Author Topic: Repair worklog : Old rusty HP 3458A (Read 245450 times)

ManateeMafia · « **Reply #125 on:** October 05, 2015, 06:03:31 pm »

I was going to suggest swapping the fiber cables from the rear trigger input. They are not needed for self-test. You can always steal the tx/rx from the trigger i/o.

TiN · « **Reply #126 on:** October 05, 2015, 06:07:01 pm »

Fiber is ok. I guess time to blow dust out of my CSA7404 scope and see some waveforms...

This weak light out of J102 not in my liking. But I could be wrong...

ManateeMafia · « **Reply #127 on:** October 05, 2015, 06:23:27 pm »

I found the datasheet and emailed it to you. The file size is too big.

TiN · « **Reply #128 on:** October 05, 2015, 07:05:59 pm »

Not my lucky day I think.

I had datasheet both HP and Agilent.

Scope decided to die on me too, NVRAM battery died, 0.4V

ManateeMafia · « **Reply #129 on:** October 05, 2015, 07:13:48 pm »

Not good. The more equipment you own, the more likely something, somewhere, is going to die when you least expect it.

Is it possible to mount a remote battery holder for the NVRAM? I am guessing it is the one you replaced under the PowerCAP? https://www.eevblog.com/forum/testgear/tek-csa7404-repair-project/

TiN · « **Reply #130 on:** October 05, 2015, 08:02:35 pm »

NVRAM chip dead, failing tests even with 3.6V LiPo bat hanging on it. $:-\$

What was multimeter repair quickly become scope repair. Not expected detour, tbh.

I guess I'd better go sleep, unless something else died more. (4am here)

Good news - no smoke from DMM. Since I'll be ordering new NVRAM for pesky scope, I'll order optical receiver/transmitters too, but will need more testing.

Macbeth · « **Reply #131 on:** October 05, 2015, 08:23:08 pm »

It's that little black gremlin I tell ya. Got to keep an eye on him!

TiN · « **Reply #132 on:** October 05, 2015, 08:29:20 pm »

Timelapse of today, one photo every 7 seconds

Sorry for glare, had to have lots of light to see what's going on.

TiN · « **Reply #133 on:** October 05, 2015, 08:51:56 pm »

Plan B for tomorrow : hook my TLA714 on optical interfaces and see whats there. I hope it's not also dead (LA), last time I turn it on half year ago.

TiN · « **Reply #134 on:** October 06, 2015, 04:07:24 am »

Here's the battle-field, to summarize in a block diagram. Hope it's not problem in ASICs, as that would be a problem

.

HighVoltage · « **Reply #135 on:** October 06, 2015, 07:43:03 am »

Quote from: TiN on October 05, 2015, 08:29:20 pm

Timelapse of today, one photo every 7 seconds

Very impressive, I watched it halve that speed in VLC.

TiN · « **Reply #136 on:** October 06, 2015, 02:54:38 pm »

Doh, LA died too. Well, LCD flickering like crazy, probably driver or panel gone bad. I don't use it locally, only by LAN anyway, but that's following bad trend here..

Ok, probed around U700..

That RSXGD- signal out of U106 PAL looks strange, causing RESETs on U700.

ManateeMafia · « **Reply #137 on:** October 06, 2015, 04:03:48 pm »

If you are pretty certain the pal is toast, do you have equipment to verify that the read protect bit is not set?

I think there may be a pal or two available from HAK. I don't have a way to r/w pals but it might not matter if the protect bit is set. Of course, there may be other solutions.

TiN · « **Reply #138 on:** October 06, 2015, 04:17:07 pm »

Their finder for parts does not have these ones (03458-88804). PAL is AMD MMI 8L14 Decoder logic.

I don't have anything, never worked with PALs before, only CPLDs. MiniPro does not support these ones either.

Found few datasheets if it's sctructure:

AMD's PAL Handbook, 31MB
8L14 series DS.

I could bodge little ALTERA MAX instead of it, but need to find out logic which goes into that decoder...

Easy things to try - disconnect PAL's pin from board and check if any difference. I will do this tomorrow to check, also capture all PAL signals.

ManateeMafia · « **Reply #139 on:** October 06, 2015, 04:50:33 pm »

I will research what I have available to me and let you know if I can come up with something.

TiN · « **Reply #140 on:** October 06, 2015, 04:56:00 pm »

If you have scope or better, LA and time, signal capture on RSXGD- during power on would be super helpful. This would narrow down suspects to PAL or PALs input circuitry.

ManateeMafia · « **Reply #141 on:** October 06, 2015, 05:21:59 pm »

I have a Saleae logic analyser. Only 8 channels but I think you were only looking at 7 (max), correct?

Did you mean U700 or U100? The clip is hard to read but I think you meant U100.

ManateeMafia · « **Reply #142 on:** October 06, 2015, 05:23:35 pm »

Never mind, found U700 in the corner of the page.

TiN · « **Reply #143 on:** October 06, 2015, 05:25:26 pm »

2 channels enough

. U700 pin 3 (reset) and U106 pin 23 (rsxgd-). U106 is top pal with 03458-88804 sticker, U700 big 40 leg ASIC.

ManateeMafia · « **Reply #144 on:** October 06, 2015, 05:32:23 pm »

OK. I will get you some screenshots. It has been a while since I used this thing.

guido · « **Reply #145 on:** October 06, 2015, 05:47:54 pm »

Quote from: TiN on October 06, 2015, 02:54:38 pm

That RSXGD- signal out of U106 PAL looks strange, causing RESETs on U700.

Mmm. No answer on R_Data, so the processor running the show resets the interface?

Try following the serial data towards U210 and see if it gets there. And if there is an answer, see where it stops.

TiN · « **Reply #146 on:** October 06, 2015, 05:50:22 pm »

Yes, will do that as well. 2 am here already, another crazy day

.

ManateeMafia · « **Reply #147 on:** October 06, 2015, 06:54:09 pm »

It took longer than planned. I had to take a pic as my software will not do screen capture with this version. It is not clear but the top line is U700.3, second is U106.23, and third is U700.1.
4th is from the 8MHz osc.

The data collection was started after I did a SHIFT-RESET. I will try again with power-on.

I will download the latest version in case more data needs to be collected.

I am not capturing a reset on U106.23. The reset line is being held low for 2us.

ManateeMafia · « **Reply #148 on:** October 06, 2015, 06:59:43 pm »

I am changing my settings, I think I am missing the trigger on u106.23

ManateeMafia · « **Reply #149 on:** October 06, 2015, 07:12:59 pm »

This time I was able to capture some traffic. The first reset does not cause any change to U106.23 . There is a change 2s10ms after the reset.
I installed some beta software and the pics are easier to look at. Sorry about the last worthless one.

I hope this helps!


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

Author Topic: Repair worklog : Old rusty HP 3458A (Read 245450 times)

Share me