Seeking help fixing TDS540

[rjmusto]

Hi all,

I am slowly working through the restoration of a TDS540.

All caps on the acquisition board replaced and board cleaned.

I now have two issues with the scope and, as I’m a bit out of my depth here, am hoping some smarter than me in the forum could offer some pointers in where to go next.

When powered up from cold, the ubiquitous acquisition failures are listed, but if left on for 5 or 10 min and then retried, it passes. During this warm-up time, the trace vertical position drifts upwards, starts out off-screen. So far have not found what is drifting. Could probably live with this one though.

More troubling is the second issue. If the timebase is increased into the extended range (>1Gs/s), the trace breaks up. Basically I end up with a dashed line making up the trace.  If more than 1 trace is active then it starts at 500Ms/s. I’m guessing maybe a memory fault? But is this going to be on the acquisition board still, or should I looking on the processor side of things?

Any help much appreciated.

Ralph

[feedback.loop]

perhaps, this might help

[rjmusto]

Hi,
Thanks for that - this is a video I have seen. And follow this Youtuber in fact.

I've been round the acqu board several times, and have already fixed a couple of missing supply lines, but can't see any other issues. All supplies look to be present and correct. Maybe yet another clean?

And I'd really like to get to the bottom of the sampling problem.  The photos attached show the scope doing a decent job of looking at a pulse waveform, plus two showing a sinewave at 1Gs/s, where all is good, then at 2.5Gs/s, where it isn't.....

[feedback.loop]

Did you run the signal path compensation?

[TERRA Operative]

I've seen that choppy waveform attributed to both bad RAM on the acquisition board and calibration issues.
If you have bad memory, as the sample rate goes up and you start using a bad chip, it can cause errors.
Also, if it has bad calibration, the interleaving of the memory at higher sample rates can be out of synchronisation causing this sort of effect.

Do you have the means to recalibrate the unit?

As mentioned above, what happens if you run SPC a few times?

[rjmusto]

Hi,
Thanks for the replies. Much appreciated.

Had a quick play with it this morning.  Should running the SPC result in some pass/fail additional message at the end. It runs through the process - shows Pass underneath the option.

Fault log screen attached.  I've read a post on the Tek Forum where some clever chap worked out what memory addresses correspond to which channel and shows:
ch1 is 0x7300000 through 0x730ffff
ch2 is 0x7340000 through 0x734ffff

So I guess my fault codes mean something wrong with Channel 1 and 2?

[TERRA Operative]

Could be a bad RAM chip then, can you pull the full error logs via GPIB? That way you can see the full messages (Unless you can scroll sideways).

Code: [Select]

To read the first line of the error log via GPIB, send the command 'errlog:first?'
To read out subsequent lines, send the command 'errlog:next?' per line in sequence.
To clear the error log, send the command 'ERR CLEAR' via GPIB.

If you can't find the exact address in the logs to narrow down to the particular faulty chip, it might be time to crack out the hot air gun and start flipping chips out for new ones until you find the bad one.

[rjmusto]

This is all a bit of a new area for me, and don't have any kit to connect to GPIB.  This scope also has an RS232 interface though. Can that be used to get this info? I could probably make up something to connect to that.....

And, yes, better get my SMD skills improved and try tackling reworking the memory chips.  These are 2k x 8 devices right?