Topics

7834 Missing Trace


sdferg7@...
 

Hello,

I recently purchased a 7834 in mostly working order, the only issue the seller listed was that the left vertical slot was not working for unknown reasons. I got it home and powered it up and it seemed to work as described. I only had about half an hour to play with it that night so I didn't give it a super thorough shake-down. The next day I pulled out the right horizontal plugin to see how the mechanism worked, as the other plugins are missing pull tabs. When I put it back in and powered it up, I found I had no trace. The readout still displays but disappears when set to gated mode. I tried adjusting the intensity and using the beam finder, neither helped. I opened it up and checked all the labeled voltage points I could find, all were good.

Any advice on what to check next would be greatly appreciated.

Thanks,

Shane


Roger Evans
 

In gated mode the readout displays between sweeps, so if you have nothing to trigger the sweep then you get no readout, I just checked this on a 7904.

It is quite easy to accidentally knock the AB horizontal buttons on the mainframe and also the small trigger select buttons on the plugin so that none of the buttons is depressed. If you have accidentally selected a mode which is not auto sweep or have nothing to trigger the timebase, then that could account for the sudden disappearance of the trace.

If you can provide an input signal, set the vertical and horizontal mainframe buttons appropriately, get the timebase to show the 'triggered' light and still have no visible trace then you have a more serious problem.

Regards,

Roger


sdferg7@...
 

Unfortunately the problem is just that. I can get a trigger light on both of the horizontals, with the display set to either CH1/CH2 or chop, but there is still no trace. With the side panels off I made a cursory inspection of the boards and the only thing unusual I could find was a capacitor with a strange glassy blue substance on top of it. I presume it is leakage, but I haven't seen one leak like this before. It is on the A25 storage board just behind the fan, so I'm not sure if that could be related. From what I can tell from the troubleshooting manual I think the problem is either in the CRT, the horizontal, or the PSU. I did notice the night before that in certain settings the display would flicker a bit, but as it warmed up the problem seemed to dissipate. Hoping the whole scope didn't become trash the day after I bought it, haha.

Here is an image of the capacitor: https://imgur.com/a/5a6zVPd

Shane


sdferg7@...
 

Actually, I just checked again and I see what you mean now Roger. While I still have not trace, if I switch to gated with the trigger light on the readout stays on. Thanks for the tip! Maybe this rules out the CRT as the problem?

Shane


Roger Evans
 

Sorry for the idiot level reply, it just seemed too much of a coincidence that you unplug a timebase and the sweep disappears.

Do you have another scope, also a high voltage DC probe capable of measuring 2kV?

A quick test which should spot a problem in the Z axis / unblanking circuitry is to locate R2135, CRT bias, on the high voltage board. Note its position, turn it slowly in each direction looking for anything on screen, and return it to its original position. I had a 7623A with several storage board problems but they did not affect non-store mode, that is not to say that there might be some issue on the storage board that would affect non-store mode. I would try the various storage modes and also the reduced scan to see if anything interesting happens and also simply to exercise switches which may be a slightly contaminated.

Regards,

Roger


Roger Evans
 

It's my turn to be the idiot! I had forgotten that the readout is still working. If the readout is in the right place and in good focus then all the high voltages are in good order and the X and Y amplifiers are working OK. You then need to check the sweep and unblanking signals are getting through so I would still say, turn the readout off completely, and tweak the grid bias. Beyond that a second scope really would be an asset.

Roger


sdferg7@...
 

Ok, sorry for my inexperience. Just to make sure, you're saying turn the readout knob all the way to the left to turn it off, then try twisting R2135 back and forth? Should I just move it a little or all the way? And I'm looking for something to appear on the display?

Thanks for the help,

Shane


Roger Evans
 

Turn the readout intensity fully anticlockwise, there should be a detent to turn the readout function off.

You need to turn R2135 slowly since if it does brighten up the missing trace you don't want to overdo it and damage the very rare 7834 storage CRT. It's probably clockwise to brighten the trace but if not try turning it slowly anti clockwise. You may see nothing at all or a trace with both X and Y signals or a dot or a line, it just depends on which of the X, Y and Z signals are missing by the time they get to the CRT. Make sure to return R2135 to its original position, you can do a final adjustment when everything is working. There are X, Y and Z signals to drive the readout, they clearly do get through to the CRT so the problem is to find the point where the main signals go missing.

Were both time bases working before the right one was removed?

Do you have a second scope for troubleshooting?

Roger


sdferg7@...
 

Alright, good news. Turning R2135 about 1/8 of a turn to the right brought the trace back. Thank you! Both time bases seem to work properly. Unfortunately this is the only scope I own, I've only used them at school before; if I really need another one I might be able to take it to a lab at my college when I go back at the end of the month, but I'm not sure if they'd let me use one for personal purposes. Is there more checking I need to do with regards to the trace issue? Not sure what would cause it to suddenly go out of adjustment like that. I also don't know the proper calibration procedure.

With regards to the storage controls, I've never used a storage analog scope before so I'm not sure if I'm using them correctly and I'm a bit afraid of burning the CRT with them. They all seem to turn the screen green at varying intensities, some at a worrying brightness.

Thanks again,

Shane


Roger Evans
 

My best guess is that there is a mechanical problem eg a switch or a plug/socket connection rather than a failed component. Switches may improve with use, plugs and sockets may benefit from removing and reseating or light cleaning with IPA or Deoxit. There is a thin coaxial cable from J2208 on the Zaxis board that goes to J4408 on the A6 Logic Board via the main interface board that the plugins mate with. If you can identify any of the plug/sockets on this signal and reach them easily, just pull the plug out and (carefully, it is easy to bend the centre connector) reinsert it.

If at any time the trace goes back to its original intensity then return R2135 to its original position (the exact setting is not critical, you should not be able to see the trace at minimum brightness setting on the front panel controls).

Can you now see the retrace (flyback) line where the spot returns from the right side of the CRT to the left side?

Do the A and B front panel brightness controls both work?

The advice I have seen here is that the safest way to use a storage mode CRT for everyday purposes is to use variable persistence mode with the persistence set to minimum. If you use normal non-store mode don't let the brightness increase to the point where the trace increases in width but not in brightness.

Regards,

Roger


sdferg7@...
 

I think the faint line between the peaks at the top is the flyback line (the trace intensity looks much higher in the image than it is in real life): https://imgur.com/a/KKS7vMp

I've discovered neither intensity control works, when either are moved to the maximum position their respective trace goes out of focus. No other change is visible when moving the knobs.

After some messing around, it does seem like the storage settings are working correctly. The only hiccup is the bright dot you can see at the lower left of the image. When I enter the save mode the image remains, but the dot creates a splotch that slowly radiates out. My work-around currently is to enter the save mode, then move the horizontal control all the way to the left so that the dot goes off-screen.

The scope also seems to have a long warm up time to me. The previous image has the calibration signal set to 4V after running perhaps 10 minutes. My other confusion, and I may just be the ultimate idiot here, is that the calibration signal seems to output half of what it is labeled to output. I confirmed this with an RMS multimeter. So the 4V setting outputs ~1.97V or so. It's a square wave so the RMS should be the same as the peak to peak voltage. The other decades behave similarly. Anyway, After running for maybe 40 minutes the scope begins to properly display 2V on one channel, then a bit later on the other. So I'm not sure if these just have a long warm up time or if something isn't right.

Shane


Roger Evans
 

It looks like the 'main' Zaxis signal, which contains both the brightness level and the unblanking pulse, is not getting through to the Zaxis amplifier. The inputs to the Zaxis amplifier from the readout system and from the 'main' signal source are completely symmetric so it looks as though it should be safe to swap the plugs on J2208 and J2203 on the Zaxis board, provided the leads are long enough.

I don't think its particularly safe to leave R2135 where it is, I would restore it to something approximating its original position. Do that first and then reinstate the readout display.

If you then swap the J2203 and J2208 leads and nothing changes it suggests there is no signal on the 'main' Z axis lead where it arrives at the Zaxis board. This could be a plug/cable fault so see if you can trace the cable originally on J2208 from the Zaxis board back to the A6 Logic board at J4408, probably via the main interface board. Check end to end continuity with your multimeter. If the cable continuity looks good then you may well have a faulty U4494. Again, if it is socketed, remove it, clean the pins and reinsert.

If you swap the J2203 / J2208 leads and the main signals work but not the readout then you probably have a fault around Q2216. Is it socketed?

If you or anyone following this thread think this suggestion is too risky (7834s and 7934s are quite rare) then please ignore the suggestion.

Regards,

Roger


Roger Evans
 

Can you just confirm which low voltage DC supplies you checked when you said they were all OK?

Roger


sdferg7@...
 

I checked the +130 and -50V on the A20 horizontal amplifier, the -50, +5, -15, and +15 on the A19 vertical amplifier, and the +15, +5, and -15 on the A13 readout board going to the powered probe connectors.

Shane


sdferg7@...
 

Ok, I have the top rear cover off, it appears the leads would be long enough to swap J2203 and J2208. To clarify, you're suggesting I return R2135 to roughly its original position, turn on the scope, turn the readout on from the front panel, turn the scope off, swap J2203 and J2208, then turn the scope on and check for a trace?

End to end continuity of the J2208-4408 connector is good. Q2216 is not socketed. Looks like I'll have to take out the PSU to get to U4494, but all the other ICs in the scope seem to be socketed, so I may try to get to that and clean its contacts first to see if that helps.

You say it may be risky, does anybody else following this have any more precautions I should take or other suggestions?

Shane


Roger Evans
 

As to your first paragraph, yes, that is the sequence I am suggesting. Give me a little more than 12 hours and I will get my 7934 out, check that the Z axis circuitry is a close match to the 7834 and try the cable swap on that.

Roger


sdferg7@...
 

Ok, I appreciate it. I almost wonder if a connection got jostled when I took out the horizontal plugin and put it back in; that would line up with the logic board.

Shane


Roger Evans
 

I have just done the swap of J2203 and J2208 on my (working) 7934 and both the signal trace and the readout still work after the swap and no need to adjust any brightness settings. The Z axis input on the 7934 looks identical to the 7834.

Regards,

Roger


sdferg7@...
 

Ok, I did the swap myself and saw no change in the trace or readout behavior. I now have the power supply pulled out to get to U4494. Unfortunately it seems to be thoroughly stuck in the socket, and I don't have one of those puller tools. There's not even a gap to get a screwdriver in, so I'm not positive a puller would be able to get at it. I ordered one that says it will get here Saturday, since I don't know of any electronics stores left around here.

Thanks,

Shane


sdferg7@...
 

Sorry about how many replies I'm making. I also noticed the PCB U4494 sits on is quite severly bent, as seen in this album: https://imgur.com/a/fsQ1DrN

I don't see anything shorted to the board behind it, but hopefully a trace isn't broken or anything. I don't believe there is high voltage on that board that could have arced across that gap; correct me if I'm wrong. Is there anything else I can try while waiting for the tool to be delivered?

Shane