[volt-nuts] Keithley 2001 Multimeter Fault - Update

Tue Mar 16 19:52:53 UTC 2010

I've had a closer look at the 'corrosion' damage from leaked capacitor 
electrolyte using a stereo microscope. Some very fine tracks have been 
completely eaten away in one area, but the path of the track is easy to 
follow in the green solder mask layer making it easy to e bypass with a 
wire.

Looking in more detail at the 'Heat Damaged Area' I can see that the 
chemical attack from the capacitor leakage is far more advanced. In addition 
to copper track, it's eaten right through the solder mask and deeper down 
into the PCB resin leaving bare woven glass fibers visible.
There are vias in this area and I'm concerned there may be internal layer 
connections that are damaged.

In addition to conductive electrolyte trapped within the fibrous region I 
think some of the resin may have carbonised and provided another leakage 
path for current, although it might be melted black plastic from the 
capacitors insulating sleeve. A fix will require returning this area of pcb 
to a non conducting state. I might just need to flush it free of electrolyte 
and reseal with epoxy resin, but a thorough fix means carefully excavating 
the damaged area out. This is a lot more trouble than I was expecting, with 
no guarantee I'll eventually get the meter working, but at least it's 
something I can try which might help. I think I'm going to have to fully 
remove the pcb and work on it under the stereo microscope. This might also 
reveal any internal layer connection problems.
The main offending capacitors are marked Nichicon  VZ(M)  470uF,63V 
&1000uF,35V.

Alan

----- Original Message ----- 
From: "Randall White" <randall.m.white at gmail.com>
To: <volt-nuts at febo.com>
Sent: Tuesday, March 16, 2010 1:55 PM
Subject: Re: [volt-nuts] Keithley 2001 Multimeter Fault - Update

> Alan,
>
> I had a similar situation with a power supply from an HP scope. In my 
> case, some of the electrolytics were overheating to the extent that 
> electrolyte had leaked out onto the PCB and was sizzling. I removed all of 
> the questionable caps, and since I do not own an ESR meter, I tested the 
> leakage current of each cap at the rated voltage. There were 2-3 caps with 
> significantly worse leakage than the others, and I replaced those. But 
> first (and more importantly), I thoroughly cleaned the area of the PCB of 
> all electrolyte. The electrolyte on the PCB was conducting current between 
> two adjacent traces, and the excessive load was pulling down the supply. 
> The traces were still OK, so no repair was necessary other than cleaning. 
> After cleaning the board and replacing the bad caps, the unit worked 
> perfectly.
> I don't know if a leakage test is a substitute for a proper ESR test, but 
> it may be a quick check you can do without special equipment.
>
> -Randall
>>
>> ------------------------------------------------------------------------
>>
>> Subject:
>> [volt-nuts] Keithley 2001 Multimeter Fault - Update
>> From:
>> "Alan Scrimgeour" <scrimgap at blueyonder.co.uk>
>> Date:
>> Mon, 15 Mar 2010 21:46:51 -0000
>> To:
>> "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>>
>> To:
>> "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>>
>>
>> Here's a progress report:
>> Searching for a possible short circuited tantalum capacitor I tried to 
>> identify if any were connected to the 5V output of the overheating 
>> regulator. None on the visible side of the analogue pcb were. I dived in 
>> and removed the pcb and did a similar search of the tants on the rear. 
>> One, located right next to the regulator was connected. I replaced and 
>> tested it. It was a bit leaky, but ok, and the replacement made no 
>> difference.
>> I started probing around with a thermocouple looking for overheating 
>> chips connected to the 5V supply, but couldn't spot any. Then I 
>> accidently discovered that 5 electrolytics in the vicinity of the 
>> overheating 5V regulator were also overheating. Unfortunately the pcb 
>> does have buried layers so it's not easy to tell what connects to what, 
>> and I cant get to connections on the rear of the pcb when it's powered 
>> up, so the whole process is proving rather frustrating. I think I'll try 
>> removing the electrolytics and if they're still ok, putting them back on 
>> long wire legs to allow me to access the powered up connections.
>> Any bright ideas welcome!
>>
>> Alan
>>
>> ------------------------------------------------------------------------
>>
>> Subject:
>> Re: [volt-nuts] Keithley 2001 Multimeter Fault - Update
>> From:
>> Brooke Clarke <brooke at pacific.net>
>> Date:
>> Mon, 15 Mar 2010 16:53:47 -0700
>> To:
>> Discussion of precise voltage measurement <volt-nuts at febo.com>
>>
>> To:
>> Discussion of precise voltage measurement <volt-nuts at febo.com>
>>
>>
>> Hi Alan:
>>
>> I'm importing and selling a combined ESR and capacitance meter. See:
>> http://www.prc68.com/P/Prod.html#ESR <- ordering
>> http://www.prc68.com/I/ESRmicro.shtml <- description
>> http://www.prc68.com/I/Capacitors.shtml <- measured results using a 
>> number of instruments
>>
>> Have Fun,
>>
>> Brooke Clarke
>> http://www.PRC68.com
>>
>>
>> Alan Scrimgeour wrote:
>>> Here's a progress report:
>>> Searching for a possible short circuited tantalum capacitor I tried to 
>>> identify if any were connected to the 5V output of the overheating 
>>> regulator. None on the visible side of the analogue pcb were. I dived in 
>>> and removed the pcb and did a similar search of the tants on the rear. 
>>> One, located right next to the regulator was connected. I replaced and 
>>> tested it. It was a bit leaky, but ok, and the replacement made no 
>>> difference.
>>> I started probing around with a thermocouple looking for overheating 
>>> chips connected to the 5V supply, but couldn't spot any. Then I 
>>> accidently discovered that 5 electrolytics in the vicinity of the 
>>> overheating 5V regulator were also overheating. Unfortunately the pcb 
>>> does have buried layers so it's not easy to tell what connects to what, 
>>> and I cant get to connections on the rear of the pcb when it's powered 
>>> up, so the whole process is proving rather frustrating. I think I'll try 
>>> removing the electrolytics and if they're still ok, putting them back on 
>>> long wire legs to allow me to access the powered up connections.
>>> Any bright ideas welcome!
>>>
>>> Alan
>>> _______________________________________________
>>> volt-nuts mailing list -- volt-nuts at febo.com
>>> To unsubscribe, go to 
>>> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
>>> and follow the instructions there.
>>>
>>>
>>>
>>
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> Subject:
>> Re: [volt-nuts] Keithley 2001 Multimeter Fault - Update
>> From:
>> Robert Atkinson <robert8rpi at yahoo.co.uk>
>> Date:
>> Tue, 16 Mar 2010 07:44:25 +0000 (GMT)
>> To:
>> Discussion of precise voltage measurement <volt-nuts at febo.com>
>>
>> To:
>> Discussion of precise voltage measurement <volt-nuts at febo.com>
>>
>>
>> Hi Alan,
>> Have you put a 'scope on the 5V rail? overheating capacitors could be a 
>> sign of an unstable (oscillating) regulator. I'd just change the hot 
>> elecrolytics anyway, they have much reduced lives when they are hot. If 
>> you are testing them, you must include an ESR test of some sort (I've no 
>> connection to any vendors of ESR meters ;-)
>>  Robert G8RPI.
>>
>> --- On Mon, 15/3/10, Alan Scrimgeour <scrimgap at blueyonder.co.uk> wrote:
>>
>>
>> From: Alan Scrimgeour <scrimgap at blueyonder.co.uk>
>> Subject: [volt-nuts] Keithley 2001 Multimeter Fault - Update
>> To: "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>> Date: Monday, 15 March, 2010, 21:46
>>
>>
>> Here's a progress report:
>> Searching for a possible short circuited tantalum capacitor I tried to 
>> identify if any were connected to the 5V output of the overheating 
>> regulator. None on the visible side of the analogue pcb were. I dived in 
>> and removed the pcb and did a similar search of the tants on the rear. 
>> One, located right next to the regulator was connected. I replaced and 
>> tested it. It was a bit leaky, but ok, and the replacement made no 
>> difference.
>> I started probing around with a thermocouple looking for overheating 
>> chips connected to the 5V supply, but couldn't spot any. Then I 
>> accidently discovered that 5 electrolytics in the vicinity of the 
>> overheating 5V regulator were also overheating. Unfortunately the pcb 
>> does have buried layers so it's not easy to tell what connects to what, 
>> and I cant get to connections on the rear of the pcb when it's powered 
>> up, so the whole process is proving rather frustrating. I think I'll try 
>> removing the electrolytics and if they're still ok, putting them back on 
>> long wire legs to allow me to access the powered up connections.
>> Any bright ideas welcome!
>>
>> Alan
>> _______________________________________________
>> volt-nuts mailing list -- volt-nuts at febo.com
>> To unsubscribe, go to 
>> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
>> and follow the instructions there.
>>
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> Subject:
>> Re: [volt-nuts] Keithley 2001 Multimeter Fault - Update
>> From:
>> "Alan Scrimgeour" <scrimgap at blueyonder.co.uk>
>> Date:
>> Tue, 16 Mar 2010 11:47:09 -0000
>> To:
>> "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>>
>> To:
>> "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>>
>>
>> I think you're right, but the capacitors in question don't connect to the 
>> 5V regulator (well, maybee one) so things are more complicated. By 
>> proximity they appear mostly to be connected to bridge rectifiers. I 
>> guess an oscillating regulator could put a high ripple current through a 
>> capacitor on its input too.
>> . I've tested the overheating electrolytics and they all seem ok 
>> (ignoring unknown ESR), but under a couple there's corrosion. There's 
>> also possibly some heat damage to the pcb at one point that needs a much 
>> closer look. I hope it's just more corrosion.
>> Unfortunately my scopes dead, which I'll have to sort that out, but for 
>> now my meters ac range and frequency measurement mode should tell me what 
>> I need to know.
>>
>> Alan
>>
>>
>> ----- Original Message ----- From: "Robert Atkinson" 
>> <robert8rpi at yahoo.co.uk>
>> To: "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>> Sent: Tuesday, March 16, 2010 7:44 AM
>> Subject: Re: [volt-nuts] Keithley 2001 Multimeter Fault - Update
>>
>>
>> Hi Alan,
>> Have you put a 'scope on the 5V rail? overheating capacitors could be a 
>> sign of an unstable (oscillating) regulator. I'd just change the hot 
>> elecrolytics anyway, they have much reduced lives when they are hot. If 
>> you are testing them, you must include an ESR test of some sort (I've no 
>> connection to any vendors of ESR meters ;-)
>>
>> Robert G8RPI.
>>
>> --- On Mon, 15/3/10, Alan Scrimgeour <scrimgap at blueyonder.co.uk> wrote:
>>
>>
>> From: Alan Scrimgeour <scrimgap at blueyonder.co.uk>
>> Subject: [volt-nuts] Keithley 2001 Multimeter Fault - Update
>> To: "Discussion of precise voltage measurement" <volt-nuts at febo.com>
>> Date: Monday, 15 March, 2010, 21:46
>>
>>
>> Here's a progress report:
>> Searching for a possible short circuited tantalum capacitor I tried to 
>> identify if any were connected to the 5V output of the overheating 
>> regulator. None on the visible side of the analogue pcb were. I dived in 
>> and removed the pcb and did a similar search of the tants on the rear. 
>> One, located right next to the regulator was connected. I replaced and 
>> tested it. It was a bit leaky, but ok, and the replacement made no 
>> difference.
>> I started probing around with a thermocouple looking for overheating 
>> chips connected to the 5V supply, but couldn't spot any. Then I 
>> accidently discovered that 5 electrolytics in the vicinity of the 
>> overheating 5V regulator were also overheating. Unfortunately the pcb 
>> does have buried layers so it's not easy to tell what connects to what, 
>> and I cant get to connections on the rear of the pcb when it's powered 
>> up, so the whole process is proving rather frustrating. I think I'll try 
>> removing the electrolytics and if they're still ok, putting them back on 
>> long wire legs to allow me to access the powered up connections.
>> Any bright ideas welcome!
>>
>> Alan
>> _______________________________________________
>> volt-nuts mailing list -- volt-nuts at febo.com
>> To unsubscribe, go to 
>> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
>> and follow the instructions there.
>>
>>
>>
>>
>> _______________________________________________
>> volt-nuts mailing list -- volt-nuts at febo.com
>> To unsubscribe, go to 
>> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
>> and follow the instructions there.
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> volt-nuts mailing list
>> volt-nuts at febo.com
>> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
>
>
> _______________________________________________
> volt-nuts mailing list -- volt-nuts at febo.com
> To unsubscribe, go to 
> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there. 