Author Topic: Keithley 2110  (Read 2058 times)

0 Members and 1 Guest are viewing this topic.

Online ch_scrTopic starter

  • Frequent Contributor
  • **
  • Posts: 812
  • Country: de
Keithley 2110
« on: October 28, 2018, 01:06:03 pm »
About three months ago, while casually browsing Ebay, i stumbled across a Keithley 2110 with recent cal, as defective "Shows 0 on all ranges", "no repair attempt made".
Can't be that bad, can it? Also, I really did not want it badly and told myself "that surely will go for way more than 200€", we can bid that just for fun, can we?
Oh well, did I learn a lesson. As you might have guessed, it ended up going for slightly over 200€ and on my bench.
As described, it displays about zero on all DC Ranges, 25° on Type K Thermocouple and half scale on the ACV Ranges. Tamper sticker was undamaged as well.

Selftest shows Errors 603 (Frontend MCU Communication) and 607 (Buffer 1 Offset) and stops there. Now, as we will see soon, these Errors didn't make a lot of sense, and the manual for the Picotest M3510 references these error codes to a more reasonable 603 (AD Offset fail) and 607 (Rundown too noisy). [The Unit has FW 2.01]  I proceeded to measure the input impedance on different DCV Ranges, 1000V/100V/10V are all around 12-14Meg. 1v/0.1V show a worrying ~45kOhm.
Now after assessing the state of affairs from the outside, its time to poke the tamper sticker and have a look inside...

Well, at least finding the fault didn't take long, and we can guess why no repair attempt was made - the smell was still lingering, but must have been quite intense when it happend.
I proceded to remove the IC (first frontend mux) with hot air, which was a mistake, as it was caked on the PCB, and the PCB material itself seems really heat sensitive.
Lesson learned, next time I will cut the legs and unsolder them one-by-one.

It seems it got so hot, the damage in the center even extends to the other side.

So i did the only thing left to do and took a Dremel to the poor PCB. The picture is with my attempt at restoring the Guard traces already in place.
I cleaned in between and after soldering with IPA and Q-Tips and otherwise tried my best to keep my greasy paws off the pcb.

After that, i took me about one and a half hour to solder the new mux IC into place.


And, to my amazement, it worked. DCV, ACV, capacitance, DCI, Thermocouple. I tested what i could. I compared it to my Keithley 2100, and it's almost spot in DCV, and on the rest of the ranges and functions it's close enough i can't tell which one is off.

 
The following users thanked this post: BFX, doktor pyta, tooki

Offline coromonadalix

  • Super Contributor
  • ***
  • Posts: 5878
  • Country: ca
Re: Keithley 2110
« Reply #1 on: October 28, 2018, 08:15:10 pm »
Be sure to clean any flux resides
 

Offline GregDunn

  • Frequent Contributor
  • **
  • Posts: 725
  • Country: us
Re: Keithley 2110
« Reply #2 on: October 28, 2018, 08:56:33 pm »
My hat's off to you for resurrecting that unit; faced with a mess like that on the board, I would probably have panicked and set it aside.  Pretty amazing that the incident didn't fry more of the surrounding circuitry!
 

Offline t1d

  • Super Contributor
  • ***
  • Posts: 1219
  • Country: us
Re: Keithley 2110
« Reply #3 on: October 31, 2018, 09:47:45 am »
Bravo! Great job.

Yes, next time, cut the legs and desolder individually. Also, you might have used an IC break-out-board to mount the component and aid in the making of connections.

Remember, what got burned may not be what caused the fault. So, retest for odd readings, heat, etc. and keep an eye on it.

Quiet impressive.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf