Author Topic: Tektronix TDS210 ASIC issue  (Read 3157 times)

0 Members and 1 Guest are viewing this topic.

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Tektronix TDS210 ASIC issue
« on: February 22, 2019, 07:53:22 pm »
I’ve got a faulty TDS210.

I am having some issues. Initially I thought I would have a faulty display board.

ISSUE 1 - my display seems to have a horizontal sync issue with horizontal lines going up and down.

But after following the troubleshooting procedure from its manual.

Checking signals from mainboard to display board. Power and digital data pulses are there. But there is an issue with 3 clock pins.
The 1.89mhz outputs 1.39mhz
The 19khz outputs on clock but duty is less than 1% when should be 16%.
The 75hz outputs nothing.

The display clock comes directly (no parasitic components between) from ASIC MM9527-VCE. The MM9527 gets its clock from a FOX3020 oscillator.

The oscillator also clocks the Processor which seems fine as I can navigate every menu/value with no problem. SO I imagine my oscillator is fine. But clearly seems I got an issue with the MM9527 internal oscillator.

ISSUE 2 - I can get the channels 1 and 2 to trigger and display a sine (WITHOUT PROBES CONNECTED) wave that seem like internal noise. But I CANNOT manage both channels to display any probed signal or the PROBECOMP signal.

I manage to capture the 5v 1Khz wave from PROBE COMP using another scope. The PROBECOMP gets its signal from the MM9527

The  MM9526-VCE deals with both channels and provides the data to MM9527-VCE which then sends data to display.

Have anyone here. Experienced the same issue or could give some light on how MM9527-VCE and the MM9526-VCE works? Pins? etc… any info is of value.
« Last Edit: February 22, 2019, 10:57:11 pm by gkmaia »
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #1 on: March 14, 2019, 05:29:24 pm »
I have a mostly-dead board-corroded TDS210 with MM9590-VJG/MM9526-VJG ASICs. Anyway, despite cleaning, mine is less alive than yours, so I can't offer any advice, but would like to ask a question, if I may?

Data and most voltages were missing at the screen flex connection (and the screen was dark), but I found I could bring them back by jiggering one of the switching transistors (grounding the base of Q130), which lights up the screen with random patterns of non-moving lines. Now on the screen flex there are voltages on pins 1 and 2, and some sort of waveforms on the four data lines. There is a signal on the three clock lines, but it's the same on all three, around 110 Khz.

I can see the 60 Mhz clock signal going into the MM9590 ASIC on pin 10, but no sign of a clock signal at the MC68K, which appears stone cold (and also seems to have HALT and RST held low. Can you offer any insight as to how the CPU is supposed to be clocked, and what the frequency should be at CLK (pin 6)?

I don't think this scope is repairable, but at least I'm trying to learn as much as I can from working on it. Thanks for any help.
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #2 on: March 15, 2019, 02:08:30 am »
I have a mostly-dead board-corroded TDS210 with MM9590-VJG/MM9526-VJG ASICs. Anyway, despite cleaning, mine is less alive than yours, so I can't offer any advice, but would like to ask a question, if I may?

Data and most voltages were missing at the screen flex connection (and the screen was dark), but I found I could bring them back by jiggering one of the switching transistors (grounding the base of Q130), which lights up the screen with random patterns of non-moving lines. Now on the screen flex there are voltages on pins 1 and 2, and some sort of waveforms on the four data lines. There is a signal on the three clock lines, but it's the same on all three, around 110 Khz.

I can see the 60 Mhz clock signal going into the MM9590 ASIC on pin 10, but no sign of a clock signal at the MC68K, which appears stone cold (and also seems to have HALT and RST held low. Can you offer any insight as to how the CPU is supposed to be clocked, and what the frequency should be at CLK (pin 6)?

I don't think this scope is repairable, but at least I'm trying to learn as much as I can from working on it. Thanks for any help.

I cant clock it now because the board is in bits. But have a look at the image bellow there are a bunch os voltages I collected from the board. Those voltages are from the board with then a functional CPU and display. But the acquisition block (IC MM9526) was still at fault.

Hope it helps. When the new MM9526 arrives I will be able to put it back and then get you some clocks.



 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #3 on: March 15, 2019, 10:37:48 pm »
Thanks for the photo. Unfortunately, my board is arranged somewhat differently. For one thing, the ASICs are 25x25 pin while yours are 20x30. The voltages on the darlingtons at the lower right also differ from yours (mine output +4.27, +4.22, and -1.6v). Also, I don't know if it's a typo, but on the power supply line pin 4 I have -4.65v while your label is for +4.85v or so.

But I am still curious about the CPU CLK frequency and where the signal comes from (the ASIC, or direct from the oscillator?).

I didn't realize that custom chips like the ASICs were available. Tektronix or scavenged from a parts board?

Also, I have to say I'm impressed with the photo presentation - can I ask what sort of hardware and software you used - it sure beats the hell out of my pencil scratched jottings.
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #4 on: March 16, 2019, 01:57:45 am »
Power supply pin 4 you are right. Should be -4.65v.

Darlingtons  "mine output +4.27, +4.22, and -1.6v". I think the voltage +4.6 is the voltage to the collector (see image) all my  Darlingtons have emitter voltage at +/-2.5v. But all their collectors get +4v. Check all your 4 darlington and make sure they collect 4V as they get this current from the power supply. If they are not getting may have a trace issue. Also if they are not emmiting then you have an issue with the transistor.

Are you able to let me know what voltage you get out of each of your 4 darlingtons? Also maybe a photo of your board?

Was pretty hard and a bit expensive to source these ICs. I could only find those in china after digging the web for days and leaving several RFQs

I do use Sketch (mac software) same software I use at work as an interface designer. So just a photo a traces on top of it. I do it for pretty much everything as it is easier to follow traces and mark what has been tested and what not.
« Last Edit: March 16, 2019, 06:26:00 am by gkmaia »
 

Offline JFJ

  • Regular Contributor
  • *
  • Posts: 236
  • Country: gb
Re: Tektronix TDS210 ASIC issue
« Reply #5 on: March 16, 2019, 04:38:27 pm »
Thanks for the photo. Unfortunately, my board is arranged somewhat differently...

Is the layout of the series voltage regulator transistors, on your board, the same as in the picture below?



The emitter voltages indicated in the picture are from a working main board.
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #6 on: March 16, 2019, 04:50:35 pm »
Bingo. That's what they look like, although the smd codes are different (two R01 and one R99) and my measured voltages are also different, at +4.27, +4.22, and +0.66 (although I was seeing -1.6 there yesterday?).

BTW, what is the purpose of the button switch S301 in that lower right corner of the board?
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #7 on: March 16, 2019, 05:43:40 pm »
Sigh. I've tried a couple of times to post a photo of the pcb, but none of the attempts get posted, and I get dumped out of the forum. Am I doing something wrong? I'm just clicking on the "Attachments and other options" link, browsing to my photo, selecting it, and then hitting "Post". Or maybe my attempts are going into moderation since I'm pretty new here?
 

Offline JFJ

  • Regular Contributor
  • *
  • Posts: 236
  • Country: gb
Re: Tektronix TDS210 ASIC issue
« Reply #8 on: March 16, 2019, 06:06:09 pm »
... my measured voltages are also different ...

The emitter voltages of transistors Q110 and Q141 are controlled by IC U110 (LMV358 dual op-amp). Measuring the voltages on all eight pins of U110 may provide a clue as to why the outputs of your Q110 and Q141 transistors are so high.

BTW, what is the purpose of the button switch S301 in that lower right corner of the board?

The button enables the service menu. Access is normally via a small hole in the back of the case. Please see page 5-2 of the service manual for details.
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #9 on: March 17, 2019, 04:46:32 am »
I traced out the local circuit for the first two pass transistors, and it looks like two instances of a pretty standard op-amp voltage regulator. Working that,  I found the problem, a corroded trace junction at a via where the reference and cathode pins of the U155 (LM431 zener shunt regulator) that provides reference voltage for the op amp regulator are joined. The trace to the reference pin was open, causing the output to float high. You can see them beautifully pristine in your photo coming from U155. Mine were not so shiny, but I really couldn't see any trace disruption or very much corrosion at the via, but it certainly failed!

When I patched that, the reference voltage fell to 2.5v, and the two rails linked to it dropped to 3.27 and 2.80 volts.  So that's good. The output from the third regulator (should be -2.6v) is still wrong, so I'll look into that next.

So unfortunately, then there's been overvoltages on most/all rails (4.2v instead of 3.3 or 2.8 ).  I wonder if that was enough to kill the ASICs, CPU, or other special chips?
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #10 on: March 17, 2019, 08:17:07 am »
Sigh. I've tried a couple of times to post a photo of the pcb, but none of the attempts get posted, and I get dumped out of the forum. Am I doing something wrong? I'm just clicking on the "Attachments and other options" link, browsing to my photo, selecting it, and then hitting "Post". Or maybe my attempts are going into moderation since I'm pretty new here?

Make sure each or all photos combined have less then 2MB. Also have the .jpg extension.
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #11 on: March 17, 2019, 05:22:09 pm »
I traced out the local circuit for the first two pass transistors, and it looks like two instances of a pretty standard op-amp voltage regulator. Working that,  I found the problem, a corroded trace junction at a via where the reference and cathode pins of the U155 (LM431 zener shunt regulator) that provides reference voltage for the op amp regulator are joined. The trace to the reference pin was open, causing the output to float high. You can see them beautifully pristine in your photo coming from U155. Mine were not so shiny, but I really couldn't see any trace disruption or very much corrosion at the via, but it certainly failed!

When I patched that, the reference voltage fell to 2.5v, and the two rails linked to it dropped to 3.27 and 2.80 volts.  So that's good. The output from the third regulator (should be -2.6v) is still wrong, so I'll look into that next.

So unfortunately, then there's been overvoltages on most/all rails (4.2v instead of 3.3 or 2.8 ).  I wonder if that was enough to kill the ASICs, CPU, or other special chips?


There are two ASICs. One for acquisition and another for the display. I believe the display sends clock to the CPU.
The display ASIC and the memory gets 5V from the power supply board directly. And I believe all the other ICs are the same including processor. As you are getting clock out of the oscillator means you are getting the right power on that side of the board.
The acquisition ASIC is the only that gets around 2.8v from these regulators and transistors from all I could find.

If they killed anything most likely just killed the acquisition.

The system will work without the acquisition, it will just fail self test and wont get any signal. But all the other things should work, like display, menus, calibration, etc...

My piece of advise would be to:
First - Fix any damaged trace. With broken traces you cannot test anything.
Second - Check if you are getting 5v on any trace on display ASIC and its memory, processor and all it's surrounding ICs.
Third - If the ASIC gets power but do not pulse any data or clock to the display connector your ASIC is likely bad.
Fourth - If the processor gets power but do not pulse any data from any connector then the processor is likely bad.

Leave these transistors and acquisition ASIC as the last thing to fix.
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #12 on: March 17, 2019, 08:57:34 pm »
Thanks for the information and suggestions. I think that ultimately board-rot with broken traces and vias will be the cause of abadonment of this attempted  salvage, but I'm not quite ready to give up just yet.

I resized the pictures, and attached them here. The "before" picture will give you an idea of the extent of the problem. The other is after some scrubbing - the bodge wires are to temporarily insert two resistors to substitute for two that rotted out in the R130 quad pack. (And yes, I see the bad pin at the three-o'clock position on the display ASIC - it carries V+ but so do several others, and they all appear to communicate internally.)
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #13 on: March 17, 2019, 09:47:22 pm »
Holy fuck! I had similar issue a while ago with a scope I bought. It had several traces gone, battery & caps leaked, but just one transistor had failed. It is now 100% working.

Do not give up. The only tricky thing you got to get an eye on is this board has a mid layer. Got to make sure it does not loose contact with surface layers.

Have you got a rework station? I would remove all those SMDs and connectors to start with. It will take some time but will be much easier to check/fix traces and that will give you the change to test each those components as well. And if you are going down this route try not to scrape the traces with tweezers, instead get some 1000 grit sand paper, spray some IPA and scrub just the bad traces avioding the good ones or silk. That will remove just the oxidation and not much copper.

And yes, I see the bad pin at the three-o'clock position on the display ASIC - it carries V+ but so do several others, and they all appear to communicate internally.)

It is quite possible your display ASIC (MM9590) pinout is the same as mine at least I can see several traces in common. Check pins 98, 99 & 100 and see if you get some clock, 1.9Mhz, 19Khz & 70hz respectively. Also if you got pulses on pins 93-96 to display connector, then 83-90 to RAM, then 41-48 to CPU (U410), then 13 pulse from oscillator. If you do you just need to find the broken traces to your display connector.


« Last Edit: March 17, 2019, 09:52:07 pm by gkmaia »
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #14 on: March 18, 2019, 01:54:07 am »
Again, thanks for the advice. I'm pretty new to smd work, so this scope is going to teach me a lot whether it lives or dies.

Quote
The only tricky thing you got to get an eye on is this board has a mid layer. Got to make sure it does not loose contact with surface layers.

OK, but how to do that? For instance, the copper/solder of some of these via's appear to have nearly disappeared on the corroded side of the board. Can I reflow them? drill them out and solder a wire through? run a bodge wire around the edge of the board? or what?

Quote
Check pins 98, 99 & 100 and see if you get some clock...

I'll look more carefully next time I have it powered up, but as I recall there were no signals, just some pins sitting at near V+, so probably a dead ASIC. But I'll check again soon.
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #15 on: March 18, 2019, 02:46:24 am »
OK, but how to do that? For instance, the copper/solder of some of these via's appear to have nearly disappeared on the corroded side of the board. Can I reflow them? drill them out and solder a wire through? run a bodge wire around the edge of the board? or what?

That is a laborious job. Need a good magnifier and some patience. Here is an example of a board that badly corroded like yours. I do use very thin wire left over from components like resistors, jumpers, etc... does not look good but is better then just doing a solder joint. The solder cracks with time. The wire do not. Then when you are done make sure you add a layer of UV green mask.



 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #16 on: March 18, 2019, 03:33:02 am »
That's very nice work.

But I'm still unclear about what to do with a via that has been eaten up. Once it's gone, there is no longer continuity with the other surface of the board (or any middle layer, as well). I can see a few places where a trace ends with what appears to be almost a very shallow small circular dark stain on the board - evidently this used to be a via and the copper and solder have been virtually destroyed.

I can try to rebuild traces as you've shown, but what do you recommend for re-establishing the vertical portion of the via connection through the board? Sorry to keep harping on it, but that seems like the worst-case situation I have to deal with here.
 

Offline gkmaiaTopic starter

  • Frequent Contributor
  • **
  • Posts: 528
  • Country: nz
  • Electronics Hobbyist
Re: Tektronix TDS210 ASIC issue
« Reply #17 on: April 16, 2019, 08:16:31 pm »
That's very nice work.

But I'm still unclear about what to do with a via that has been eaten up. Once it's gone, there is no longer continuity with the other surface of the board (or any middle layer, as well). I can see a few places where a trace ends with what appears to be almost a very shallow small circular dark stain on the board - evidently this used to be a via and the copper and solder have been virtually destroyed.

I can try to rebuild traces as you've shown, but what do you recommend for re-establishing the vertical portion of the via connection through the board? Sorry to keep harping on it, but that seems like the worst-case situation I have to deal with here.

Hi Neil, did you figure out your tektronix? any progress?
 

Offline neil

  • Contributor
  • Posts: 47
  • Country: us
Re: Tektronix TDS210 ASIC issue
« Reply #18 on: April 17, 2019, 05:17:21 am »
Yes, thanks. I've been very fortunate with it lately. It's in another thread named "LCD display flaw - TDS210", but basically, I was eventually able to get the scope up and running (while connected to an external -2.6v power supply). I'm waiting on parts now to rework the on-board regulator for that -2.6v rail. 

The failure points were all from things eaten by the corrosion: half a dozen via's, one SOT23 transistor, a SOT23-5 op amp, and two bias resistors in a quad-pack. Miraculously, all of the Unobtanium components survived!

Thanks for your interest and help.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf