No.091 - Keithley 2001 7.5 digit Multimeter Repair - Part 5 

Absolutely incredible. This instrument is worth a second mortgage by the time you're finished repairing it. What a battle, but you came out on top. I bet most people - including me - would have the launched the odious thing into low Earth orbit whilst hysterically screaming. Deep respect for your patience, insight, and skills, Ian. You're one heck of an engineer. Have a great weekend!

This series is 'solid gold' I'm loving it as its quite a meaty fix to get your teeth into and a great prize at the end, you doing a great job Ian and I'm looking forward to the weekend and some glorious PPM's (as Marco would say !)...cheers.

Great walkthrough of this repair Ian 👍 I'm sure you did the right thing choosing the donor board to repair, as that original looked like it could have needed everything replacing, then still not worked fully in the end. Cheers John

I've really enjoyed this repair series - so many problems! Hopefully it will be working soon!

Excellent! Thank you for your efforts to share and teach. Much appreciated.

Unbelievable endurance you show in this repair. My 2420 SMU still waits for repair...

I used to repair welding power supplies and have seen a number or variations of the 'auto linking' circuit for auto-configuring input voltage selection and have to say this Keithley circuit is the screwiest of them all. Given the havoc that a malfunctioning input selection can have why would they choose to go this route? As an example: Miller Electric had a fairly robust auto-link system on the XMT-304. As I recall it used a separate secondary winding that fed a small switch type supply which powered the auto-link circuitry. This worked well by all accounts except for one edge case. We had one customer that had a rack of 6-8 units that was used for assembling large brewery tanks on site. They would hire a very large commercial gas driven generator to power all the welders with 480VAC. This worked fine for months on several job sites until a generator had a brief brownout which caused the welders to automatically relink to 240VAC and stay there even as the generator's output recovered to 480VAC. Most of the power supplies were damaged and Miller Electric said "Oh, we didn't think of that!" Given the more common sense design of their system they were able to update it to handle that edge case. I wonder what would happen with the auto-linking scheme on this Keithley meter in a brown out? It seems they went to a lot of trouble to design an overly complicated input selection scheme for an instrument which would normally be configured once for input voltage and then never be touched again for 20 years.

looking good, all digits crossed for the big power up.

Sometimes your power supply is just out of reach so your 'emergency stop' for the supply voltages was a great idea 👍

I don’t know how you have the staying power for such in depth repairs. But it is very impressive. Thanks regards Chris

cliff hanger for part 6🙂

Yea, first one on the list! Another great video!

Nice video Ian. For next time i would suggest to use heatsinks for that white resistor..It gets pretty hot if there isn't any heatsink connected to it ..So i would suggest that..and also that transistor that is close to it..(i'm not sure if it was hot or not when i did it..but for sure i know that resistor got pretty hot ..actually i burned my finger..lol..)

epic stuff!

Yay, another part

The circuit with the 470R and the mosfet will also act as a current limit if the system has a fault stopping the rails to come up, say a short circuit secondary diode. I have seen this arrangement in industrial PSU's which tend to have a thermal fuse in the resistor also.

I'm rooting for this thing to sing like a song bird, when you power it up...

I think if it was me id come up with a crowbar to protect that meter from possible future power supply failure .

I noticed the donor A/D board had a cross marked on it... might have a fault.

We are lcd manufacturer for more than 19 years, hope can put some video onyour chanel for advisetime, we can pay, Looking forward to your reply.

I find it hard to believe that mosfet isn't fully on or fully off depending on mode. makes no sense otherwise - have you checked the transformer hasn't got a short on the primary or something?

Can we see your script.?

These things cost a fortune in the first place. I would expect a million to fix one of these. We had a 60k RSA analyzer fron Tek that they said they couldn't repair. Not be using them again...

Sorry bro Ian but I didn't like your new tactic ... You shouldn't replace many parts in that are already working just because you suspect things with and also the modification to main supply cct..... No , this is not REPAIR work anymore ... Wish you luck to have it working at the end ...