Need help repairing couple 'scopes (different faults)

how do i test it to verify it works as intended?

Click to expand...

Apply some known voltages/waveforms at the Y input. Does the trace show what vertical deflection it should?

i suppose you mean to use scope as normally, and does it show what it should? yeah, at first, when fault hasn't occured. when fault occurs however, then display gets distorted. i'll try to get video of sort
oh, and do i observe screen in x-y mode, or in normal? sorry, i just want to be sure to measure right things....

edit: managed to take video, hopefully it explains issue better. i'll post it as soon upload is done to onedrive

here's the video:
**broken link removed**

Fez:

Do you have another scope to take p-p AC ripple readings of those supplies before and after? Particularly, the +12 before the fault.

sure, give a moment. (3 working scopes, spare ones to blow up, just kidding here...)

My thoughts (Thinking out loud), The 12 V supply does wierd things. If it's flakey, it may show up with increased ripple even when it seemingly behaves. Now, it could mean something is taking th 12 V supply down.

-16v: 2v (seen even on dc-mode on scope)
+12v: next to none, ouple mv
-12v: same as +12v
+140v: none

huh? weird, i wonder did i measure it right, i always thought i have understanding of measuring ac ripple: scope channel is set at AC setting, ground clip to reference earth (kikusui's chassis) and ''hot'' to test point. and probe doesnt have x10 on, double checked

but, about your guess on 12v: there was one melted off resistor very near -12v test point, also, other solderings are/were cooked-looking, along pcb. heat haze is felt, no matter fault state or not

Which resistor on the schematic? In circuit resistance?

You missed the 5V supply where I would expect more ripple, but low mV is good.

oh, sorry, 5v is very low ripple. only ripple of note was that 16v
resistor is r623 (pcb marking) resistance is marked as 2.7k (two resistors in parallel) the melted off one was melted off under pcb (solder side) so i piggypacked it on top of its pair, and secured with zip-tie. now it at least won't run loose if it melts off and cause s/c

fez: I haven't had a chance to print the schematics out and I was out of it for the past couple of days. When it rains, I get migraines and yesterday was a particularly bad day. it usually takes me at least a day to "recover".

take your day buddy, i too have migrens occasionally, its then situation: gotta go to dark room and just rest it off. it's insane pain!

edit: i mean take your time, not day...

Fez:

I don't think this is going to be an easy fix. I'm only going to be guessing. I think the video suggests a problem in the horizontal or the +12 V supply. I can't really see any evidence of current limiting in the +-12V supplies.

So, I'd like you do one or two power on type tests. Compare the ripple on C626(*) and C627 before and after a fault with a scope. If they are way different stop. Look at the base to emitter voltage of Q604 and Q605 and compare, Use a handheld battery powered meter. If you can compare the ESR on the caps, great. Look at the ESR of C628 (+12 supply). Compare the temperatures of Q604/Q605 right after a fault.

Place the scope in x-y mode with the inputs in the gnd position and turn down the intensity. Try putting the dot in 3 positions:
a) full left
b) center
c) full right

See if anything happens in fault mode or when the fault is supposed to happen. What I'm HOPING for is the fault won;t occur when the beam is centered and will occur when it's to one side. leave enough time between tests so the scope will go from no-fault to fault.

Then with the unit off, start at Y+ and Y- on the schematic and compare the B-E junction voltage as measured with the diode test of a DVM to the symetric counter-part. e.g. Q3028 to Q309 etc.

With the controls set at ground, in x-y mode and the beam centered and the intensity way down.
You may want to to look at the the voltage BETWEEN the Emitter of Q105 and the Emitter of Q106 and the emitters of Q308 and Q309 and see if there are changes between no fault and fault.

With the power supply being off balance,

PS
Printing the schematic was a pain, I did page by page.

The idea is to use symmetry and the B-E junction voltage as a measure of the health of the transistors. I'm going on the premise that it's a horizontal amplifier fault and you don't want a steady horizontal signal. With the +12 V supply being affected, it could still be the supply.

Replacing the high value caps in the power supply may not be a bad idea. I've seen lots of issues where the negative supply ripple on say U603 can wreak havoc on a system.

woa, thanks a lot for such list. Appreciate it a lot if someone has spare time to help out, rookie as i am.
yeah, printing was pain for me too...
i'll do my best to measure those

sadly, nothing fishy of note yet,

KeepItSimpleStupid said:

Compare the temperatures of Q604/Q605 right after a fault.

Click to expand...

temperatures are steady between those, before fault, haven't tested it during, hard as i need to flip scope over to be able take a shot with ir meter.

KeepItSimpleStupid said:

So, I'd like you do one or two power on type tests. Compare the ripple on C626(*) and C627 before and after a fault with a scope. If they are way different stop. Look at the base to emitter voltage of Q604 and Q605 and compare, Use a handheld battery powered meter. If you can compare the ESR on the caps, great. Look at the ESR of C628 (+12 supply)

Click to expand...

all transistor's base-emitter tested out-circuit, as well as those caps ESR. bigger ones were 0.1R and smaller 1R. transistor's b-e voltage were pretty much ~0.7v.
ripple was 2vpp at caps, in normal state. not tested yet in fault state

My recent experience (today) repairing my own scope is don't trust transistor b-e b-c readings as to whether the tranny is good. I just had to replace one that looked fine when tested with the meter, voltage levels didn't indicate a faulty transistor either, but the scope only worked after I replaced it. So (based on this experience) I would try replacing the suspect transistors anyway.

sounds familiar....trouble is, haven't located faulty section yet...
can you tell about that your situation with scope today?

2 V p-p might be too big. This is on the +12 and -12 supply caps? Do the waveforms look different?

throbs: The 12 V supply oscillates badly at fault, supposedly.

Don't worry about pulling stuff out to test just yet.

KeepItSimpleStupid said:

Don't worry about pulling stuff out to test just yet.

Click to expand...

and i thought ESR and most transistor's tests are out circuit....hehe

KeepItSimpleStupid said:

2 V p-p might be too big. This is on the +12 and -12 supply caps? Do the waveforms look different?

Click to expand...

oh yes, main filter caps, and no, waveforms are like sawtooth, only bit rounded on edges, like normal ripple (at least i think so....)
cap voltage is seen at photo

oh, and caps voltage oscillates same time as fault ''blinks''

fezder said:

and i thought ESR and most transistor's tests are out circuit....hehe

Click to expand...

Well, the diode test was basically invented to test transistors in circuit. The ohmmeter doesn't work. Hence, I said to "compare" readings. Later, the circuit is basically symmetric so any extraneous circuits might affect your measurement, but the effects should be similar. If you find a b-e junction that's zero, it may be the surrounding stuff causing it. If complementary placed ones are say 0.2 V, note it, but don't worry about it yet. If the complementary ones were 0.7 and 0.2, then you need to investigate. But, don't get too far ahead.

Cut of circuit tests are far more accurate, but don't get out your Swiss Army Knife yet.

Not sure how to separate a power supply issue from something down stream. You could TRY to freeze spray the op amp in the power supply with freeze spray when it acts up.

This issue is tough and I'm not making any promises, but we don't want to make it worse.

I wonder if we should shelve this until the main filter caps are replaced?