA?
 

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 1980
  • Country: us
  • Cramming the magic smoke back in...
Re: Repair : HP 3456A - fixes and improvements
« Reply #4 on: October 29, 2015, 11:26:02 PM »
That is fantastic, thanks for sharing TiN.   :-+

Online Vgkid

  • Super Contributor
  • ***
  • Posts: 2229
  • Country: us
Re: Repair : HP 3456A - fixes and improvements
« Reply #5 on: October 30, 2015, 02:03:43 AM »
That is a rather extreme modification. I will keep my 3456a stock(over 24hr it is 2-3ppm more stable than my 34401). For the money invested i would rather run a ltz1k reference.
If you own any North Hills Electronics gear, message me.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 2890
  • Country: tw
  • Volt-bolt
    • xDevs.com
Re: Repair : HP 3456A - fixes and improvements
« Reply #6 on: October 30, 2015, 02:15:47 AM »
3456A feels a bit too vintage for me

But 3456A have much nicer warmer LED feel display, than those unholy LCDs without backlight. ;)

Quote
How much would I gain by replacing original well aged LM299 board with 3x LM399 or maybe building a LTZ1000 replacement board? I suspect some noise reduction surely, but even LTZ1000 awesome tempco would be meaningless because frontend tempco is much worse (that's why 8.5 digits have more precise parts or ACAL) and ADC linearity is specified to 1-2ppm AFAIR anyway.

In my opinion, proper LTZ1000 reference cost would blow up cost of such meter way too much (unless you snag one of those 3458A's refs for <100$), and like you clearly see, rest of circuitry do not match up to reference performance.

Quote
BTW Noob question on ADC linearity measurements. Doesn't the function generator must be more precise than the ADC? Either in ramp or sine + histogram method? For 6.5 digits it would mean 140 dB THD? The same way why one would need 0.1ppm Kelvin-Varley divider. That's why 3458A had to be verified against JJA?

Well, generator's output is divided by factor of 10000, together with it's errors. This test was not targeting absolute measurement values, but rather relative result, given particular instrument and setup at given time.
3458A A/D is way more challenging, due to it's much better linearity and sensitivity, to be sure of which JJA is required.
xDevs.com | Have test gear documentation to share? Upload here! No size limits, firmware dumps and teardown photos welcome.
 

Offline lukier

  • Supporter
  • ****
  • Posts: 495
  • Country: gb
    • Homepage
Re: Repair : HP 3456A - fixes and improvements
« Reply #7 on: October 30, 2015, 02:48:30 AM »
But 3456A have much nicer warmer LED feel display, than those unholy LCDs without backlight. ;)

That is one of the mods I'm planning to do. I saw someone added backlight to 3457A and 3478A, it is somewhere on the eevblog forum and also I saw it on bbs.38hot.net forum as well.

In my opinion, proper LTZ1000 reference cost would blow up cost of such meter way too much (unless you snag one of those 3458A's refs for <100$), and like you clearly see, rest of circuitry do not match up to reference performance.

I thought so. I think I'll just add some foam to the 20 year old LM299 and it should be fine. Hopefully, such an old instrument, if I decide to send it for calibration it will be in spec for quite a while and sort of good enough to transfer some calibration points to Keithley 2015 which I suspect is a tiny bit off.

I would send K2015 for calibration, as it is more modern, and my primary workhorse in the lab, but I doubt it is as stable and as accurate as 3457A.

Well, generator's output is divided by factor of 10000, together with it's errors. This test was not targeting absolute measurement values, but rather relative result, given particular instrument and setup at given time.
3458A A/D is way more challenging, due to it's much better linearity and sensitivity, to be sure of which JJA is required.

Ahh, I think I understand now. You were sweeping only around 100ppm (~1mV) so it was more noise and DNL performance, not standard INL measurement.

I was reading a bit recently about measuring INL of 24 bit ADCs and multimeters and the classical way is either precise Kelvin-Varley divider or JJA. Another approach is measuring repetitive sine or ramp signals and accumulating histograms for all the codes. Both approaches seem to take very long time and require top notch equipment (e.g. K-V / JJA or >140 dB THD sine generator).

I was reading some papers that claim to make this easier:

"Low-Resolution DAC-Driven Linearity Testing of Higher Resolution ADCs Using Polynomial Fitting Measurements"
"High-Performance ADC Linearity Test Using Low-Precision Signals in Non-Stationary Environments"
"Algorithm for Dramatically Improved Efficiency in ADC Linearity Test"
"ADC Integral Non-Linearity Testing with Low Linearity Monotonic Signals"

I need to read them in depth and see if the contribution is really there. My thinking goes that long time has passed since 3458A development, plenty of research out there, most multislope conversion patents are now long expired and it seems that the only special thing left that 3458A has are the matched channel length FET switches. Everything else seems to be DIY doable (does not imply it would be cheap - e.g. vishay resistor networks for slopes).

The problem then would be how to measure DIY ADC INL/DNL parameters without JJA.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 2890
  • Country: tw
  • Volt-bolt
    • xDevs.com
Re: Repair : HP 3456A - fixes and improvements
« Reply #8 on: October 30, 2015, 03:10:53 AM »
Quote
seems that the only special thing left that 3458A has are the matched channel length FET switches

That's a bold statement to say at least. :) Main stronghold of 3458A - decades of recognized stability and accuracy, which are hard to beat without spending few decades again on getting reallife data. Yes, 3458A is not ideal and there are instruments which surpass it in performance in one or another point, but not as a whole kit. I'm not seeing another 8.5 digit or thing like 3459A coming anytime soon, there is reason why during all those years we see only 11 meters of this level were in production. And it's not cost of components or parts. But let's not go into this in this thread, there is dedicated 3458A's thread just 2 clicks away ;)
xDevs.com | Have test gear documentation to share? Upload here! No size limits, firmware dumps and teardown photos welcome.
 

Online Vgkid

  • Super Contributor
  • ***
  • Posts: 2229
  • Country: us
Re: Repair : HP 3456A - fixes and improvements
« Reply #9 on: October 30, 2015, 04:05:56 AM »
On the slight OT discussion of 8.5 digit meters, it would have been interesting to see what solartron could have come up with, if they stuck with DMM's. Considering the 7081 predated the 3458 by at least 5 years.
If you own any North Hills Electronics gear, message me.
 

Offline gilbenl

  • Regular Contributor
  • *
  • Posts: 170
  • Country: us
Re: Repair : HP 3456A - fixes and improvements
« Reply #10 on: October 30, 2015, 04:39:47 AM »
This is awesome. Thanks for sharing.
« Last Edit: October 30, 2015, 09:26:43 AM by gilbenl »
What doesn't kill you, probably hurts a lot.
 

Offline guido

  • Regular Contributor
  • *
  • Posts: 180
  • Country: nl
Re: Repair : HP 3456A - fixes and improvements
« Reply #11 on: October 30, 2015, 05:01:49 AM »
Nice mods. I've encountered ROM rot in a 3456A, but was able to fix it with old eproms (going from 3 to 6 :)).
It only failed when pressing specific keys, going to bad code in the affected rom(s).

PS, think i've spotted a typo: "while our modified, more stable 3457A cannot"

And the service note links all point to the 15B doc. Thx
« Last Edit: October 31, 2015, 05:22:36 AM by guido »
 

Offline pelule

  • Frequent Contributor
  • **
  • Posts: 392
  • Country: de
  • Business? Very simple: it’s other people’s money
Re: Repair : HP 3456A - fixes and improvements
« Reply #12 on: October 30, 2015, 07:47:57 AM »
Great Repair and Mod.
You will learn something new every single day
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 2890
  • Country: tw
  • Volt-bolt
    • xDevs.com
Re: Repair : HP 3456A - fixes and improvements
« Reply #13 on: October 30, 2015, 11:03:49 AM »
Service notes all in one PDF. I have links pointing same dox for search purposes only.
xDevs.com | Have test gear documentation to share? Upload here! No size limits, firmware dumps and teardown photos welcome.
 

Offline nidlaX

  • Frequent Contributor
  • **
  • Posts: 522
  • Country: us
Re: Repair : HP 3456A - fixes and improvements
« Reply #14 on: October 30, 2015, 01:22:38 PM »
I just finished reading the whole article, it was thoughtfully written! Poor guy really went down the rabbit hole with the initial repair attempts, but the ingenuity for salvaging random old parts was impressive.  :popcorn:
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf