Author Topic: Keithley 486 - Repair (Need help) (Read 6772 times)

Sbampato12 · « **on:** August 04, 2017, 06:31:16 pm »

Hello,

I'm trying to repair a Keithley Model 486 Picoammeter.

Until now I've found the manual, opened it and start probing.

I've isolated a problem (at least one, but it seems to be the 'only' one) to the digital board.
Probing the voltages, I've saw a problem in a specific rail, and managed to find a very hot +5V regulator. I've take out the regulator and put a DC Power Supply (with current limit, at the maximum of the VReg), and the Power Supply showed that it is in CC mode, telling me that something is drawing to much power.
Then with a termal camera I've found 1 very hot IC (with 10 seconds it get over 100ºC). The problem is: It is a PAL (IC U507 wich is a PALCE16V8H). In the manual, there is a Keithley part number, but I've got in touch with them and they told me that there is no chance to get this part, since it has been discontinued since 90' and since 2008 they do not do service in this model.

I've never worked with any PAL circuit, but someone have any clue if it could be possible to get this program in any way?

Thank in advance!

alm · « **Reply #1 on:** August 05, 2017, 11:55:31 am »

Cloning might be possible if you had access to a parts / known good unit and the PAL does not have its security bit set (probably unlikely). Alternatives are a procedure like this (also requires access to a working PAL, possibly from a pal

), or trying to reverse engineer its programming from the schematics. Maybe you could compare the schematics to earlier units like the 480 or 485 in the hope that they might have a similar circuit in discrete logic? Is there a theory of operations describing what the PAL does? Is it function obvious from the schematics or circuit around it?

Pre-programmed parts are unfortunately parts that are often very hard to source. At least EEPROMs can be easily read and copied.

Ordinaryman1971 · « **Reply #2 on:** August 05, 2017, 12:01:31 pm »

I have 487, the difference is that mine has a voltage source... is there a difference in the software, I don't know, I've never even seen 486

Sbampato12 · « **Reply #3 on:** August 06, 2017, 02:29:54 pm »

Quote from: alm on August 05, 2017, 11:55:31 am

Cloning might be possible if you had access to a parts / known good unit and the PAL does not have its security bit set (probably unlikely). Alternatives are a procedure like this (also requires access to a working PAL, possibly from a pal ), or trying to reverse engineer its programming from the schematics. Maybe you could compare the schematics to earlier units like the 480 or 485 in the hope that they might have a similar circuit in discrete logic? Is there a theory of operations describing what the PAL does? Is it function obvious from the schematics or circuit around it?

Pre-programmed parts are unfortunately parts that are often very hard to source. At least EEPROMs can be easily read and copied.

Thank you for the answer,

interesting link about PAL, I've checked it, it has an interesting way to solve, but as you said I would need a pal with a PAL

I've been thinkin about reverse engineering the circuit, but it seems to be a little complex. It has 8 outputs signals (Status, IEEE 488, GCLK, Display, VIA, EEROM, RAM and ROM), I don't think I'm able to replicate these signals without more info.

I will try to find 480 or 485 info, to see if this could clarify a bit.

Sbampato12 · « **Reply #4 on:** August 06, 2017, 02:33:05 pm »

Quote from: Ordinaryman1971 on August 05, 2017, 12:01:31 pm

I have 487, the difference is that mine has a voltage source... is there a difference in the software, I don't know, I've never even seen 486

Ordinaryman1971, the 487 has the same 'logic board' as the 486. So the PAL should has the same program.
Could you do some tests, or try to help with info (I know it is time consuming, so I really understand if you couldn't)?

Thank you.

Ordinaryman1971 · « **Reply #5 on:** August 08, 2017, 01:49:13 am »

Let me take some pictures of the inside so you can stare and compare. Will post some picts later on today or tomorrow.

Ordinaryman1971 · « **Reply #6 on:** November 30, 2019, 03:27:30 am »

Well, two years later I’m getting back to the subject of 486. I’ve got so busy with other projects I haven’t even open the meter.
And I’ve realized actually I have 486 that has been waiting for its turn on the bench.
The problem with mine is lack of reading. The meter zeroes out but there is no meaningful read outs on any of the ranges. Most of the ranges show overload.
I was wondering if somebody experienced this problem.

Ordinaryman1971 · « **Reply #7 on:** November 30, 2019, 03:54:42 am »

I guess this picoammeter I have was overloaded. I’ve bought it long time ago but I’ve never had time to look at it and here we are...
I’m suspecting one of the opamps. I see two of them right by the input. One I can’t see clearly, have to take the board out to be able to read it but the other is AD711kn
Any chance on schematics of the input stage..... is it designed similarly to the electrometer input?

Ordinaryman1971 · « **Reply #8 on:** November 30, 2019, 03:59:57 am »

Here are some pictures of board

Ordinaryman1971 · « **Reply #9 on:** November 30, 2019, 04:00:38 am »

And closeups.

picburner · « **Reply #10 on:** November 30, 2019, 03:36:39 pm »

Wishing you to find the fault soon, could you post the eprom dump?
The K486 firmware is one of the few that is not on the net yet.

Ordinaryman1971 · « **Reply #11 on:** December 03, 2019, 04:53:18 am »

Waiting for some parts, as soon as I have board out.

Ordinaryman1971 · « **Reply #12 on:** December 18, 2019, 08:22:58 pm »

I've checked and the Eprom says 487 800...

Ordinaryman1971 · « **Reply #13 on:** December 18, 2019, 09:20:22 pm »

Here is a schematic of the analog board for both meters. Still getting basically erroneous readings out of mine... have changed two AD711KN opamps...

Sbampato12 · « **Reply #14 on:** March 31, 2020, 07:52:28 pm »

Thank you Ordinaryman1971 for the photos. Right now this would be a little complicated for me.

As I think many of us have a time now, with this corona virus...

I entered in contact with Keithley in Brazil, but they said that as this equipment is no longer manufactured since 1997 (or something like that), they no more do any kind of repair, (imagine selling parts...), except for calibration.

Then, since august last year I moved from Brazil to Italy, and didn't have time (neither money) to bring my lab yet. Just some small things.

But I will search for some photos I've taken that time.
But could you dump the memory of that PAL (I mean, have access to a programmer, and the availability for do that)?

Sbampato12 · « **Reply #15 on:** March 31, 2020, 07:58:42 pm »

I would say to check carefully the 'mechanical' parts too. Because any little grease, moisture and things like that can also inflect errors.

But as you said, the problem looks like a more big problem.

Tomorrow I will download the schematics you've posted and try to check possibles causes.
There is some months I didn't opened the forum, unfortunately.

Ordinaryman1971 · « **Reply #16 on:** April 01, 2020, 03:47:20 pm »

I've done some testing two days ago... it's been a month since I've looked at it... and I've noticed that the overload is present only when certain relays are engaged. All the switching is done with U145 which is DG411DJ DIP-16 Precision Monolithic Quad SPST CMOS Analog Switch.
That would be my next suspect. Unfortunately I don't have any of those... maybe somebody out there has some in stock.
I will make a copy of the eprom tonight and post it.

Kleinstein · « **Reply #17 on:** April 01, 2020, 05:15:59 pm »

For quick check one could use other switch chips instead of the DG411. They have a little higher resistance but the main function should be OK with DG211, DG441 or DG444. DG201 could also work.

Instead of random switching parts one could do a few more measurements to narrow down the fault.

Ordinaryman1971 · « **Reply #18 on:** April 02, 2020, 07:05:08 am »

So, here is firmware, I have C1 on mine.

Ordinaryman1971 · « **Reply #19 on:** April 02, 2020, 07:12:33 am »

and here is first meaningful measurement. Well, I know it's unconventional but.... there were two DG411DJs. I've just desoldered them both and swapped them.
There is still few ranges that are showing overload but I've proved the point....

Sbampato12 · « **Reply #20 on:** April 04, 2020, 01:10:24 pm »

Nice to hear that!

thank you soo much for the dump.

Sbampato12 · « **Reply #21 on:** April 07, 2020, 09:56:51 am »

Ordinaryman1971, can I disturb you a little bit more?

Could you dump the PAL16 IC from the digital board?

I don't know if you are available to do that (I could argue saying that this could serve as backup even for you, rsrs

).
If you are, I don't know what is your programmer.
There is an adapter that could make this PAL readable with a bunch of programmers in this link (only need some resistors, and a socket for commodity) as a 27C020:
http://dreamjam.co.uk/emuviews/readpal.php

I attached the adapter schematic anyways.

I've checked and this IC is the same for the models 486 and 487.

Ordinaryman1971 · « **Reply #22 on:** April 10, 2020, 06:51:58 pm »

Desoldered the PAL16 and I’m trying to find it in the ic choices on my programmer I have mini pro tl866cs... have a newer one also but didn’t check with the other one.
Can’t find it in cs, am I missing something?

miken · « **Reply #23 on:** April 11, 2020, 05:43:00 am »

I don't know if any modern programmers support PALs "natively". The page Sbampato12 linked to shows a method that requires making an adapter to read the PAL as a 27C020, and uses software to reinterpret the readout into logic.

On the schematic it looks like it says 16L8, so hopefully the implementation in the 16V8 is all combinational logic. Another way to read it out is to use a microcontroller to drive the inputs, while reading the outputs. Then the logic can be derived from inspection of the results. I think I may even have some Arduino code somewhere for doing this.

Ordinaryman1971 · « **Reply #24 on:** April 12, 2020, 03:25:29 am »

I wonder if maybe somebody made adapters for it.... want to buy one.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Keithley 486 - Repair (Need help) (Read 6772 times)

Share me