-
Advantest R6144 volt/current gen, bad output on lower ranges.
Posted by TERRA Operative on 17 Oct, 2022 15:35 -
I have an Advantest R6144 here that I've finally got back around to repairing after having it sit on the shelf for years....
So far it is all working ok in regards to the user interface, no errors, battery is good, etc etc. I found a bad chip or two on the top logic board which have been replaced.
I tested every.single.chip on this board besides the CPU, and a one or two others that I can't test, using my Retro Chip Tester Pro, so as far as I can tell, all the logic board seems to be working fine.
My problem is when I try to use the 1.6V, 160mV and 16mV ranges I get bad output.
On the lower main board, I originally found a bunch of removed parts in a previous repair attempt, so after replacing those parts with new, I went on to test every diode, every transistor, and every relay and all are good.
I have also tested all the precision resistors, and all are also well under tolerance by an order of magnitude, save the 2ohm R56, that is reading 2.3033ohms (4-wire measurement) or 15.17% high.
The voltage reference (LT1021BCN8-7) also appears to be good, as the upper ranges works perfectly and are spot on.
I haven't tried replacing the associated opamp yet (LT1008CN8/NA) as that part of the circuit seeeeemssss ok, but maybe that could be an issue?
I thought maybe something in the feedback circuit somewhere (oh how I wish for schematics!) so as part of replacing missing parts, I also replaced U9 (LM324N), U10 (LM339N) and U8 (C812C) opamps so they are good.
Today for the sake of it, I replaced U7 (OP77) which was an original part with a better spec OP177 part to no change.
So, besides the 2 ohm resistor I have to figure out, does anyone have any ideas where I can poke to get this thing running?
I have made a graph of the output in the affected ranges (all other ranges work perfectly), linked below.
-
I was about to say / yell : " WOULD HELP IF YOU POSTED THE SCHEMATICS " !
.... but OK, there are none, too bad !
Well my dear Terra looks like this one won't be fixed in 10 minutes.
Will need to spend many hours reverse engineering the thing 'til at least you can understand the basic topology of the thing... then you can start troubleshooting, rather than replacing parts blindly.
It's harder much more interesting, you learn more
Have you tired researching Advantest product line of the day ? Maybe they had several models in that category, and maybe one of them has schematics available. The topology would probably be the same and could therefore help you understand how your unit works.
On the bright side, looks like iut must be something minor, given it works fine on the other ranges... and they are the most sensitive ones as well, weirdly.
It's a simple design, double sided, discrete, lots of space to probe.... you shall be able to achieve victory I would think !
-
Yeah, it's a pain without schematics...
The earlier TR6142 (which I also have) had schematics released, but it is an entirely different beast inside, all trimpots and analogue. The R6142/R6144 are digitally controlled with a microprocessor so nothing is the same inside.
However, I do have an idea I'm going to try.
One of the opamps I replaced way back when was a UPC812C, which is basically unobtanium in a DIP8 package for any sane price these days, even after searching all my secret Akihabara sources...
So I had bought some from ebay as a last resort at the time.
I'm becinning to wonder just how genuine they might be...........
BUT, I can get UPC812G2 opamps, the same part but in a SOP8 surface mount package, genuine off-the-shelf in Akihabara for a couple bucks (few hundred yen) each, so I'll buy some of those and stick them on a SOP8 to DIP8 adapter PCB (also from Akihabara) to make them fit.
That will let me know straight away if my DIP8 chips are good or bad at least.
Too bad Advantest can be right stingy aholes about releasing documentation for their older products. Trust me, I've tried... -
I have repaired a couple of these a few years ago and they are still working well.
The last problem was one of the logic chips that seemed to be ok but was not.
It took me a long time to find the problem.
https://www.eevblog.com/forum/repair/advantest-r6144-dc-voltagecurrent-source-needs-repair/
-
BUT, I can get UPC812G2 opamps, the same part but in a SOP8 surface mount package, genuine off-the-shelf in Akihabara for a couple bucks (few hundred yen) each, so I'll buy some of those and stick them on a SOP8 to DIP8 adapter PCB (also from Akihabara) to make them fit.
That will let me know straight away if my DIP8 chips are good or bad at least.
Yeah sounds like the very first thing to do !
I also noticed that the faulty 1.6V range has a "pattern" to it, sawtooth like. Something to chew...
-
Yeah, I noticed that in the numbers as I was writing them down, the graph confirmed it. It's such a regular sawtooth that it must be significant somehow.
At this point, I'll swap that opamp and see what happens. Without a schematic, it's easier at the moment to swap the easy obvious things than spend hours tracing the circuit. -
I can check what opamp is used in ADCMT 6144 if it helps. Maybe the construction is same only with newer parts.
-
Without a schematic, it's easier at the moment to swap the easy obvious things than spend hours tracing the circuit.
You are in Japan right and speak Japanese, I guess.
Maybe you can call Advantest and ask for a service manual with schematics ?
-
Maintanance manual is without schematic
.
-
Yeah, I have that manual too.
Incidentally, I have optimised it and added bookmarks for easier navigation and smaller filesize with no loss of quality. Link here:
https://drive.google.com/file/d/1TH5QvZ2TVfdWgvaHCu2NG3AvpXM1Yqjo/view?usp=sharing
Ok, so I've narrowed the fault(s) down a bit further now...
Testing between TP4 (Output voltage control PWM) and TP2 (GND) on the top Logic board has some interesting results.
In the ranges that work (16V and 32V), the PWM signal shifts smoothly between roughly 0% and 100% as I raise and lower the voltage accordingly.
However, in the 1.6V range, the PWM signal jumps around in correlation with the graph shown in the PDF file in the original post, in the 160mV range it also jumps around but the output doesn't change accordingly, and in the 16mV range it seems to change smoothly but also with no corresponding change in the output.
So, I am thinking there could maybe be two issues. One is the PWM signal jumping around which is affecting the 1.6V range and would be affecting the 160mV range if it weren't for the second issue in that both the 160mV and 16mV ranges have no change in their output regardless of what the PWM signal says.
As an aside, can anyone dump the contents of the EPROM from their unit? I would like to try a different ROM to see if there's any possible change, just to rule out corruption as unlikely as it is.. -
So I had made a picture of ADCMT 6144. It is the same PCB. I cannot read the ICs names of U7, U8 and U9 on your picture, but it seems that at least U8 and U9 are from different manufacture on my PCB.
Incidentally I had found out that if you disconnect the J8 cable (the red one which need to be disconnected for disassembly) than the output gives full voltage for selected range.
By the way, I have downloaded the manual from your google drive and it is user manual for ADCMT 6144.
-
Comparing the pictures, R56 precision resistor (on the right side, center) is missing on TERRA's unit? Is it a clue or was it never soldered?
-
So I had made a picture of ADCMT 6144. It is the same PCB. I cannot read the ICs names of U7, U8 and U9 on your picture, but it seems that at least U8 and U9 are from different manufacture on my PCB.
Incidentally I had found out that if you disconnect the J8 cable (the red one which need to be disconnected for disassembly) than the output gives full voltage for selected range.
By the way, I have downloaded the manual from your google drive and it is user manual for ADCMT 6144.
Ah, my bad. I uploaded the wrong manual....
The link above has been corrected so you can re-download it now. 
It looks like all my chips are the same numbers as yours except for the OP177. It was originally OP77 in my unit, but as that part is now obsolete (I guess the discontinuation caught Advantest back then too) I've upgraded mine to the better OP177 part now.
One question though, that group of 4 orange diodes just below OP177 in roughly the center of the PCB (D17, D18, D19, D20), can you let me know what is written on them? It appears to be 3636 or 3638?
Mine are plain black with a yellow stripe and white dot, but no numbers at all, and I found one of mine was leaky, but I don't know exactly what part number I should replace it with as the part number in the Advantest service manual doesn't turn up much in Google....Comparing the pictures, R56 precision resistor (on the right side, center) is missing on TERRA's unit? Is it a clue or was it never soldered?
Ah, I removed that one from my unit before I took the photo as it is reading 15% high (is specified as an 0.5% part) so I currently have a 10ohm multiturn trimpot dialled to 2.000ohms there now for testing purposes. -
The marking on the diodes is 36 from other pictures I have taken. But it cannot be seen if there is another row of numbers. I need to check it on the device.
It can be the OP amp imput protection, at least D19 and D20. Question is if the diode have a leakage or if it is simply the effect of surrounding parts. -
Hi TERRA,
I attached the serial eeprom (which holds the calibration values) and firmware rom content. Also the datasheet for the serial eeproms (original OKI and Atmel, which was in my unit). I hope it helps.
BR,
Gyorgy
-
Great, thanks for that!
What version number does your unit show on boot? (The xx.xx number after it shows R6144) -
All my units have version 02.04
BR,
Gyorgy -
Hi TERRA,
I attached the serial eeprom (which holds the calibration values) and firmware rom content. Also the datasheet for the serial eeproms (original OKI and Atmel, which was in my unit). I hope it helps.
BR,
Gyorgy
My R6142 reported an Err 04 error when booting up. After checking, the U5 (AT59C11) firmware was damaged. I used TL866II plus to select the chip AT93C46 (X8). Reading the chip was normal, but the programming verification reported an error, and it seemed that the chip could not be erased. May I ask what programmer you use to program AT59C11? -
Hi,
I used a Beeprog (ELNEC) for reading and programming the AT59C11. The 93C46 is not an exact compatible device. But you can try to replace the original AT59C11 in the R6142 with a AT93C46, which you can program with your programmer. As long as the R6142 will be used to read only the eeprom (normal operation), it will work. And you can write the content available above to the AT93C46 with your programmer.
BR,
Gyorgy -
Thanks!
-
Hi,
I used a Beeprog (ELNEC) for reading and programming the AT59C11. The 93C46 is not an exact compatible device. But you can try to replace the original AT59C11 in the R6142 with a AT93C46, which you can program with your programmer. As long as the R6142 will be used to read only the eeprom (normal operation), it will work. And you can write the content available above to the AT93C46 with your programmer.
BR,
Gyorgy
Hi! Gyorgy, thank you very much! I borrowed a friend's Superpro 500P to program AT59C11, everything turned on normally, and upgraded it to R6144.
-
Hi,
In the end, this is the best solution. In the case you will perform a calibration (and the values should be saved in the EEPROM at the end), you will not have a failure as it would happen with the AT93C46. Since the values in the bin file are the calibration constants for my unit, probably you will have some differences between the set values, and the actual output voltage/current. Using a 6 digit DMM, you can perform the calibration manually, it is documented in the user manual, and you can have spot on output values.
BR,
Gyorgy -
Yes, I calibrated it manually with the 34401. Thanks for your help!