Keithley 192 AC transformer

[John3354]

I've had a Keithley 192 in my to-do pile for quite some time.  The other night I decided to give it a go and replaced the fuse with the wrong value (I really should have gotten out my optovisor, and long story short I cooked the AC transformer, along with the 6808 processor, and who knows what else?

All of my Googling has been a dead end so far in finding a replacement transformer.  The processor will be easy to get, I can pick one up for under $10 off of eBay from a US source.  But unless I can procure a replacement transformer it won't matter.  Does anyone know of a good source for them?  The manual shows it as Keithley part number TR-180, I think.  They also list TR-189 for the same part.

According to the manual I need +/-15VDC, +/-30VDC and 5VDC.  I can get those with my selection of lab power supplies, no problem.  I am tempted to get the 6808 replaced and power it via lab power supplies and see what else might be cooked to see if getting a transformer would even be worth it.  I have plenty of other bench meters, many of even higher spec, but I just like old Keithley gear and really want to bring this one back from the dead if it would not be completely pissing money away to do so.

Thank you for any help anyone might be able to lend.  I do appreciate it.

[Kleinstein]

The transformer is a rather speical one. It is not just multiple voltages, but also with extra shield windings.

The windings for the +-30 V could be replaces by a votlage doubler from the 15 V part if really needed.
The 5 V digital side could use a separate transformer - this part is not critical and does not need the shield winding.

[trobbins]

Can you elaborate on what happened?  It seems like there may have been a fault in circuitry for starters (although a blow fuse could occur by chance)?  What windings were damaged? 

Just asking, as I got lucky with a Keithley that had dual 115V primary windings, and one winding was open - although not good for 230V operation where I am, the meter has served me well for over a decade now by using it with a step down transformer and running it from just the one 115V winding.

[srb1954]

I've had a Keithley 192 in my to-do pile for quite some time.  The other night I decided to give it a go and replaced the fuse with the wrong value (I really should have gotten out my optovisor, and long story short I cooked the AC transformer, along with the 6808 processor, and who knows what else?

It would be interesting to know what the fault was that cooked both the transformer and the processor. The only thing that I can see that would do that is a shorted series pass transistor in the digital supply +5V regulator. If that failed it would have likely done a lot more damage to the circuitry than just killing the processor chip.

The whole digital +5V regulator circuit looks pretty dodgy - it doesn't seem to have any form of overload protection. I would pull it out and replace it with something like an LT1085 LDO and you would get proper current limiting, thermal protection and overall better performance.

[John3354]

Can you elaborate on what happened?  It seems like there may have been a fault in circuitry for starters (although a blow fuse could occur by chance)? 

I assume there is some issue with the circuitry, and that is what I was about to dive in to to attempt to figure out.  Unfortunately I did not make it that far.

What windings were damaged? 

All of the secondary windings are toast.  They all now read as a direct short.  The primary windings appear to be fine, but that would be expected.

Just asking, as I got lucky with a Keithley that had dual 115V primary windings, and one winding was open - although not good for 230V operation where I am, the meter has served me well for over a decade now by using it with a step down transformer and running it from just the one 115V winding.

Unfortunately I did not get that lucky.  If it were just one of the primary windings that had failed non-catastrophically I could bypass the 110V-220V switch and get on with it.  But I cooked all of the secondary windings so I may be up a moving stream of feces with no means of conveyance.

The transformer is a rather speical one. It is not just multiple voltages, but also with extra shield windings.

The windings for the +-30 V could be replaces by a votlage doubler from the 15 V part if really needed.
The 5 V digital side could use a separate transformer - this part is not critical and does not need the shield winding.

How did you find such specs on the transformer?  I cannot find anything but a part number with zero specifications.  If you could point me in the right direction I would have a better idea of what I am looking for.  Thanks in advance for any help you might lend.

It would be interesting to know what the fault was that cooked both the transformer and the processor. The only thing that I can see that would do that is a shorted series pass transistor in the digital supply +5V regulator. If that failed it would have likely done a lot more damage to the circuitry than just killing the processor chip.

The whole digital +5V regulator circuit looks pretty dodgy - it doesn't seem to have any form of overload protection. I would pull it out and replace it with something like an LT1085 LDO and you would get proper current limiting, thermal protection and overall better performance.

That seems like a good place to start.  I may desolder it here in a bit and test it on the bench and I will look in to that part number for a possible replacement.  Thank you.

[wishboneash]

I came across this thread recently, so excuse my rather late reply. I got a transformer made at Heyboer Transformers in Michigan (you can do an online search for more info). I had a Keithley 610A electrometer with blown secondary windings and some other incorrect connections. Being an electrometer application, the transformer had shields for the two high high voltage secondary windings and primary winding. There were a couple of filament windings as well. I have a video on youtube https://youtu.be/TLpwEaU6wow
I had to basically come up with specs for it based on the schematic. Mine was a TR-23. After I put the replacement transformer in, the electrometer worked beautifully and it is now fully functional. I thought Heyboer did a good job on it with a lot of communication back and forth.