Author Topic: Repair : 94' Keithley 2002 8.5-digit DMM , good ol' cap leaks destuction derby..  (Read 14352 times)

0 Members and 1 Guest are viewing this topic.

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Some of 8.5d-nuts might notice this poor 2002 gone recently on ebay.. And wonder no more, which pesky Keithley-maniac got it..

Just got back home after trip, could not resist to take a sneak peak inside, to get idea what's we dealing with.

Appetizer photos for starters...



Hood removed..



Firmware is A05, unit already equipped with MEM2 DS1248Y option (very old NVRAM chip though, likely dead/near death).



A/D board 2002-162 Revision B, very same as my newer 2007-year Model 2002. :



LTZ1000A reference circuitry, DCV/Ohm blocks with VPG foil resistors, pair of LTC1043-based reference converters... All looks good in this area. Date codes on chips are 1994, so this meter is 22-year old.



What this means? It means capacitor disasters! This is reason of why unit get destroyed and require bit of advanced repair now...



1000uF 50V cap at earth-references circuitry leaked it's nasty electrolyte on PCB, corroding PCB tracks, components pins and causing short. Stuff die and blows up as a result.



More crispy tasty photos:




So if you still have some instruments more than 10 years old and did not replace all electrolytic capacitors, close this thread and do it NOW. :)

To be continued...

YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 719
  • Country: us
Do the transformer windings still measure ok? It looks like it cooked in its own juices for a while before someone noticed the smell of burning pcb.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Yes, xfrmr is OK. Resistance measurement fine and it's working ok in one of 2001's.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TheSteve

  • Supporter
  • ****
  • Posts: 3106
  • Country: ca
  • Living the Dream
Wow, this one should be quite the challenge - those are some seriously hungry capacitors, they want to eat everything.
VE7FM
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: de
Looking forward to follow another excellent thread from you.
Thanks for sharing and good luck
 

Offline ketil b

  • Regular Contributor
  • *
  • Posts: 64
  • Country: au
Hi

So the pic of the bottom of the power supply looks familiar. I go a 2001 just before the new year that wood not power up, turns out that the big mosfet and power resistor on the primary side of the power supply had both failed open, one of the symptoms of this was R102 and R103 where being cooked, almost got throw the pcb before I caught it.

thanks

ketil
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
One of my 2001's had exactly same issue before, and working well repaired, so that was one of reasons why I hit bid button on this 2002. Second reason was due to my wish of making complete schematics of 2002 as well, not only 2001, and I did not want to risk my first 2002.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Day 1 - cleaning analog board, check digital & FP

Repair will start from removing all affected by electrolyte gunk parts and cleaning PCB till it's shiny.



No visible damages on secondary analog power, which is good news, making high chances that all analog circuits is still nice and working.



Methodology is simple. Desolder EVERYTHING in all areas where electrolyte was spread. Jelly-bean parts replace with brand new components without any testing, expensive/custom parts to be cleaned and washed in IPA, checked and then soldered back. I'll make a list of everything need replacement and place order on Digikey next days.

Now, test digital board and FP, using recently bought Keithley 2304A PSU to provide +6.3V at U629 LDO input.



Good news, digital board works fine, VFD is OK too.



It's bit aged on edge segments, but overall pretty good shape for 22-year old unit. I'll use it as is. Overall current consumption of digital+FP boards = ~0.66A.

YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline krivx

  • Frequent Contributor
  • **
  • Posts: 763
  • Country: ie
You have some very nice images of that 2-sided power board, probably good enough to make some reverse-engineered gerbers to provide cheap replacement boards. Have you seen this thread? https://www.eevblog.com/forum/eda/image-to-gerber-converter/
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
I'm not really seeing how that can help in this particular case :) Nice idea though.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline krivx

  • Frequent Contributor
  • **
  • Posts: 763
  • Country: ie
No, I guess it's not needed in this case. It just seems likely that might end up a few more of these units at some point  ;D
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Ordered parts from DK.

Started assembly of what I already have.



Mains area cleaned and ready for repairs...

YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Someone forgot this poor DMM? That ain't me.

Yesterday night I replaced all components in earth-referenced area before the transformer, as well as 2xDG444,DG408 (interesting to note, 2002 using different switches there, instead of DG211 on K2001, shooting for lower leakage?), 1xAD711, 1xLM311, pair of 4094 SRs, 79L05 etc.

Also enforced few vias by putting small copper strand thru them and soldering both sides, as well as few traces were repaired as they were eaten away. One of pads of bridge rectifier was loose, so I had to jump wire it on bottom side. Blown resistors were fixed on same location above hi-temp mylar using piece of copper plate.



PCB was delaminated in few areas, but nothing we can do about that. Should be still fine, as these areas are not high importance of analog path.

End result look like this:



Power switch, Q101 transistor, 2 and 5 pin jacks, fuse holder, LM317T were checked, thoroughly cleaned and reused. It ain't looking pretty, but should do the job well.

Bottom side:



Every single resistor and cap were replaced here, some parts taken from my donor 2001, most of parts brand new from DigiKey.

And, first power on, smoke test passed, no surprises. Moment of truth, self-diagnostics test..



Yay, this fella want to play! I ran it 10 times in a row, all tests pass flying colors.

Hooked it up in parallel with 3458A, which is now calibrated to measured LTZ1000A module (live log data), it read 10ppm high. Not bad, considering last calibration in 2009 and repairs :)

So the outcome of this project was just like I expected and hoped for, just fix of electrolyte mess, replace gunked parts and all is back again.  :-DMM
Seems it's time to sell few 2001's..
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 719
  • Country: us
 :-+ Another Keithley returned to service.

Interesting fix with the burnt area. Will have to remember that.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Interesting fix with the burnt area. Will have to remember that.

Professional way of doing that would be milling/cutting complete burnt area, putting FR4 patch inside and fixing it with board epoxy. Then jump-wiring missing/broken connections with proper wire around. Carbonized FR4 is very leaky/partially conductive, so you don't want any signals running over it.

I think we will be back to proper way of doing it, on some 3458A board which you may know of. Time, time...  :phew:
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 719
  • Country: us
I thought something like this anchored to the pcb with epoxy ...

http://www.robotshop.com/en/prototyping-board-sop8-dip8.html?gclid=CNGruPOc08sCFUc2gQodnhMJ7g

Couple bits of wire through the pcb on each side and you are back in business.
 

Online doktor pyta

  • Frequent Contributor
  • **
  • Posts: 287
  • Country: pl
    • RF Scientific
In my K2001 I used plastic spacer to lower heat transfer from the hot PCB to the body of the capacitor.
I may find the TME part number if someone is interested.

Offline NANDBlog

  • Super Contributor
  • ***
  • Posts: 4596
  • Country: nl
  • Current job: ATEX certified product design
Very nice fix so far.
But if corrosion started on a trace, it will slowly eat the trace the entire length. Would not it be better if you would cut the traces somewhere and completely remove them?
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
All open traces were cleaned to shining copper and I had tinned them with solder. There was a lot of cleaning with IPA many times as well, to ensure all electrolyte is removed.
I'll let you know if that's not good enough in 2026 of before that if meter fails earlier :)

doktor pyta

Neat tweak, but I'm afraid it would not help much, as case is closed and there is nowhere for heat to escape. So with spacer or without it will all be cooking there.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Online doktor pyta

  • Frequent Contributor
  • **
  • Posts: 287
  • Country: pl
    • RF Scientific
Tin
1. For sure this changes the ratio of radiated heat to conducted heat. I haven't made the measurements so I cannot give You the numbers.
2. In case of the leak the spacer should slow down the PCB destruction process (+ I hope that the leak could be easier seen on the surface of the white spacer).

All in all we will see in few years how much this helps.
« Last Edit: March 22, 2016, 01:18:53 pm by doktor pyta »
 

Offline NANDBlog

  • Super Contributor
  • ***
  • Posts: 4596
  • Country: nl
  • Current job: ATEX certified product design
All open traces were cleaned to shining copper and I had tinned them with solder. There was a lot of cleaning with IPA many times as well, to ensure all electrolyte is removed.
I'll let you know if that's not good enough in 2026 of before that if meter fails earlier :)

doktor pyta

Neat tweak, but I'm afraid it would not help much, as case is closed and there is nowhere for heat to escape. So with spacer or without it will all be cooking there.
Yes, I'll make sure to check back in ten years whether or not it is working. Although probably in ten years we will not be using FR4 anymore, but the information is still going to be useful for repairs.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
I'd expect FR4 would still be around just as we have it still. And overall concept (remove damaged parts, put good ones back in, mate the connections) is to remain valid more decades from then. You and me will have a gray beard by then, so bigger question remain would we care about this meter then :).
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TheSteve

  • Supporter
  • ****
  • Posts: 3106
  • Country: ca
  • Living the Dream
Nicely done, seemed just a little too easy!
VE7FM
 

Offline Le_Bassiste

  • Regular Contributor
  • *
  • Posts: 164
  • Country: de
seemingly not entirely on topic, but still...
there are a lot of tear down pictures of KEI2002 and KEI2001 out there, but only a very few give a clear view on the top black plastic shroud that covers most of the analog circuitry. what i've noticed so far:
the shrouds for KEI2002 and KEI2001 are different:
KEI2001 shroud doesn't have any air vents at its circumference, thus protecting the analog circuitry from any "busy" air.
KEI2002 shroud has a vent at the fan-side and an exhaust vent at the opposite side, somewhere in the area of connector P1027. also there is a transparent air duct installed over the fan.

and here is my question: what is the correct placement of the transparent air duct in a KEI2002?
is it like this, allowing for some airflow onto the mains controller V105?
https://xdevs.com/doc/Keithley/2002/photo/Chuck/img_0113_fan%20baffle%20detail.jpg

or is it like this, where the entire airflow is directed into the shroud?







 
The following users thanked this post: smoothVTer

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
On my 2007 year unit - yes, it's like on your photo, all air goes into the shroud (which is confusing to me, to say at least).
On older unit shown here in thread - shroud is 100% same as Model 2001, without any vents or openings.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Quick video of this 2002 repairs:


YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Just a little follow-up on whitening FP on this 2002..

First thing - remove PCB, rubber pad and current input jack/fuse:



Then tape logo, model number, jacks with tape to avoid contact with chemicals



Put panel into proper sized box, pour some warm water with TAED (better not, see later) and puke some Vol.40 peroxide over the top. Make sure everything covered evenly and put it for few hours on direct sun for some UV magic



And....3 hours after if no one stolen your panel out there, it's done:



TAED seem to eat black silkscreen a bit, i'll try without it next time.

Also gold plated terminal pins getting nice shine as well now:



And happy (or not) whitened crowd smile to camera again:

YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline Le_Bassiste

  • Regular Contributor
  • *
  • Posts: 164
  • Country: de
aaand another question that i hope belongs in here:
found a small area on the analog board beneath Q241/Q250 JFET pair  that is covered by a conformal coating. it's transparent and flexible, starting to peel off at one of the solder joins. (see picture below, i added the blue tint to the picture to make the area in question more obvious)
any advice on what substance would be best to apply as new coating?

« Last Edit: June 09, 2016, 08:33:33 am by Le_Bassiste »
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
On my older 2002 there is nothing there, and on newer (07') there is thermal sponge pad there. Might be better just to carefully clean the area and leave it alone, if you don't have fan blowing under the cover hood.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline Le_Bassiste

  • Regular Contributor
  • *
  • Posts: 164
  • Country: de
thanks, TiN.
interesting, seems that they had to improvise some protective measures after the first units. may be a leakage problem? my analog baord is a 1994 rev. D, serial A45xxx
i'll be giving this one a try, ordered the 15 ml bottle:
http://www.electrolube.in/products/conformal-coatings/fsc/modified_silicone/

 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
I'd be reluctant to add any coating over guarded circuits, as that defeat guarding purpose in a way.
Let us know your results though.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline Le_Bassiste

  • Regular Contributor
  • *
  • Posts: 164
  • Country: de
understood. my unit has the thermal pad, too. that pad seems to have "sweatened out" some silicone, i noticed some fine oily film on that  pcb area.
 

Offline Towger

  • Super Contributor
  • ***
  • Posts: 1576
  • Country: ie
'Silicone' absorbs moisture overtime. It would be much better not to have any. I have see equipment where it was used to protect xtals and the build up of moisture/capacitance over the years would stop them from oscillating.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
There is another sick 2002 on fleaBay for $1.5K take-me-now in case someone want 8.5 digits  :-DMM

Seem to be in fairly decent condition, made in mid-90's. Usual treatment with capacitor replacement and PCB inspection is mandatory.
« Last Edit: June 14, 2016, 04:34:24 am by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Attempted calibration yesterday of this fixed Model 2002. Had once gotcha though.
During calibration of 200uA current range, I sourced 200uA from Time Electronics 9823 MFC (range 200uA) into 2002 but I gave error.
Second 2002 indeed show Overflow on same condition, whem MFC sourced 200uA on 200uA range into meter.
If I switch range on MFC to 2mA and just source 0.2mA - 200uA range on DMM works and calibrates OK.
3458A had no problem measuring current on 2mA range as well.

Shot youtube vid covering this:



Why is it so?  :-DMM :-/O

YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 7045
  • Country: de
The problem could be due to some rather high ripple on the current from the calibrator (so a unnoticed defect in the calibrator).

Another possible trap could be if the calibrator does not like the impedance of the DMM as a load and starts to get unstable. Adding something like 10 Ohms in series or about 10 Ohms + 1 µF in parallel might fix this.

 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
3458A reading is steady to +/-2nA, on same range, so I'd expect ripple is not the issue.
HP's impedance on 2mA range for current is 100 ohm, while 2002 on 200uA range is 1KOhm.
Output compliance on 9823 per specifications is 15V with 10G output impedance on DCI.

I'll try with RC network and get back to you.


YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 7045
  • Country: de
It's not low frequency ripple that can be a problem, but more like short spikes. So the reading can be stable but during peaks the current might be well over the nominal 200 µA and thus cause an overflow as some of the current might flow through protective elements. So I would look at the current with a scope.

With 100 Ohms of 1 K impedance (e.g. classical shunt with 200 mV drop) this should not be a problem for the current source. It could be a problem, if the meters were using a TIA input - some meters (e.g. HP3457) do that with low currents. The impedance in this case may be not so well behaved and cause oscillations. Just like a voltage source does not like a pure capacitive load very much, a current source can get trouble with a highly inductive load. If the is the case, I would consider this more like a defect of the current source.
 

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
seemingly not entirely on topic, but still...
there are a lot of tear down pictures of KEI2002 and KEI2001 out there, but only a very few give a clear view on the top black plastic shroud that covers most of the analog circuitry. what i've noticed so far:
the shrouds for KEI2002 and KEI2001 are different:
KEI2001 shroud doesn't have any air vents at its circumference, thus protecting the analog circuitry from any "busy" air.
KEI2002 shroud has a vent at the fan-side and an exhaust vent at the opposite side, somewhere in the area of connector P1027. also there is a transparent air duct installed over the fan.

and here is my question: what is the correct placement of the transparent air duct in a KEI2002?
is it like this, allowing for some airflow onto the mains controller V105?
https://xdevs.com/doc/Keithley/2002/photo/Chuck/img_0113_fan%20baffle%20detail.jpg

or is it like this, where the entire airflow is directed into the shroud?


I recently scored a 2002, totally dead, power supply issue.   Opened up the top analog shield to have a look at cap's condition.  Looking good, not bad, all Nichicon 105C,  see zero evidence of leaks.   Looks like IC date codes are 1999. 

Confusing that the 2002 would have the design decision to divert/force air along through the analog section.  It is bad to have metal junctions at different temperatures cause by air currents, is it not?   Maybe it was getting too hot under the shield and those caps were dying?   Regardless this has at least one problem:  very fine dust collecting over all the sensitive analog circuitry.   Either fan filter was never replaced or it was removed totally... but some dust gets in and over the years it does collect.   

Also wondering how to place this transparent air diverter,  now that I have the entire unit open.   To divert the air away from that analog section is my preference ... but then again I don't know. 

 :-// What's the best way to clean this PCB?   PC air spray can?     Wash in  IPA / methanol?     
« Last Edit: August 13, 2016, 07:49:15 am by smoothVTer »
 

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
I recently scored a 2002, totally dead, power supply issue.   Opened up the top analog shield to have a look at cap's condition.  Looking good, not bad, all Nichicon 105C,  see zero evidence of leaks.   Looks like IC date codes are 1999. 

Confusing that the 2002 would have the design decision to divert/force air along through the analog section.  It is bad to have metal junctions at different temperatures cause by air currents, is it not?   Maybe it was getting too hot under the shield and those caps were dying?   Regardless this has at least one problem:  very fine dust collecting over all the sensitive analog circuitry.   Either fan filter was never replaced or it was removed totally... but some dust gets in and over the years it does collect.   

Also wondering how to place this transparent air diverter,  now that I have the entire unit open.   To divert the air away from that analog section is my preference ... but then again I don't know. 

 :-// What's the best way to clean this PCB?   PC air spray can?     Wash in  IPA / methanol?   

I will not clean the PCB in current state. Rather connect  external power supply to check if meter is working and perform self test.
 Can you post some pictures, because the complete analog section is covered by top and bottom covers to prevent dust. LTZ has its own cover.
In your case compressed air is way how to clean it. IPA will dissolve the dust and after this cleaning you can have more issues.
In general is safe to clean PCB with IPA mixed with water and after that second batch in IPA followed by baking @ 50°C/2h. Avoid rinsing.
 
The following users thanked this post: smoothVTer

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
+1 on photos.
I'd go with dry compressed air first and then leave it as is, if it's not looking too bad. If you can avoid liquids on board - better to avoid them.

Capacitors... they always look good before it's already to late. Replace them. I don't quiet understand why some owners of 2001/2002 would want risk their meters over 10$ worth of caps...
I posted number of horrid photos of what are the consequences, but this question still arise every time. ( :rant: over).
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: Samogon

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
+1 on photos.
I'd go with dry compressed air first and then leave it as is, if it's not looking too bad. If you can avoid liquids on board - better to avoid them.

Capacitors... they always look good before it's already to late. Replace them. I don't quiet understand why some owners of 2001/2002 would want risk their meters over 10$ worth of caps...
I posted number of horrid photos of what are the consequences, but this question still arise every time. ( :rant: over).

I will change out all the electrolytic capacitors for sure;    the big 15000uF cap on the digital board I wanted to replace with a 105C rated cap, but unfortunately I cannot find this same axial cap in 105C version on Mouser, Digikey, or Newark.   Might just replace with a 85C like original.

OK SO Pictures that may explain:

Reference image with the cover in place above the analog board


Cover off now


Cutout in black cover showing the clear plastic fan air diverter and the cutout right across from the fan.  When received, it seems like all the air flow was directed AWAY from the electrolytics/regulators outside of the black cover, instead forcing air in and out thru cutouts in the cover:


Cutout from side view, close proximity to analog section caps & regulators:


Dust over a portion of the analog section;  representative of dust under the black cover in almost all the analog PCB.  Probably will blow this off with canned compressed air.


 :-DD   I'm not 100% sure but ... it looks almost ... as if a prior owner or service tech took a dremel tool to the sides of the cover and cut vent holes into it?   Nice move, Mr. Prior Owner guy.    Can anyone post if their 2002 cover is the same, different or does it lack this air-flow-thru ventilation completely?   






 

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
You are right, the two vent holes are not present on 2002/2001 units. The most recent unit on this forum has TiN, mine one revision older. None of them has these external holes. Cover can be purchased from Tek for $50, If I remember it correctly.
The contamination on picture I will remove with plastic brush ( ESD) and compressed air. It is around AC circuit, so not so critical.
Did you power it up?
 

Offline Le_Bassiste

  • Regular Contributor
  • *
  • Posts: 164
  • Country: de
as the others already mentioned: avoid cleaning if possible. i'd also suggest to wear gloves while handling the analog board. the repair manual mentions "dry nitrogen" for cleaning. pressurized "air" from cans may be ok (but is most likely not air but gas), while i'd not take pressurized air from compressors (even with oil / water separators installed) to dust off the pcb.
as for the shroud, the specimen on my KEI2002 looks the same as yours, having the same port holes, without the "handicraft" marks, though. i had the same concerns as you wrt the airflow going over the analog components, so i took my KEI2002 out for a wild ambient temperature ride and measured offset variation on the 200 mV range over several days. 
see attached diagram, with 1000 points on the x-axis being three hours. y-axis reading is temperature in degC, 2nd y-axis is offset deviation with low-thermal input short on 200 mV range, in ppm of 200 mV. measurement took place on fully assembled unit, PT100 sensor for temp sensing located slightly off center and 2 cm away from rear fan intake.
would really like to know whether others' KEI2002s have similar offset drifts. at diagram start, KEI2002 was already on for several hours.
« Last Edit: August 13, 2016, 07:05:01 pm by Le_Bassiste »
 
The following users thanked this post: smoothVTer

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
You are right, the two vent holes are not present on 2002/2001 units. The most recent unit on this forum has TiN, mine one revision older. None of them has these external holes. Cover can be purchased from Tek for $50, If I remember it correctly.
The contamination on picture I will remove with plastic brush ( ESD) and compressed air. It is around AC circuit, so not so critical.
Did you power it up?

Thanks everyone with your responses so far.  I am going to use the conservative route and do as little touching of the PCB as possible, relying instead on compressed air to get some of the dust out.   

Certainly did power it up, yes.   No voltage on any power rails on secondary side of T100.   I annotated some measurements and voltages from the primary/secondary when I first applied power to the unit.  So far here's what I've found as problems:

Optocoupler that drives the big power MOSFET in the primary pre-regulator circuit is totally blown - open circuit on the input LED.    Testing S-D of Q101 for a short;  no short present, good news, tested gate drive too;  Q101 appears to be functional.   


For the optocoupler replacement, I could find only  the  very similar TLP591B on Digikey.   Same as the TLP590B except that it has a shunt resistor across the output pins "for optimized switching speed" according to the datasheet.   Don't have much experience with opto replacements ... any idea if the TLP591B will work as a drop-in here?    Was also considering a similar pinout but different topology opto like a transistor output opto but again, not certain this will work as a replacement.


On the digital board, I found a single diode short on rectifier CR622 ( red arrow points to the shorted diode )   CR624, CR625 tested good.   Applied 7.5V to U619 and U629, received 5V out.   Appear to be good, I will replace anyway just to be safe.   Could not test comparators U627, U628, U629 yet.   Obviously all caps on the + output of CR622 I'm going to replace as a matter of course.     Also will probably replace Q608 which might have destroyed the opto's input. 


My guess at what happened is probably a voltage spike on the lines coming in from J1032 caused the rectifier CR622 to blow, also perhaps sending the spike further down the line on the +5VC rail and blowing the opto. 


« Last Edit: August 14, 2016, 03:27:17 am by smoothVTer »
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2565
  • Country: us
You might not have to worry about that 15000uf cap. In a bunch of my gear, it is often the physically smaller caps thst tend to go bad.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
My newer 2002 has the vents holes in cover and air deflector for airflow. Older 2002 does not have that, so i was under impression that only newer units got this "tweak". Which maybe still true, perhaps someone with unit made in other dates can confirm it?

I did not see DCV performance difference on 20V range either with vent hole or without though (tested within 0.6ppm uncertainty vs 3458A).

Rest of questions I'll reply later, waiting for boarding in LAX atm :)
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Beefy 470ohm resistor is usual blown too, if you FET dead. And 20k 1210 resistors on bottom side bbq the pcb, often taking zener nearby out too. Check all that.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: smoothVTer

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
Dead relay coil found, K101.    Seems like it was permanently stuck to 220V tap.

Relay K102 was working,  however the flyback diode CR108 ( BAV103 ) was reading 0.4V Vf ... both forward and reverse polarity.  :palm:  Ah, think it is toasted.   All other BAV103's read at 0.6 Vf.   

As you mentioned in previous posts TiN,   its not worth a few dollars to risk some other power supply component being bad or overstressed due to a transient overvoltage condition.   Better to replace entire power section.  That is my plan now.  Need to learn to use a desoldering pump ( no experience with these )  for the relays and power switch removal ( VR101/102  & Q102 hidden underneath -  sneaky sneaky )     








 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 7045
  • Country: de
Q608 can hardly do any damage to the optocoupler. If at all the 150 Ohms resistor could.
The TLP591 should work as a replacement - it's rather similar. R104 is maller than the extra internal resistor.
 
The following users thanked this post: smoothVTer

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
My newer 2002 has the vents holes in cover and air deflector for airflow. Older 2002 does not have that, so i was under impression that only newer units got this "tweak". Which maybe still true, perhaps someone with unit made in other dates can confirm it?

I did not see DCV performance difference on 20V range either with vent hole or without though (tested within 0.6ppm uncertainty vs 3458A).

Rest of questions I'll reply later, waiting for boarding in LAX atm :)

My 2002 is made in roughly in 2008 (serial number starts 09). And it does not have any vent holes. I did not noticed from your pictures this difference when we compare units in past  :palm:
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Well, then it's random :). What's firmware your unit running?
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
Well, then it's random :). What's firmware your unit running?

It has A09, but I upgraded it to A10.
 

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
Past week or so I've been steadily working on replacing components on K2002 main board:   super-paranoid careful to not touch/contaminate the front or back of the PCB, especially areas around the guarded traces.   While replacing C247,250,253  I saw clear, sticky goop around that sponge which TiN also removed in another K2002 post & I just had to share this because I burst out laughing holding the soldering iron in my hand.

TiN :  Check out how Mr. Evil Spongebob Squarepants poops all over the nice guard traces on the rear of the K2002 mainboard:



Who decided to place Spongebob on top of these guarded traces?  I'm wondering about the engineer who decided its time to melt marshmallows inside an 8.5 digit DMM!  Its like the sponge half-melted into ectoplasm and spilled all over. This residue is tacky and sticky to the touch.   I'm thinking years of heat inside the plastic shield maybe caused the glue/polymers of the sponge to semi-liquify. 

Does anybody have an opinion on whether or not this sponge-from-hell is a necessity?   Can we do without it?   Should I clean with IPA?   T




« Last Edit: September 02, 2016, 01:41:15 pm by smoothVTer »
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
That's why guard traces are always should be exposed metal.
You can clean it, your calibration already corrupted anyway by big enough impact :)
I did not clean that spot in my units, decided to live with it, hoping Keithley people know what they are doing.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
It is difficult to imagine this was some mistake by Keithley ... perhaps this was a designed-in feature, as somehow to defeat this guarding?

Hard to believe.  If I find more information I will post it here.
 

Offline Le_Bassiste

  • Regular Contributor
  • *
  • Posts: 164
  • Country: de

Who decided to place Spongebob on top of these guarded traces?  I'm wondering about the engineer who decided its time to melt marshmallows inside an 8.5 digit DMM!  Its like the sponge half-melted into ectoplasm and spilled all over. This residue is tacky and sticky to the touch.   I'm thinking years of heat inside the plastic shield maybe caused the glue/polymers of the sponge to semi-liquify. 

Does anybody have an opinion on whether or not this sponge-from-hell is a necessity?   Can we do without it?   Should I clean with IPA?   T

 what you have there is a deliberately applied conformal coating area under the differential JFET pair. it isolates the thermal pad from the circuitry.  i concur with you that the whole contraption doesn't look too trustworthy, but it's probably put in there to improve thermal equalization of the two JFETs. if you look on top of pcb, you will find the small "silver" box in proximity of the differential pair. that's a burr brown DCDC converter, which gets really hot (at least on my specimen). i'm suspecting that the DCDC converter does really spoil the temperature profile on the pcb, so they decided in an afterthought to implement the thermal pad and the coating to improve thermal equalization. i'm just guessing here based on my findings, though. https://www.eevblog.com/forum/repair/repair-94'-keithley-2002-8-5-digit-dmm-good-ol'-cap-leaks-destuction-derby/msg958245/#msg958245

them voltnut folx may be able to better explain this makeshift "improvement". i also noticed, that without the coating (but with pad), the meter initially reads some wild xx mV in 200mV range, inputs shorted, and takes roughly an hour to stabilize to a repeatable reading of a couple of 100 nV (meter is not calibrated). maybe, i contaminated the meter while replacing caps, or the conformal coating simply needs to be there, or i should remove the pad entirely. i don't know. any help welcome.
« Last Edit: September 02, 2016, 08:48:56 pm by Le_Bassiste »
 
The following users thanked this post: smoothVTer

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
 :-+   Thanks Bassiste,

I failed to notice the jfet diff pair in that area ... good eyes man.   My 2002 does not appear to have that conformal coating on it ... it was just bare PCB / green LPI / gooey mess covering it all.   Just like the user in the post you referenced ...    But your explanation strikes me as the most plausible explanation I've read so far.   I'll sit on this idea for a while and try to think of a hack or mod to achieve the same effect but without the goopy, viscous gluey mess.   

Ideas welcome ...





 

Offline smoothVTer

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
Success!

Checked all replaced components for orientation / type 3x.   Partially re-assembled K2002, pushed power button on. 

Meter springs to life.   Added bonus:  all self-tests pass.   Seems like no errors.  I let it run ~10 hours over night, re-ran self tests again and all self-tests pass.  Preliminary check done on 10V range:



The 34401A  in the photo was last calibrated 6/2014.   Ignore the K2700 above ... seems  out of cal ...

Well I like the fact that the K2002 I restored ( last cal:  2004! ) matches my most recently cal'd 34401A, at least to 6.5 digits.   Especially since I removed the thermal sponge and then cleaned with IPA the area left with all the goo from the sponge/adhesive.

Some observations:
-- U105 (5V reg) gets uncomfortably hot to the touch.  I can see now why the cap next to it tends to leak/explode.  I am going to put a low profile heat sink on U105, it is in the air channel anyway.
-- Testing 200mV range:   shorted HI to LO with bare copper wire, pushed into contact.   Tried all combinations of filtering, A/D sampling window, Line Sync etc    but  meter seems very noisy.   Readings jumping around between 100uV - 800uV.  I have the metal case cover off and I didn't put the ferrite beads back on the input wires yet ... not sure if that would make this much noise.    I can post a video soon of this behavior soon.




 

 

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
The 5V regulator is hot on all 2002, but when covered the airflow will decrease temperature little bit.
For noise you can check the DMM noise by ( EZGPIB) script.
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 7045
  • Country: de
Cleaning of the input section may have an influence on input currents, not so much on the calibration / DC accuracy. The suitable test would be a 10 M resistor or a capacitor to measure the bias.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf