Fluke 8600A troubleshooting help needed (main chip failure?)

[pevedete]

Recently I picked up an old fluke 8600A, however it does not seem to be working. I have tried to look where the fault comes from and for now it would seem that sadly the main custom IC is faulty.

At first not even the display was working, only the decimal points lit up. After some searching I found that the data pins for the display assembly were not low enough to be registered as a logic low. I have managed to get a bodge to work by pulling the signals down a bit by inserting a 1,5kΩ resistor to ground on each of the data tracks (picture 1). On the schematic these are the tracks from U8 pins 31-32-33-34 to U10 pins 7-1-2-6.

And that is about where the good stuff ends. Even when nothing is connected to the input terminals there is a reading on the meter which is wrong. The reading seems to be still partially dependent on the voltage applied to the inputs, however that also seems to be very hit and miss. The value displayed on the display is also the same independent from the range selected, so if I select the 200mV range and read 34,62mV and then switch to the 2V range the display will still read ,3462V. This voltage changes proportionally to the integration time, so the readout of the meter is still functioning correctly. Picture 2 is the integration slope on C28 with disconnected inputs.

Now the reason why I'm suspecting the main IC U8 is because while I was probing around looking for odd signals I came across two pins (22 and 23) on that chip that did not have a nice squarewave at all, they have a sharp rising edge but the falling edge falls very slowly and has small differences in voltage corresponding to which range switch is pressed. In picture 3 you can see the distorted squarewave from pin 22 and in picture 4 the wave from pin 23.

Also if I select the 20V/Ω/mA range a relay starts to buzz and a second decimal point shows up on the screen.

If someone could provide more insight or comparative measurements on a working multimeter I would be very grateful.

Service manual: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=2ahUKEwjOlMnA5cPhAhXNKlAKHdaFBv4QFjAAegQIABAC&url=http%3A%2F%2Fassets.fluke.com%2Fmanuals%2F8600a___imeng0300.pdf&usg=AOvVaw3t6BOi6Zw180Qm_jxhpmy6.
Full schematics are on the last pages.

[anachrocomputer]

Was that a battery-powered Fluke 8600? I have one that had the battery option (four D-cells with dates codes of 1974), and it suffered damage due to the corrosive gases vented from the battery. I'll probably have to replace all the IC sockets, because they're no longer making good contact with the ICs. Some IC leads have broken off, too, and I've had to make a repair to the ROM chip.

So your problems may be down to bad IC connections, rather than defective ICs.

Photos here, if they help at all: https://www.flickr.com/photos/anachrocomputer/albums/72157699195174382

Video clips of the repair:
And:

[DC1MC]

Uhmm, you didn't say anything about the power supplies and if the elcos are OK, before doing any other stuff, make sure that all the power supplies are in spec.

 Cheers,
 DC1MC

[pevedete]

Yes, it is a battery powered unit, however when I got it there were no batteries in it when I got it. There indeed was some corrosion on the battery contacts, but I tried to clean them off as well as I could. I also reseated the IC's in their sockets before turning it on, checking for continuity with the solder joints on the underside of the board.

I did indeed forget to mention that the power supplies are indeed working correctly, the 5V is usually supplied from the batteries, but since these are missing I am powering the meter from a 5V power supply. The + and - 15V lines are also well within spec.