[SOLVED] Help with Analog oscilloscope problem

[hendrix]

Hello,

I have a kinda old GwInstek GRS-6032A oscilloscope and I am having a problem with it, was wondering if I could get some help on repairing it.

The problem started the other day when one of the probes touched a live AC line thanks to my cat jumping from place to place. The circuit braker went off and when I turned it on afterwards, the screen of the oscilloscope started flickering. I can still get measurements but, the screen keeps flickering. The service stated that the tube was fine but the board needs to be changed with some other things and asked for 500 bucks to repair it.

I was not able to find the service manual for it, so I am having a hard time locating the problem. It would be great if I could get some help or maybe someone could direct me towards a service manual. Thank you.

Regards,

[AllanMN]

I can try. Others may be better.

Describe the flickering. Is there a progression to it's returning? (grows vertically, horizontally or from dim to bright, even very, very fast?)

 I'm thinking it's a power supply dropping out, like the screen high voltage. I see it's listed as 2 Kv which seems like it could be carefully measured with a voltage divider setup and a DMM. Of course, KNOW THAT THIS IS DEADLY VOLTAGE, so... don't touch it!

[hendrix]

I will try to explain it, if that doesn't work I ll post a video.

The flicker is almost like, if I can make you imagine this, that there is another layer under the screen and on that layer imagine a video of sea waves, every one or two seconds, the waves on the under layer seem to grab what's on the top layer and pull it towards the right side of the screen (horizontally), then it all comes back. Every couple minutes or so, this flicker stops and everything goes back to normal for a few seconds then it starts again, it doesn't do it periodically, its inconsistent. It sometimes comes back if I lightly whack it, but its not consistent.

I dont know if that means anything to you but, thanks anyway.

[hendrix]

Ok, found the problem, its a connector on the bottom board that has a loose connector somewhere. When I press on the board slightly, the flicker goes away. However, the service has totally screwed the focus on the tube, there are some trimmers that change the picture, anyone know what I should be fiddling with:

- ADD BAL
- ASTIG
- CRT BIAS
- V CENT
- V GAIN
- H CENT
- H GAIN

There are also some trimmers labeled with number:

- VR3321
- VR3311
- VR3301
- VR3322
- VR3313
- VR3312
- VR2101
- VR2102

Anyone know what these mean? Any help would be appreciated. Thanks.

[cybermaus]

I am not an expert on this, but on my old & cheap analog scope, I found it was better to clean and refit all the connectors then to start trimming stuff.

In fact, I had already started trimming to get focus back, but the range if the trim was not enough. But after ensuring all connectors were clean and snug, I had to trim them all back. I should have started with the connectors, not with trimming. Luckily, I had marked all trimmer's original positions with a fine marker pen

[tautech]

Best advice it to only clean and check connections. You could re-seat any socketed devices. Get Service and User manuals. It may have PS voltages printed on the PCB that you could check and adjust on the PSU board. Maybe even check PSU for ripple in case of dying PS Caps. That's as far as I would go without manuals. It is likely its calibration has drifted a little but it will still be very usable if not the adjustments are not mucked with!

[GreyWoolfe]

  on the cleaning/reseating connectors.  I checked out a Kikusui o-scope that the display would fade out after about 20 minutes.  I wiggled off and reseated all connectors and the problem went away. 

[SoundTech-LG]

From your description of blowing the circuit breaker (at your panel I presume)I would be looking for scorched traces, particularly the ground traces. If a big fat one is blown off, a smaller one may be trying to carry the current, or maybe not at all. That would certainly explain why your screen probably appears to be mainly 60 Hz components. Look closely at the path of least resistance from your probe ground to the mains ground plug. Power off and continuity test from probe ground to mains plug ground. Should be near zero ohms.

The current from your live wire traveled from the scope probe ground through the path circled at the inputs to the power plug ground (lots of current). I'd bet you find your problem in that path.

[hendrix]

Just a progress report.

I took apart the oscilloscope, took pictures of the whole process of course:) I did not see any separate ground trace, the ground of the BNC input connectors are connected directly to the chassis; the whole chassis is used as the ground. Thats why the scope still works, well kinda. Anyway, I am looking carefully at the traces and components to see if anything is out of the ordinary. I will then lightly sand the plug type contacts and the put it back together. Any advice while the thing is open?

Thanks.

[grumpydoc]

I will then lightly sand the plug type contacts and the put it back together.

I would not "sand" any contacts, it will ruin any plating and not be good for longer term reliability (if you manage to fix the 'scope).

Clean with IPA, if tarnished a contact cleaner or something like "deoxit" could be used. People use to say use a pencil eraser but I think opinions on that vary.

[hendrix]

hmmm...ok no sanding then:)

[hendrix]

Thats all she wrote...problem solved...saved about $550...woot:)

So, the problem was in the high voltage transformer, as suspected but, what was not suspected is this piece of **** quality control over at Gwinstek being so poor. I opened up the small metal box to find a crooked mounted transformer. the transformer was almost touching the metal housing and was creating an arc. I suspect, this was giving the earth of the scope a signal that was messing with everything, including the tube.

The picture is below, you can see in the top part of the housing where it is almost touching. There is a tapped hole, probably made to heatsing a 220 case regulator, which was not used in this scope, and that part was arcing to the transformer, what a piece of **** job.

Well, hope this thread helps someone else out some day.

Thanks everyone.

[SoundTech-LG]

It appears in your photo there are two very close spots there. Also there is a regulator or something mounted to one of them. I arrowed the 2 spots, and the reg.
I wonder why the casing of the transformer would be arcing though. Normally that is isolated from the internal windings, but I guess if it is soldered to the high side circuit, and the shield is ground/neutral, that would explain. Does sound like a POS. I guess your cat with it's live wire antic began the process of the arc???

[hendrix]

The photo might be somewhat misleading, because its not on an angle. I wanted to show the arcing spot, so I tried to shoot it from the top as much as possible without creating shadows, thats why the screw, for the regulator mount, seems to be on the same level with the transformer and touching it. The screw sits above the transformer in actuality and does not touch the transformer, I changed it anyway and put a shorter screw, just to be on the safe side. The metal box is grounded, as you can guess. I do not know how the actions of the cat; the probes touching live 230V and all this had to do with the arcing, but my guess is that the 230V 16 Amp electrical current from the mains, passing through the ground, right before the braker went off, was enough for the current to go through the very thin coat of potting varnish that the transformer had on it. When the varnish disappeared, the arcing began between the ground and the transformer, thats my best guess.