Old Easter Germany Oscilloscope repair

[XPModder]

Hello everyone,
I recenctly got my hands on an old eastern Germany oscilloscope. Its an RFT EO 211 and it is in relatively good condition. When I got it there was only a little bit of dust inside. I cleaned it up anyways and tried to turn it on. The good thing is, that the CRT works perfectly fine and the powersupply also has all the correct voltages. I could confirm that the Y-Axis works correctly - at leastfor DC Voltage. Now the big problem is, that the X-Axis is not working correctly. To be exact, the beam moves from left to right correctly, but then it moves further right (out of sight and to the side of the tube) and then just gets stuck there. So the problem is, that the beam never gets moved back to the left.
When the scope is on manual X-input, the beam just stays in the center of the screen as there is nothing hooked up to the X-input. When I then switch to 30ms/T timing I can see the beam move from the left edge of the screen to the right edge, then moving further right out of view and never returning unless I switch back to manual X and it appears in the center of the screen once again. Yesterday I had the scope fully taken apart and did a visual inspection of everything. Then I also heated all the solder points and let them reflow with a bit of new flux, so it cant be a cold/bad solder joint either. After putting everything back together I tested it and it worked perfectly for about half a minute where it broke again. Now the fault is back in the exct same way it was before.

The attachements is a full schematic I found online. I have the user manual with component overlays as well, but its to large to attach it here.

I hope you can help me fix this scope.
Just before someone asks: I dont have another oscilloscope, so I cant verify the waveforms shown on the schematic.

[PA0PBZ]

Is the trigger set to AUTO? Did you try to feed a signal into the input?

[XPModder]

Yes the Trigger is set to AUTO and I just hooked up a 10V AC on the X-input, but nothing at all happend. So the beam just stayed centered on the screen.

[PA0PBZ]

Yes the Trigger is set to AUTO and I just hooked up a 10V AC on the X-input, but nothing at all happend. So the beam just stayed centered on the screen.

I was referring to the Y input. So if you switch the X to external do you have a dot on the screen? Can you move the dot with the X and Y position knobs?
What happens when you feed the Y input with a signal?

Picture for those playing along at home:

[XPModder]

When the switch is set to X I have a dot in the middle of the screen and I can move it everywhere on the screen using the horizontal and vertical adjustment knobs. When I feed a signal into the Y input the dot moves to the corresponding height (as specified by the V/T selector). It is still farely accurate when it comes to measuring the voltage. So the CRT and all the high voltage circuitry is fine. Therefore the problem is somewhere in the circuit for the timebase. It also isnt a problem with the selector for the timebase as the speed of the one sweep I get is faster when I quickly switch to 10ms/T  before the sweep is over. I am very sure that the problem is somewhere responsible for moving the beam back to the left edge of the screen.

[Kleinstein]

The description sound like a problem with the trigger, not necessary the time base generator.

The ext/it switch should be there to switch from internal to external trigger. Auto trigger could be broken too, so I would try manual trigger too. Having an external signal on the trigger input could also help.

[XPModder]

I agree that it probably is a problem with the trigger circuit. And as I said before a AC voltage fed into the external trigger input does not move the beam across the screen at all. Seeing that the beam moves across the screen once (when on internal trigger and a timebase selected) and the speed of it depends on the selected timebase I would think that the timebase generator is actually fine.
The thing that confuses me the most is that it breefly worked two days ago. Because of that I thought it might be a lose connection somewhere. It could be a thermal issue as well, as it worked only just after reheating and reflowing all the solder joints.

[Kleinstein]

If not used for a long time contact problems at the switches could be a problem. It may help to operate the switches some 10 times. Quite a few of the switch contacts are so called self cleaning, which means they clean by using them.

An internal connector could be another problem.

[XPModder]

The thing with the internal connector is, that there is no internal connector apart from the fuses at the back. Everything is just soldered in.
The switches should have been used enough by now to clean themself... I can however spray some contact cleaner on them and press them a couple times to make sure theire ok.

[pbarton]

Once the horizontal timebase circuit, ramp waveform has reached the right hand side of the CRT screen, the source of the horizontal timebase ramp waveform needs resetting.
It appears that the horizontal timebase circuit is not being reset.
Looking at the schematic, the reset may be performed by the diode GR2007 (SAY30) which is shown just below the timebase rotary switch.
I would investigate in that area, including T2014 (SC236E) which may reset the horizontal timebase ramp waveform.

[PA0PBZ]

The user manual is more like a service manual, it has the schematics, the block diagram and a description of all the blocks.

(In German, but hey...  )

[XPModder]

I do have the manual as well and have now measured every component that is mentioned in the paragraph. All the resistors are fine and I took most of the transistors out to test them and they seem to be good as well. The capacitors are ok as well, all though some measure a bit high, but that might just be my old meter having driftet out of spec a bit, as all the small caps with under 1uF are the ones that are a bit high...
I also took a good look at the board, as I found a single broken trace yesterday and apart from that one trace, that I got working again by soldering over the crack, I cant find any other problem with the board.

So it doesnt seem to be a problem in that part of the circuit.
The good thing is, that I should be able to measure all the signals using a good oscilloscope at uni tomorrow. So any suggestions where I should take a look at the signal, apart from the spots marked on the schematic?

[jdragoset]

If GR 2007 is in the sweep reset signal path, it is fed by R2075, which may have a non-contacting wiper.
Try a small movement. left and right

[XPModder]

I will try that.
I did find something odd, when measuring the resistors. Measuring R2056 in-circuit showed about 2,2k, while measuring it out of circuit showed that it is 30k as its supposed to be. So maybe one of the other resistors has gone short-circuit for some reason, as I cant find a path with about 2k from one end of it to the other...

[jdragoset]

If C 2020 were shorted and time base switch is in "X" position, 2200 ohms could be expected, depending on R 2054 setting.

[XPModder]

So C2020 does not seem to be shorted, as it measures about 10kOhms.
Some maesurements with an oscilloscope revealed, that the sawtooth waveform on the base of T2015 is not present. When switching from X to a internal timebase and the beam runs across the screen once, I get a single ramp of voltage on that point and then the voltage just stays on the higher level at a few volt. I also measured between R2043 and R2044 and the square wave signal is not present, instead there is a steady dc voltage of about 3 volt, that changes slightly with the position of the timebase switch.

[pbarton]

I now realise that the circuit formed by T301, T302, T303 and T304 is involved in conveying the fact that the timebase ramp has reached the right limit of the CRT, back to the resetting of the ramp initiation for successive horizontal sweeps. The gate of T301 has both internal and external sweep inputs. The collector of T303 feeds the junction between R2043 and R2044 (via C308/C309). Has the circuit formed by T301, T302, T303 and T304 (which includes a couple of slide switches) been checked?

[XPModder]

That circuit had not been checked before, so I tried to measure the componenets in-circuit with my multimeter. Some value seem to be quite a bit of... I will need to take the board out and desolder some componenets to make sure, that its not the other circuitry though...

[pbarton]

If you have an oscillator with a ramp, or triangle, or sine wave output, of about 4 or 5 volts and 1KHz and you apply it to the external sweep input BNC (marked 0,5 V/T) with the timebase knob in the 'external' position and the INT/EXT switch in the EXT position, in theory you will have a repetitive horizontal timebase trace on the CRT. No trace? Then follow the signal path to identify the problem?

[XPModder]

So I found out why the external x input was not working. A cable connecting the capacitor on the bnc to the pcb had become loose. I attached the wire to the correct point on the pcb again.
I then decided to just test all the components on the small daughter board with the switches on, as thats the board with the circuit around T301-T304 and all the capacitors seem to be fine and the resistors are also just fine. Some of the resistors measured very low in-circuit though. This then brought my attention to power-rails 5 and 13 as the resistors measuring low, where R304-R308. So measuring the resistance across C2019 in the power-supply area on the main board shows just 690 Ohms. Looking around the schematic further, I could not find any reason why the resistance would be so low here.
So maybe its a bad Capacitor?

Edit: A couple minutes later the resistance starts to climb, as if a capacitor is charging and slowly getting higher in resistance...

[pbarton]

Your capacitor is OK. Your meter is charging it up, which confirms that it's healthy.

[XPModder]

So further testing revealed that the two transistors T302 and T303 have broken down internally, wich allows current flow from base to collector. That explains why the resistors were measuring low when in-circuit as this provides a low impedence path across the transistors. I am not sure what the frequency these are running at is and I dont have npn-transistors that would match the old once.
So I cant put in new once to see if that fixes the problem...

[hli]

So further testing revealed that the two transistors T302 and T303 have broken down internally, wich allows current flow from base to collector.

These are SS216, which are "general switching transistors for digital applications". See e.g. https://www-user.tu-chemnitz.de/~heha/basteln/Konsumg%C3%BCter/DDR-Halbleiter/ for a (German) data sheet . Their transit frequency is quite high (350MHz), so you would need to replace it with some RF transistor. Just try it and see what happens (if the replacement is not fast enough it probably just means you lose on the higher frequency inputs)

[XPModder]

So its been quite a while since my last post here, but I am still working on this oscilloscope and since the last post I have replaced the two transistors and I have tested the X-input of the scope. The X-input works perfectly fine now and with an external signal hooked up I can get a horizontal line and even some waveforms to show up on the screen. So the problem can only be somewhere before the section used by both internal and external signal.