Tektronics 465 repair

[Ortzinator]

Hello all! It's been a long journey with this 465 ("low S/N") but I think I need help now. It's been a great learning experience but I'm still a little new at this.

I bought the scope from the original owner so it was in good working condition at first. But he said it hadn't been used since the 80s and unfortunately it didn't stay working for long. First it was a twitchy trace and then no trace. I replaced the filter caps as well as the diff amp for the 5 and -8 rails, which were found to be intermittently faulty. Now all the low voltage rails are within spec and have around 0.5mV or less ripple.

Then to fix the no-trace issue I swapped out the HV multiplier with a used one from eBay. Which brings me to the current problem. With full intensity I only have a soft glow on the right 2/3 of the screen. Pressing beam find I get a short vertical line on the far right edge that can be moved up and down with the position knob, although this line moves off screen as the scope warms up.

Shorting the left and right leads to the deflection plate together brings the line to the center of the screen. This led me to investigate the horizontal amplifier. The voltages here are all a tad off but around Q1274 they are way off. The base reads -2.4 and the emitter -3. Luckily it's a 2N2222 and I have a bag full of those but swapping it made no difference.

Unfortunately the only other scope I have to test with is one of those cheap handheld ones from China (2.5MS/s sample rate) so I'm not sure if that would be useful. I also don't have a HV probe to measure the -2450V rail.

I've been writing down voltages if you need to see those, and if you need any other pictures let me know! Thank you!

[TimFox]

-2.4 V on the base and -3.0 V on the emitter is a normal condition for a NPN transistor such as the 2N2222 (+0.6 V from base to emitter).
What voltages are indicated in the wiring diagram?

[Ortzinator]

-2.4 V on the base and -3.0 V on the emitter is a normal condition for a NPN transistor such as the 2N2222 (+0.6 V from base to emitter).
What voltages are indicated in the wiring diagram?

It should be 0V for the base and -0.7 for the emitter.

I've attached the horizontal amplifier schematic with my annotated voltage readings.

[TimFox]

Is CR1275 blown open?
It should prevent that emitter from going more negative than -0.7 V.
The purpose of that diode is to bias the base of the transistor to approximately 0 V--a cheap version of a long-tailed pair.

[Ortzinator]

Is CR1275 blown open?
It should prevent that emitter from going more negative than -0.7 V.
The purpose of that diode is to bias the base of the transistor to approximately 0 V--a cheap version of a long-tailed pair.

Looks fine to me. It reads about the same as CR1255

[Bud]

Check its resistor R1275.
With -2.4v on the base of that transistor, the diode CR1273 will be reverse biased cutting off the signal path.

[TimFox]

The -3V across CR1275 in that location is impossible if the diode is intact and conducting (forward biased).
The way the circuit is supposed to work is that the current through R1275 should flow roughly evenly divided between that diode and the emitter of Q1274 to get about 0V bias on its base.

[Bud]

Yes, should also check for cold solder joints on CR1275 PCB pads.

[Ortzinator]

I lifted R1275 and it read 223 ohms.

Also touched up the solder on CR1275 and the voltages didn't change. To double check, I lifted the anode and the diode mode on my DMM read 0.59V which matches the datasheet for a 1N4152.

I'm gonna try cleaning the sockets on all the surrounding transistors.

[floobydust]

CR1252 must be open. -11V cathode and -0.44V anode.
CR1273 must be open. -11V cathode and -2.4V anode.

[Kim Christensen]

Also touched up the solder on CR1275 and the voltages didn't change. To double check, I lifted the anode and the diode mode on my DMM read 0.59V which matches the datasheet for a 1N4152.

I wouldn't trust "diode check" readings from any DMM. The only time I find that diode check is 99.9% correct is when it shows a shorted diode out of circuit. Any other readings or situations are not a reliable indication of anything much.

TimFox is correct: 3V across CR1275 when it's forward biased in circuit means that it's faulty.

[Bud]

I lifted R1275 and it read 223 ohms.

Also touched up the solder on CR1275 and the voltages didn't change. To double check, I lifted the anode and the diode mode on my DMM read 0.59V which matches the datasheet for a 1N4152.

I'm gonna try cleaning the sockets on all the surrounding transistors.

The discrepancy in voltage drop across that diode when in circuit vs tested on the multimeter does not make sense. Try checking for continuity between the diode terminals and respective connection points, e.g. check if the grounded lead is indeed has continuity to the chassis. PCB traces may be broken.

[TimFox]

I lifted R1275 and it read 223 ohms.

Also touched up the solder on CR1275 and the voltages didn't change. To double check, I lifted the anode and the diode mode on my DMM read 0.59V which matches the datasheet for a 1N4152.

I'm gonna try cleaning the sockets on all the surrounding transistors.

Is the diode installed backwards then?  Again, 3 V forward-bias is not possible.
Note that CR1275 (biasing an NPN transistor) and CR1255 (biasing a PNP transistor) are opposite polarity with respect to ground.
The other possibility is mentioned above by Bud:  broken trace or discontinuity to the chassis or voltmeter common lead.

[floobydust]

[...] Shorting the left and right leads to the deflection plate together brings the line to the center of the screen. [...]

I don't think that was a good idea, it pits the L and R plate deflection amps arm-wrestling each other.
So I would expect new damage beyond the original problem, such as Q1282 or Q1262 roasted with more possible collateral damage.
During a repair don't short stuff out- it's an aggressive strategy and there are consequences with solid-state gear. Older tube scopes don't mind overloads and shorts but it's not something I would do here.

[Ortzinator]

I lifted R1275 and it read 223 ohms.

Also touched up the solder on CR1275 and the voltages didn't change. To double check, I lifted the anode and the diode mode on my DMM read 0.59V which matches the datasheet for a 1N4152.

I'm gonna try cleaning the sockets on all the surrounding transistors.

Is the diode installed backwards then?

Wow that was it! How embarrassing! I do now remember removing the diode to check it but that was a few months ago. I'll take this as a sign that I need to keep a log. (And pay more attention to the schematic)

The scope is still having problems but I have horizontal line now and the horizontal position knob works.

Thank you to everyone who replied!

[Kim Christensen]

The discrepancy in voltage drop across that diode when in circuit vs tested on the multimeter does not make sense. Try checking for continuity between the diode terminals and respective connection points, e.g. check if the grounded lead is indeed has continuity to the chassis. PCB traces may be broken.

The problem with "diode check" on DMMs is that it doesn't pass much forward current through the diode. 1-2mA is typical. I've seen this many times: Diode checks OK with DMM, but as soon as you pass more current through it, the forward voltage drop climbs to 3V and beyond.
Same with using "diode check" to check the reverse characteristics of a diode. The DMM will only put 2-3V across it. Your 50V rated diode could be breaking over, or be very leaky, at 10V but your DMM is never going to show that.
Diode check is a handy function, but the user should be aware of it's limitations and not take the indication as the gospel truth.

My next pet peeve is blindly using the beeper "continuity test" on a DMM. A 10-100 ohm resistance will make them beep, the display will be near zero, and the user moves on thinking there's no problem. The continuity beeper is fine for quickly ringing out wiring (while keeping it's limitations in mind), but the lowest ohms function should always be used for checking bad/suspect connections and traces.

Not a rant at you, but just at "diode check" in general. 
Anyway, the OP seems to have found his problem.

[Ortzinator]

Alright, I’m back after some more work. Thanks to everyone who has helped so far.

Right now I’m trying to fix a triggering problem. Instead of triggering near the center of the sensitivity knob, it only triggers with the knob turned far to the right and only in a very narrow position. It also only triggers with trigger coupling set to DC.

I probed around and found 4.5V (!) at R618 where the trigger signal comes into the trigger generator circuit (attached) after the coupling switch. Every coupling position reads 4.5V except for DC, which reads 80mV at that resistor. If this is a short, I cannot for the life of me figure out where the voltage is coming from. I fear it is under the switch itself but I removed the plastic slider and switch contacts looks clean and I can’t see anything like a solder bridge. I’m assuming removing these switches are difficult to impossible?

Any help would be appreciated.

[Kim Christensen]

Every coupling position reads 4.5V except for DC, which reads 80mV at that resistor.

When set to DC, what is the voltage drop across R618? How about R617? Both should be 0.0mVdc. Even a small drop of 0.5mV across either of these 100 resistors would indicate a problem from that path.

[Ortzinator]

Across R618 is reading 300mV and R617 is reading 0V. Does that mean Q622 is bad?

[Kim Christensen]

Most likely. Especially if the end of R618 that is connected to Q622's gate is more positive than the other end of R618. (Q622's gate junction should normally be reversed biased)

[Ortzinator]

I swapped the fet pair with the ones from the B sweep and it fixed the problem. Thanks!

[Ortzinator]

Now I’m trying to fix the fan. I thought it was fixed but after replacing the transistor array (U1690) with 4 discreet 2n2222, it works at first but after a few minutes the board starts to draw a lot of current and the motor stops turning. I tried replacing Q1698, which is a 2n2907, and also tried replacing the thermistor with a 5k resistor, none of which had an effect.

My only clue is that transistor D and A get hot, whereas B and C do not. In other words the transistor in those positions, even if I move them around. However, it doesn’t seem like there’s any imbalance voltage-wise between the two hall sensors.

Edit: Sorry, forgot to attach the schematic

[Kim Christensen]

Problem might be the hall sensor associated with the D & A transistor pairs not providing enough base current. This would result is more voltage drop across the transistors creating more heating.
It would be interesting to compare the average current though R1693 vs R1694 by measuring the voltage across each while the fan is running.

[Ortzinator]

It’s hard to get exact readings cause they move around, but the base voltage on D and A are within 5mV of each other (about 1.1V) but the emitter-collector voltage drop is 100mV more on D than E.

I’m just going to assume the motor is broken and move on. This scope still has other problems.