Intermittent Fault :(

Gentlemen,

I'd welcome any constructive suggestions on how to proceed with tracking down this elusive and annoying fault. The problem presents as a blank CRT screen on the network analyser I'm working on currently. Inititally I discovered the screen was being blanked not just during re-traces, but *permanently* - there appeared to be a rogue logic signal somewhere responsible. However, it turns out the problem goes away and the screen instantly comes back when I press and tap on various areas of the motherboard or the cards that slot into it. The trouble is, there doesn't seem to be one specific area that's susceptible to this tapping and bending. First it appears to be a bad joint/ connection on the motherboard, then it's one of the slot-in boards and then next time again, a different board - but all capable of misleadingly appearing to be the site of the poor connection. Any ideas as to how to narrow the search area right down to something manageable? Thanks.

Reply to
Cursitor Doom
Loading thread data ...

My first guess is that there is a line on the bus that carries that signal and it is being interrupted. Check for any cards that are in pressure type connectors where the board provides the "jumper" between the fingers. Cou ld be as simple as some corrosion on the fingers or the pads on the boards.

Look for the simple stuff first. Something as simple as removing and re-ins erting the cards a few times might be enough. It doesn't take much to mess up a signal.

Dan

Reply to
abrsvc

Monitor the picture tube pins.

You may have a problem with the HV transformer that is losing one of various supplies that the tube needs to work.

Or a simple connection failure - such as is often found with circuit boards that have only a single sided PCB and Molex style pin connectors (solder cracks at the terminals)...

A dual trace scope helps with this sort of troubleshooting. Leave one lead tied to the picture tube cathode so you can see the signal at the tube, and use the other lead to check the signal path while tapping. The idea is to divide in half the circuits relating to the tube, while watching the cathode to make sure it is something unrelated to the video signal that is causing the image to vanish.

John :-#)#

Reply to
John Robertson

I'd prefer to avoid the EHT areas until every other possibility has been eliminated! I don't think that will be necessary anyway.

In the absence of any better ideas, I've decided I'm going to have at it with Arctic Spray in the hope of exposing a dry joint somewhere. The appeal here is that there's no mechanical stress being applied except in the immediate area of application and I've already found that this fault is unusually sensitive to any form of physical stress.

Reply to
Cursitor Doom

I was going to suggest spray for exactly that reason. Maybe in conbination with a heat gun to increase the thermal shock.

For poking EHT I suggest large knitting needles, even if not EHT certified. That's why my long-late Dad had one in his toolkit, anyway...

Mike.

Reply to
Mike Coon

I have a couple of 30kV high voltage AVO probes with long spikes for that sort of thing if necessary. Anyway, I didn't need the spray. It looks like the problem is/was an insufficiently snug edge connector to the motherboard. These are pretty much identical to the ones you see in desktop computers; same pitch etc. I took the suspect board out and cleaned it's edge connections with IPA but noticed on re-insertion there is no positivity about the fit whatsoever unlike the other boards I've removed and re-fitted before. You cannot really tell whether it's seated fully by feel. Not good. Anyway, problem's gone away -- but for how long, who knows?

Reply to
Cursitor Doom

CRT tube cathode is typically under 200VDC, not exactly EHV, and the screen may run in at 400V. Focus is likely 2K or so, but that is why one stocks a HV meter. The 'scope is for monitoring the cathode, not the EHT.

John :-#)#

Reply to
John Robertson

I think you need to be more careful with the advice you're dishing out here. The cathode on this tube runs at -2850VDC with the anode at only

+84VDC. Please don't generalise about the tube potentials on these pieces of test equipment. Someone else may read your remarks on Google a year or more in the future and end up killing themselves.
Reply to
Cursitor Doom

I was not aware that those tubes used such a high negative voltage. Thank you for pointing it out. I am certainly not afraid to admit when I make mistakes!

John :-#)#

Reply to
John Robertson

That marks you out as totally unique in the Usenet community; well done! :-D

Reply to
Cursitor Doom

If the logic is CMOS, you could also have an OPEN input someplace.

I have seen these change state seemingly at random due to static electricity.

m
Reply to
makolber

Well, this is certainly the time of the year for plenty of static - if you're in the northern hemisphere anyway. ;-)

Reply to
Cursitor Doom

.

No, seriously. Its just tracing circuitry. Or are you scared of the fire marshall bill image from the show "In Living Color" (that showed from 1990-

2006) where them and electricians have spiked hair and are accidentally sho cked or burned every day. Gimmie a break.
Reply to
bruce2bowser

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.