Author Topic: Help needed: Assessing Solartron 7081 performance  (Read 407 times)

djac and 1 Guest are viewing this topic.

Offline cozdas

  • Contributor
  • Posts: 25
Help needed: Assessing Solartron 7081 performance
« on: July 12, 2019, 03:15:55 pm »
Hi all,

I need expert opinions on my new Solartron 7081 toytool. I bought this unit from ebay for ~$300 for the typical voltnuttery reasons. My immediate goal is to test the noise, stability and tempco of the four DAS-57AL units I recently repaired (so that I can keep the best one).

Initially this 7081 was not working but after cleaning up the mess of the infamous NiMH battery leak around the SRAM area and installing supercaps instead, I was able to resurrect it. It passes the self-test and all functions seems to work. I kept it mostly on for several days (occasionally turned it off and/or removed the cover for cleanup, dip switch settings etc). Finally I warmed it for 24+hrs and did some measurements to characterize the performance. I was hoping that it'd be a big improvement over my new DMM6500 for ppm related tasks.

These are the things I've done so far
  • cleared the mess, replaced the dead RAM battery with supercaps
  • refreshed NVRAM
  • warmed up for 24+hrs but didn't perform the self-baking procedure as described in the manual

There are few apparent issues:
  • It seems +214ppm off. It measures my 10V source (calibrated to 10.00000V agains factrory-calibrated DMM6500) as 10.002141V. The offset halves for 5V: 5.001034V. So looks like scaling/calibration issue(?)
  • Internal 10V test (Test 10vdc) measures 9.886882V . This is very close to min limit of the 9.8V-10.2V range. Is this normal?
  • Quick and dirty preliminary tempco tests (left overnight in the garage for large temp-swing) show that it's much worse than my DMM6500. Still not sure if this is because I adjusted the tempco of the DAS57 against DMM6500 or not. But since DMM6500 has 1ppm/C tempco, looks like Solartron 7081 has at leas 4-5ppm/C which looks too much to me.
  • After the warm-up, I did some basic tests and it looks more noisy than I was expecting. Maybe it's normal. I'll provide the measurement results below and please let me know if it looks normal or not.

For testing the noise I connected my own-refurbished DAS-57AL with a shielded cable (still don't have the proper fischer connector, just inserted the copper ends to the holes). Room temp was pretty stable and monitored the source with DMM6500 connected in-parallel (high imp mode). I measured these sources:
  • TEST 0Vdc mode
  • TEST 10Vdc mode
  • inputs shorted with short copper wire in 0.1V range
  • inputs shorted with short copper wire in 1V range
  • inputs shorted with short copper wire in 10V range
  • DAS-57AL sourcing (presumably) 10.00000V

with these settings (all without any filtering)
  • 6x9: 200 samples
  • 7x9: 200 samples
  • 8x9: 10 samples

I summarized the results in the attached table and I'm very curious if they are as expected or outside of the expected performance.


Also I'd like to know your opinions about if the +214ppm error. Should I try to recalibrate it or might it be the amplifier gain issue? How to test it?

Here are the current calibration constants:
Code: [Select]
Vdc calibration constants
-26.6432762E-06
 164.274653E-03
-4.64916229E-06
 1.64188692E+00
-1.43051147E-06
 16.4116128E+00
-5.12599945E-06
 164.202978E+00
-1.96695327E-06
 1.64130529E+03
Vac calibration constants
 327.219739E-03
 3.27368424E+00
 32.6906788E+00
 327.042840E+00
 3.27177220E+03
kOhm calibration constants
-26.6432762E-06
 166.135736E-03
-4.64916229E-06
 1.66048612E+00
-1.43051147E-06
 16.5974973E+00
-4.64916229E-06
 166.075607E+00
-1.43051147E-06
 1.66003549E+03
Top Ohms range calibration constants
-1.43051147E-06
 9.89989842E+03
 602.414965E-03
Ratio terminals calibration constant
 16.4131605E+00
Zener current token value
083

Thanks in advance.

Ah also: I didn't have the cal key but it took me only 10 seconds to lock-pick it 8) . It's super easy, much easier than hunting down a key in eBay. You may want to give it a try.
« Last Edit: July 12, 2019, 03:59:24 pm by cozdas »
 

Offline e61_phil

  • Frequent Contributor
  • **
  • Posts: 778
  • Country: de
Re: Help needed: Assessing Solartron 7081 performance
« Reply #1 on: July 12, 2019, 07:27:20 pm »
I had a 7071 and a 7081 both were very noisy (have to look for some data)

After calibration of my 7081 (zero and FS) it failed the selftest. The problem was an oscillating input amplifier.
 
The following users thanked this post: cozdas

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 9174
  • Country: gb
    • Having fun doing more, with less
Re: Help needed: Assessing Solartron 7081 performance
« Reply #2 on: July 12, 2019, 08:17:56 pm »
I had a 7071 and a 7081 both were very noisy (have to look for some data)

My 7081's noise improved noticably after:
  • baking it for 24 hours or longer, as per manual
  • tweaking the trimcap so the VCO locks onto 50Hz correctly

Now it is noticably less noisy when measuring a Weston standard cell than when measuring a zener diode reference.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: cozdas

Offline cozdas

  • Contributor
  • Posts: 25
Re: Help needed: Assessing Solartron 7081 performance
« Reply #3 on: July 13, 2019, 07:29:51 pm »
It was a productive session today: I found a problem which explains the error in the reading and the bad tempco I'm seeing.

While testing stuff around the voltage reference I realized that the voltage between -10V Ref and +10V Ref points doesn't match the LK301, LK302 link configuration. Link configuration suggest the in the factory the reference was between 20.00V and 20.31V but  I'm measuring it to be 19.76V. (This value is matching the "TEST 10VDC" reading posted above: 9.88V BTW).

So I started to test around and found out that the Zener Current adjustment circuit is bogus. TP301 should read -9.28V for Zener token=124 but I read a mere -0.6V. I confirmed that the DAC reference and bit inputs are correct. So currently I'm suspecting IC304 (741 buffer opamp) or the IC301 (AD7533 DAC itself). Too bad that they are not sitting in sockets.

I'm glad that this finding explains both the +214ppm error as the reference voltage is lower, and the terrible tempco as the zener is not biased optimally.

Let's see what happens when I replace both ICs.  :-/O
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf