Author Topic: HP 3245A universal source refurbishment and upgrades  (Read 2417 times)

0 Members and 1 Guest are viewing this topic.

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
HP 3245A universal source refurbishment and upgrades
« on: October 20, 2023, 11:07:54 pm »
Hi,

I recently acquired a HP3245A in reasonable condition. It looks like someone's dropped something heavy on it, some dings and gouging of outer metal cover. The front bezel's a bit crushed causing the plastic decal to detach. The VFD display's a bit dim, so replacing the electrolytic caps and the display is first on my list.

I have read through TiN's blog post on the 3245A repair, it sounds like the replacement VFD's on eBay for 3458A would be compatible - is that correct ?

Also any chance of upgrading this with Option 002 (x10 amplifier) or is that going to be impossible ?

Against my better judgement, I did power this on briefly to spot check some DCV and DCI values and seems to be in the ballpark of what I'd expect.
 

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #1 on: October 21, 2023, 02:04:23 am »
and it looks like someone's been inside and replaced at least one of the electrolytic caps with a dodgy repair job after the last calibration done by "Agilent"

the board is in good condition otherwise, can't spot anything dodgy other than the soldering job on the 199/399 reference but don't want to touch that for now.
 

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #2 on: October 21, 2023, 06:14:15 am »
Ian's made an excellent series on repairs.

Part 1 -
Part 2 -

Thanks Ian, much appreciated!

I'll try and get a 3458A VFD. Also interested in the other upgrades, FRAM and the Opamps. I might look into them next.
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: HP 3245A universal source refurbishment and upgrades
« Reply #3 on: October 21, 2023, 06:56:18 am »
Also any chance of upgrading this with Option 002 (x10 amplifier) or is that going to be impossible ?
I'm sure it's possible to create a clone of the board, like someone has done for the HP 3325, but looking at TiN's article the BOM costs look pretty high to me. Power amplifier is $145, the five hermetic VPG resistors and metal can OPA627. I doubt the total BOM will be under $500.
 
The following users thanked this post: deepfryed

Offline TheSteve

  • Supporter
  • ****
  • Posts: 3785
  • Country: ca
  • Living the Dream
Re: HP 3245A universal source refurbishment and upgrades
« Reply #4 on: October 21, 2023, 07:13:57 am »
I'd use FM1808 DIP FRAM devices, no adapter boards needed. They work fine in mine.
VE7FM
 
The following users thanked this post: deepfryed

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #5 on: October 21, 2023, 07:23:51 am »
I'd use FM1808 DIP FRAM devices, no adapter boards needed. They work fine in mine.

I'll have to get hold of them, the usual suspects (Digikey, Mouser, Arrow, Element14) don't stock them. Rochester wants me to buy the whole tube :/
 

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #6 on: October 21, 2023, 07:25:48 am »
BOM costs look pretty high to me. Power amplifier is $145, the five hermetic VPG resistors and metal can OPA627. I doubt the total BOM will be under $500.

Yikes, it's a bigger project that I'm comfortable with at this stage. Perhaps one to add to the long backlog of things I never manage to get to.
 

Offline TheSteve

  • Supporter
  • ****
  • Posts: 3785
  • Country: ca
  • Living the Dream
Re: HP 3245A universal source refurbishment and upgrades
« Reply #7 on: October 24, 2023, 03:08:40 pm »
I'd use FM1808 DIP FRAM devices, no adapter boards needed. They work fine in mine.

I'll have to get hold of them, the usual suspects (Digikey, Mouser, Arrow, Element14) don't stock them. Rochester wants me to buy the whole tube :/

The only way to get them in DIP is to buy used/surplus from china as they aren't made anymore. I've had no issues buying them from china via ebay or aliexpress.
VE7FM
 
The following users thanked this post: deepfryed

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #8 on: October 26, 2023, 08:33:12 am »
I've let the 3245A run for a few days now to stabilise in a temperature controlled room (~25-25.5C, 35-45%RH) and today took some quick measurements with my old 34401A - I had the 10V range "calibrated" a few months back using a friend's DMM7510.

I didn't get any paperwork with the 3245A but there's a calibration reference sticker inside, not sure if I can get anything from Keysight based on that.

I fixed the range to 10V in the 3245A to prevent it auto ranging, the readings look reasonably close to spec. Looks like a sweet spot between 2 and 4V, where the disagreement is lower which is a tad interesting.

I'm looking to get this and one of my relatively stable (less drifty) 34401A re-calibrated at the local Keysight service center down the road, not looking forward to the $$$ quote though ... anyone have experience dealing with their calibration services in Melbourne directly ?


34401A

- Powered on for 4 hours to stabilise
- Manual 10V range
- NLPC 10

3245A

 Powered on for 3 days
- Manual 10V range


Code: [Select]
     Set      Read      Delta
 0.00000  0.000005   0.000005
 0.25000  0.250006   0.000006
 0.50000  0.500008   0.000008
 0.75000  0.750006   0.000006
 1.00000  1.000004   0.000004
 1.25000  1.250005   0.000005
 1.50000  1.500005   0.000005
 1.75000  1.750004   0.000004
 2.00000  2.000004   0.000004
 2.25000  2.250002   0.000002
 2.50000  2.500003   0.000003
 2.75000  2.750000   0.000000
 3.00000  2.999998  -0.000002
 3.25000  3.250000  -0.000000
 3.50000  3.499998  -0.000002
 3.75000  3.749996  -0.000004
 4.00000  3.999992  -0.000008
 4.25000  4.249993  -0.000007
 4.50000  4.499989  -0.000011
 4.75000  4.749993  -0.000007
 5.00000  4.999991  -0.000009
 5.25000  5.249992  -0.000008
 5.50000  5.499989  -0.000011
 5.75000  5.749993  -0.000007
 6.00000  5.999992  -0.000008
 6.25000  6.249992  -0.000008
 6.50000  6.499993  -0.000007
 6.75000  6.749993  -0.000007
 7.00000  6.999990  -0.000009
 7.25000  7.249985  -0.000015
 7.50000  7.499993  -0.000007
 7.75000  7.749983  -0.000017
 8.00000  7.999983  -0.000017
 8.25000  8.249981  -0.000019
 8.50000  8.499981  -0.000019
 8.75000  8.749975  -0.000025
 9.00000  8.999985  -0.000015
 9.25000  9.249983  -0.000017
 9.50000  9.499987  -0.000013
 9.75000  9.749977  -0.000023
10.00000  9.999985  -0.000015
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: HP 3245A universal source refurbishment and upgrades
« Reply #9 on: October 26, 2023, 08:51:20 am »
In terms of accuracy the deltas (a few uV/V) are likely lower than any claim of accuracy the 34401A can make. So you can't say which one is closer, though it gives confidence that both are close.

What might be more useful is the integral non-linearity. There have been some posts on this forum about linearity of the 34401A, and I thought that was pretty good, though I don't remember numbers. You could plot the INL of the 3245A against the 34401A and see how that compares to INL figures other people have measured.

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #10 on: October 26, 2023, 08:55:02 am »
Yeah good idea re. plotting the INL, will do that.
 

Offline PartialDischarge

  • Super Contributor
  • ***
  • Posts: 1625
  • Country: 00
Re: HP 3245A universal source refurbishment and upgrades
« Reply #11 on: October 26, 2023, 09:14:12 am »
Analog has the ADHV4702-1, +-110V supplies , with 1mV offset and an impressive 170dB of open loop gain, so it seems to be very precise and apt for a x10 gain circuit.  Also not expensive.



https://www.analog.com/media/en/technical-documentation/data-sheets/adhv4702-1.pdf


The evaluation board is 200€ and even includes a bipolar flyback converter :  https://www.analog.com/media/en/technical-documentation/user-guides/eval-adhv4702-1cpz-ug-1444.pdf
« Last Edit: October 26, 2023, 11:11:44 am by PartialDischarge »
 
The following users thanked this post: deepfryed

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 15153
  • Country: de
Re: HP 3245A universal source refurbishment and upgrades
« Reply #12 on: October 26, 2023, 01:57:18 pm »
One point worth noting is some 5 µV of offset and slightly lower scale factor for the 34401. After that there is not really much difference left.
The main type of INL expected for the PWM type DAC of the 3245 is a square contribution from a mismatch in switch resistance. There is not much seen of this. So the DAC looks OK with a good match (or a correct working numerical correction).

The data are essentially testing the relative INL - a plot may be easier to see than the raw data. Still it looks like well in specs of both instruments.
 
The following users thanked this post: deepfryed

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: HP 3245A universal source refurbishment and upgrades
« Reply #13 on: October 26, 2023, 09:16:31 pm »
The data are essentially testing the relative INL - a plot may be easier to see than the raw data. Still it looks like well in specs of both instruments.
I agree the data is pretty much what's necessary for INL (in the positive quadrant), except you would subtract a linear fit and report the differences instead of absolute values.

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #14 on: November 04, 2023, 07:25:36 am »
I finally got around to replacing most of the electrolytic caps except the massive 8300uF 35V axial on the A6 outguard power supply board. I up sized some of them and most of the replacements are higher voltage spec'd 105c temperature rated rubycon or panasonic parts. Also replaced the RIFA mains capacitors.

Since I was doing all of this, I also replaced the elec. caps in my 34401A's. One interesting thing I noticed is that the settling time has drastically reduced in both the DMMs and the 3245A. It used to take a couple of hours previously before I could get the last digit to settle down in all of them, now it's about 20-30mins and the 3245A seems to more accurately agree with my 34401A's, one of which was calibrated a while back.

I'm going to let them all stay on overnight in a room with reasonable stable temp ~25c +/- 2c and do some measurements tomorrow. 


Attaching the replacememt parts list I used if it's of use to someone else.
 

Offline deepfryedTopic starter

  • Supporter
  • ****
  • Posts: 130
  • Country: au
Re: HP 3245A universal source refurbishment and upgrades
« Reply #15 on: November 05, 2023, 01:07:58 am »
after about 18h of up time, source seems to agree well with my more recently calibrated 34401a. I wouldn't have expected the capacitor replacement to have such an effect, perhaps the old capacitors had more leakage currents ? I might get some from the bin and check them.

Code: [Select]
 Set       Read       Delta
 0.00000   0.000007   0.000007
 0.25000   0.250006   0.000006
 0.50000   0.500008   0.000008
 0.75000   0.750002   0.000002
 1.00000   1.000002   0.000002
 1.25000   1.250002   0.000002
 1.50000   1.500003   0.000003
 1.75000   1.750004   0.000004
 2.00000   2.000005   0.000005
 2.25000   2.250003   0.000003
 2.50000   2.500004   0.000004
 2.75000   2.750003   0.000003
 3.00000   2.999999  -0.000001
 3.25000   3.250004   0.000004
 3.50000   3.500002   0.000002
 3.75000   3.750005   0.000005
 4.00000   4.000002   0.000002
 4.25000   4.250004   0.000004
 4.50000   4.499996  -0.000004
 4.75000   4.750002   0.000002
 5.00000   4.999995  -0.000005
 5.25000   5.249996  -0.000004
 5.50000   5.499999  -0.000001
 5.75000   5.750001   0.000001
 6.00000   6.000004   0.000004
 6.25000   6.250003   0.000003
 6.50000   6.500004   0.000004
 6.75000   6.750003   0.000003
 7.00000   7.000000   0.000000
 7.25000   7.249998  -0.000002
 7.50000   7.500007   0.000007
 7.75000   7.749999  -0.000001
 8.00000   8.000004   0.000004
 8.25000   8.250005   0.000005
 8.50000   8.500000   0.000000
 8.75000   8.749995  -0.000005
 9.00000   8.999998  -0.000002
 9.25000   9.250000   0.000000
 9.50000   9.500008   0.000008
 9.75000   9.749999  -0.000001
10.00000  10.000008   0.000008
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf