-
Come on a ride as Dave repairs an intermittent Keithley 177 Microvolt DMM, garden paths and all.
Service manual & Schematic: http://www.utwente.nl/tnw/slt/doc/apparatuur/multimeters/keithley177.pdf
Datasheet: http://www.datasheets.pl/integrated_circuits/I/ICL/ICL71C03.pdf
555 Timer T-Shirt:
http://teespring.com/555timer-grey3
Negative Feedback T-Shirt in the EU:
https://fabrily.com/negativefeedback
-
1) If you look down the back of some of the switches there's a hole, squirt some switch cleaner down there with the meter front panel downwards.
2) Units like this have many switch contacts that are not used so wire a spare set in parallel to the burnt contacts.
Finally..........................
Keep your switch cleaner well away from the ADC circuits, if you contaminate the PCB it causes all sorts of strange readings (been there, done that) -
Great video! I'm proud to be the one who fixed the schematics from my bad copy. Very glad to see that someone else had a use for it. It actually took many, many hours to draw those lines!
I had the exactly same problem, bad switches, and yes, the ohms and voltage switches were the worst. But, you actually can apply some contact spray on the ends (opposing ends from the front) and you also can take the switches apart without desoldering. They are a mess though. The pins, as you said, go across and the wipers contacts the pins directly. If you remove the front, then the metal bar in the front of the switches, the bar holding the wipers can be removed. The wipers are just loose metal tongues and will fall off when the wiper holding bar is removed. It's not trivial to put it all together again, but this is what I was forced to do to fix those intermittent problems.
After the fix, the instrument worked again for many years, until this very summer (northern hemisphere) , when the instrument suddenly fell out of calibration, showing about 20% to high values on all voltage ranges. This is outside the possible calibration ranges of the trimpots. I suspect it may be due to the summer here in Sweden, which has been unusual wet with RHs approaching 80%, so I will wait to the autumn and try to recalibrate it.
The unpopulated socket was correctly identified in the video as a plug-in socket for a serial- or HPIB-interface, not for an IC. I actually made my own interface for it to my own specification and has therefore invested a lot in the instrument, why I also will go the extra mile to recalibrate it once more (even if it takes replacement of components).
I believe I have discussed the RS232 interface in another thread (the thread where I posted the schematics).
Again, great video! Great recognition factor!
Edit: And yes, all the same confusion. I also looked at the chopper circuits, discovered the two chip chipset, and also desoldered and checked all the pots. And, after the fix, made some modifications with my interface board. Negative input and system ground is the same potential, so it was possible to have a relay with a resistor to +5V and voliá - free diode measurement range! (Control from RS232, not from front panel.)
-
If you're quick, there's a 179 (TRMS) going REAL cheap on ebay right now... US$15 is the current bid with < 2hrs remaining!
I was tempted to bid myself, but I want at least 5 1/2 digits (OK, I actually WANT 8 1/2 digits, but can only afford 5 1/2... LOL) -
Great video! I'm proud to be the one who fixed the schematics from my bad copy. Very glad to see that someone else had a use for it. It actually took many, many hours to draw those lines!
Thanks for that, proved useful! -
Short a test point with a multimeter on the amps range? ...because it's fused?
Be sure to use a Fluke with those $12 fuses in it.
-
Probably too late now but when it randomly came good during testing, could it have been when you touched one of the terminals with your hand (and grounded the charge build up)?
-
2) Units like this have many switch contacts that are not used so wire a spare set in parallel to the burnt contacts.
The manufacturers don't already do that to increase reliability?
-
Probably too late now but when it randomly came good during testing, could it have been when you touched one of the terminals with your hand (and grounded the charge build up)?
Nope, it often just came good (or went bad) without getting close to it. -
What about the exploded resistor? Didn't that break anything?
-
Short a test point with a multimeter on the amps range? ...because it's fused?
Be sure to use a Fluke with those $12 fuses in it.
If shorting a test point blows a 10A fuse, then might just be thankful you used a fused meter and safe probes to do that instead of just using a bit of wire
-
What about the exploded resistor? Didn't that break anything?
Yes, it seems it took out the diode bridge. -
What about the exploded resistor? Didn't that break anything?
Yes, it seems it took out the diode bridge.
But doesn't it need a new resistor?
-
It's quite nice to have such a meter on my bench but unfortunately I'm too far away from AUS (I'm from Philippines) and a good friend of mine ( who happens to be an Australian and a member here) just sent a box yesterday containing an analog scope and some other stuff that he donated to me. Well I guess someone out there deserves that meter.
-
I suspect that good troubleshooting procedures and making a video simultaneously is an impossible task. As a 'professional troubleshooter' to component level for most of my working career (now retired), I was yelling at you almost non-stop watching this video.
So many fail walks into the tall weeds, jumping to false conclusions, improper symptom recognition. However I couldn't stop watching the whole thing.
It was like reliving past battles.
I think good troubleshooting skills is as much an art as learned skill. I feel you proceeded much too quickly and didn't think carefully enough about each analysis, decision, and path you made as you walked down the problem. That's probably because you also have to keep feeding audio talking points while trying to troubleshooting at the same time. Most brains are too small to pull that off successfully.
To be useful as a teaching moment, such a video troubleshooting task should be 'scripted' where you previously did the complete troubleshooting process and then create and follow a show and tell script to share your decision steps that lead to solving the problem.
All that said, intermittent problems are the worst case symptoms to have to deal with.
-
I have always hated those slide switches, even though by design they are self cleaning.
My experience working on now older Motorola two way radio gear that used those switches. (Micor systems 90 Syntor and Syntor X radio gear.) Their self cleaning feature becomes less effective as the switches are used.
You actually can get contact cleaner/lube in to the sealed ones since the ends are open..
I have to confess I was the one who started screaming "clean the switches" at 10:00 into the video.
Sorry....
-
Nice job with the troubleshooting Dave and David, I was the one who sent you the DMM almost a year ago. I am also glad to hear that you decided to give it away to someone who needs it more!
I want to explain the lack of finesse on the diode bridge repair by saying that single diodes was the only thing available in the drawers in the lab at the time and the only available soldering iron was a large-tipped mains powered one! Maybe Dave has a rectifying bridge in a SIL package on hand which he can replace the diodes with at the same time as he replaces the blown resistor before giving it away?
I can also clarify that 30 kV was never connected to the DMM, we only had a low power 30 kV arc/discharge through the air a few meters away from the DMM. But the electromagnetic radiation from this arc probably caused some component to latch-up which shorten the supply rail which in turn burned the diode bridge (my theory). I believe that the burned resistor close to the output has been like that for a much longer time and is not related to any faults you saw now.
I am equally surprised as you were regarding that the solution was only to clean the contacts. Because the meter worked fine just a day or two before the diode bridge broke and I would never have guessed that the contacts could become bad from that. After all maybe there was an ESD inside the DMMs contacts even though the DMM did not have any leads attached to it but was connected to mains at the time of the 30 kV arcing event which took place in the same room.
/Richard -
double post....
-
Could it be that the ADC checks should have been done with TP1 and TP2 still shorted?
-
Could it be that the ADC checks should have been done with TP1 and TP2 still shorted?
Yes, but he did flash a message box stating that (after the fact) even before the long trip into the weeds.
-
Nice job with the troubleshooting Dave and David, I was the one who sent you the DMM almost a year ago. I am also glad to hear that you decided to give it away to someone who needs it more!
There may have been some other reason the diodes blew, as for that resistor over in the right rear corner.... who knows... The switch issue is common among test equipment that is old enough to vote and certainly among gear that is old enough to have grandchildren, such as my Tek 466 storage scope. The primary culprit is oxidation, even in a Class 100 clean room (back when I spent some time working in one) oxidation was an issue.
I want to explain the lack of finesse on the diode bridge repair by saying that single diodes was the only thing available in the drawers in the lab at the time and the only available soldering iron was a large-tipped mains powered one! Maybe Dave has a rectifying bridge in a SIL package on hand which he can replace the diodes with at the same time as he replaces the blown resistor before giving it away?
I can also clarify that 30 kV was never connected to the DMM, we only had a low power 30 kV arc/discharge through the air a few meters away from the DMM. But the electromagnetic radiation from this arc probably caused some component to latch-up which shorten the supply rail which in turn burned the diode bridge (my theory). I believe that the burned resistor close to the output has been like that for a much longer time and is not related to any faults you saw now.
I am equally surprised as you were regarding that the solution was only to clean the contacts. Because the meter worked fine just a day or two before the diode bridge broke and I would never have guessed that the contacts could become bad from that. After all maybe there was an ESD inside the DMMs contacts even though the DMM did not have any leads attached to it but was connected to mains at the time of the 30 kV arcing event which took place in the same room.
/Richard
I have a friend who owns several Keithley meters including a 177 from that time period and once and a while he just has to exercise the switches to clean the oxidation out. Those are good meters, going to have to add one to my bench -
but he did flash a message box stating that
Ok, that was a YouTube note and not in the downloaded video I've watched. -
Good job Dave/David. I'm an avid hater of those ganged type switches.
-
Thanks for doing the video Dave. I too was screaming at the monitor - clean the switches!
I purchased one of these 177s last year from eBay. Seller claimed it worked OK, but when I got it, it was doing the same as yours. The first thing I did was sprayed some contact cleaner into the switches and worked them. Problem solved! I now have a nice bench meter.
Early in my electronics career (early 1980s) I was troubleshooting and repairing many different types of equipment, similar vintage to the 177. Often I solved some of the faults by cleaning all pots and switches with contact cleaner. That's why I went first for the cleaner while fixing my 177.
Keep trying with the troubleshooting videos.
Craig. -
but he did flash a message box stating that
Ok, that was a YouTube note and not in the downloaded video I've watched.
I cannot add text overlays to videos I have already uploaded to youtube, only youtube text annotations.
If you are watching a downloaded version of any video will miss those.
I often add annotations like that.