Tektronix 465M Repair

[sean0118]

Calling all Tek experts. 

I'm a bit stuck. 

I recently bought a Tektronix 465M for $50au, it was meant to be working, got it home and it doesn't work. That's ok though, as I already have a rigol scope and it gives me a chance to improve my testing and repair knowledge.

Following this manual.


All the LEDs come on but the CRT is dead. On closer inspection I found fuse F558 is open, this fuse connects the 32V-unreg to the HV transformer T550 (schematic on pg.166). I can't find any faulty discrete components and I'm starting to think the problems in the HV module...   ...which is potted...   

Although, I did notice the voltage applied to the fuse is 55V, is this higher than normal? Should it be closer to 32V? I noticed there is +52.2 written in one of the schematics above the 32V-unreg line, is that referring to the expected voltage?   

[tautech]

Normally the primary of the HV switcher is unregulated and the PSU specs will tell you what it should be.
The switching bipolar is under a bit of stess and they are known to fail, but also check the primary and secondary windings for shorts.

Also check this recent thread:
https://www.eevblog.com/forum/repair/tektronix-465/

[poida_pie]

Even though the 465 sounds very similar to 465M, they are completely different beasts.
I agree with Tautech, check the HV transformer. I used a LCR meter to confirm winding inductance and DVM for resistance and shorts.

I have a 465M that I repaired, the symptom being fuse F558 that blows. This "32V unreg" is actually closer to 50V.
Eventually I found that in my case cap C522 was bad, very low resistance. This caused the 2 transistor inverter (Q552, Q556) to run way too hard, blowing one and/or the other transistor. And replacing components is a bugger since the circuit board is rather old school with easy to lift traces...
When debugging this, I found that using a current limited 30V supply, bypassing F558 and set to maybe 100mA to be very educational. It permitted me to run this at a current that did not over heat the transistors but still see where the excess current was being drawn from. It was reassuring to see the good HV transformer provide correct but scaled voltages.

I bought a Tek 455 (50MHz brother of 465M) and again there were HV issues, this time a dud HV multiplier.
Whilst sorting that out, I blew the HV transformer. It's very easy to do this so be careful.

[tautech]

I'm very happy to hand this one to you poida_pie, sounds like you've been there before. 
 

[sean0118]

Thanks guys 

Ah, ok so it is meant to be near 50V, well that removes one possible fault...   


I'll check the HV transformer, I have a multimeter, but no LCR meter. Do I just check for shorts between the primary and secondary? Or are there other tests to do as well?

When debugging this, I found that using a current limited 30V supply, bypassing F558 and set to maybe 100mA to be very educational. It permitted me to run this at a current that did not over heat the transistors but still see where the excess current was being drawn from. It was reassuring to see the good HV transformer provide correct but scaled voltages.

I was thinking about doing this, but I don't really want to go poking around the HV part while it's live. Unless you mean the voltages are scaled down to a safer level?

[tautech]

I'll check the HV transformer, I have a multimeter, but no LCR meter. Do I just check for shorts between the primary and secondary? Or are there other tests to do as well?

Resistance of both, note them.
Shorts between (unlikely) and to laminations.

[sean0118]

Eventually I found that in my case cap C522 was bad, very low resistance. This caused the 2 transistor inverter (Q552, Q556) to run way too hard, blowing one and/or the other transistor. And replacing components is a bugger since the circuit board is rather old school with easy to lift traces...

Actually, I hadn't considered the fault could be in that part of the circuit. I've checked everything around Q552 & Q556, but now I'll go look around C522. Out of interest, if it's shorted does that make Q526 sink way too much current? 

edit: Also, I meant to ask if a faulty/damaged CRT could cause this fault? I tried sliding it out but it's stuck.

[sean0118]

Ok, I'm still not sure if I tested the transformer correctly, but heres what I found with reference to the schematic:


Pins 4&5: 0ohm (primary, winding 1)
3&6: 0ohm (primary, winding 2)

1&10: 142ohm (secondary)
1&2: 3ohm (secondary)

4&10: 11Mohm (primary to secondary)
3&10: o/c (primary to secondary)

I don't really know what values to expect for these? Is the 142ohm value too high? With regards to shorts to the laminations, the bobbins  ferrite, that means there's no laminations to test right? Unless I'm missing something (highly likely  ).

Also, good news! While poking around C522, I noticed that R523 was an odd colour, some scraping with the probes showed it was completely toast! Now I just need to investigate further and find what toasted it. 

[poida_pie]

Eventually I found that in my case cap C522 was bad, very low resistance. This caused the 2 transistor inverter (Q552, Q556) to run way too hard, blowing one and/or the other transistor. And replacing components is a bugger since the circuit board is rather old school with easy to lift traces...

Actually, I hadn't considered the fault could be in that part of the circuit. I've checked everything around Q552 & Q556, but now I'll go look around C522. Out of interest, if it's shorted does that make Q526 sink way too much current? 

I found that to be the case. After fussing over the Q552/Q556 part for hours and hours, I found that C522 is in series via a few other that connects +95V to -5V and some resistors nearby were stinking hot.

I recall from a good look at the tekscopes forum at yahoo on this and the Tek 455 failure modes and they usually involve
- HV tripler
- HV transformer
- that cap C522
- C582 and or C584
- C548
- Q522 or Q556
It seems the order of events is failure of one of the above 5 options, which then cause the inverter LV oscillator transistors to overload.

Always the symptom was the blown 1/4A F558 fuse.

edit: Also, I meant to ask if a faulty/damaged CRT could cause this fault? I tried sliding it out but it's stuck.

Rarely do these CRTs fail in this way (i.e. present big-ass current sinks to the HV supply. typical anode current is of the order of 10s of uA)

When debugging this problem, I found it a great reason to purchase a Fluke 27 F/M with HV probe from ebay. But then I spent all my time looking at the 95V output, thinking that if there is 95V on pin 2 of T550 then the 200V, 2000V and approx 3000V on pins 8,9 and 10 should be nearly there. Assuming the U550 module is all OK...
So no need for the HV to repair it!

This happened 18 months ago and my memory is a bit rusty AND I am not an EE, just a hobbyist.

[poida_pie]

Ok, I'm still not sure if I tested the transformer correctly, but heres what I found with reference to the schematic:


Pins 4&5: 0ohm (primary, winding 1)
3&6: 0ohm (primary, winding 2)

1&10: 142ohm (secondary)
1&2: 3ohm (secondary)

4&10: 11Mohm (primary to secondary)
3&10: o/c (primary to secondary)

I don't really know what values to expect for these? Is the 142ohm value too high? With regards to shorts to the laminations, the bobbins  ferrite, that means there's no laminations to test right? Unless I'm missing something (highly likely  ).

Also, good news! While poking around C522, I noticed that R523 was an odd colour, some scraping with the probes showed it was completely toast! Now I just need to investigate further and find what toasted it. 

Well I think T550 is good from what I see from your measurements.
It's most excellent you found a toasted R523, I recall one of those resistors around there in my device was discoloured from overheating but still functional and close to correct value. (close enough for government work anyway)

Lift a leg of C522 and see if it's got any resistance?
My dud cap was something like 10 ohms. I can't recall and I don't take notes

[sean0118]

Ok, I think C522 might have been the culprit!? 


It measured around 8k in circuit, but I unsoldered it and placed it in series with a resistor. It had massive leakage even with only 5V supplied!

Although, I might have ruined it by not using the correct polarity (is it polarised?). Even if it were polarised surely 5V reverse bias wouldn't hurt a 500V rated cap?

[poida_pie]

Ok, I think C522 might have been the culprit!? 


It measured around 8k in circuit, but I unsoldered it and placed it in series with a resistor. It had massive leakage even with only 5V supplied!

Although, I might have ruined it by not using the correct polarity (is it polarised?). Even if it were polarised surely 5V reverse bias wouldn't hurt a 500V rated cap?

It's clearly not polarised, according to the schematic. Any polarised caps on it have a little "+" on the symbol. 5V test voltage will not hurt these caps.

I replaced it with a similar value polyester 100V rated part. It worked fine.
jaycar has these available on a Sunday.
http://www.jaycar.com.au/Passive-Components/Capacitors/Greencaps/0-0033uF-3-3nF-Greencap-Metallised-Polyester-Capacitor/p/RG5033
Maybe it's better to get something with higher voltage rating, I do not see any need to go for 500V

[tautech]

edit: Also, I meant to ask if a faulty/damaged CRT could cause this fault? I tried sliding it out but it's stuck.

Please understand, these things are delicate inside, everywhere.
DON"T rush in.
Study, consider, consult.

Failure to be very careful will have you buying another scope, been there done that. 

[sean0118]

It's clearly not polarised, according to the schematic. Any polarised caps on it have a little "+" on the symbol. 5V test voltage will not hurt these caps.

Good point, I must have missed that, in my defence I did notice there were "+" printed next to the polarised caps on the PCB. 


One thing I've learnt from this is not to trust resistance measurements of suspect capacitors. It was originally 8kohm in circuit, after testing its leakage it fell to 400ohm, now after sitting overnight it's up to 800ohm. It must temporally heal itself?

Please understand, these things are delicate inside, everywhere.
DON"T rush in.
Study, consider, consult.

It's ok I was very gentle and only applied light pressure. I think there's some rubber seals at the front it's stuck too. Although, I think I should ban myself from working on scopes after midnight...   


EDIT: Ok, this capacitor never ceases to amaze me. I played around with testing its leakage a little more and removed the series resistor, and it quickly stopped leaking! Now it doesn't leak even with 50V applied across it, it's capacitance is now bang-on and its resistance is to high to measure with my multimeter...   but I'll replace it anyway to be safe.

[tautech]

I really did mean everything inside the cover is delicate, proceed with real care.

[sean0118]

Ok, I replaced the cap and supplied the 32V unreg from a current limited power supply... and it worked!

Well, a dot appeared on the CRT and the power supply was only supplying 0.1-0.2A. I didnt have time to hook up the deflection yokes, I'll post a proper update in the next few days.

 

[sean0118]

IT'S ALIVE!!!   

[tautech]

Nice. 
You have one of the classic Teks, look after it.
Always use only enough trace brightness (Intensity) to do the job at hand and then adjust out of Focus if there is more to do.
This will preserve the CRT for a long time.

But do remember to use her every month or two to keep her in good shape.

[poida_pie]

I quite like my 465M, the crt draws a nice sharp trace. No fans, little weight, it just works.

So it was the cap I suggested you look at?

[sean0118]

Nice. 
You have one of the classic Teks, look after it.
Always use only enough trace brightness (Intensity) to do the job at hand and then adjust out of Focus if there is more to do.
This will preserve the CRT for a long time.

But do remember to use her every month or two to keep her in good shape.

Thanks, yep it's in a good home now. I'll give it a clean and do some calibration at some stage, but first I have to have something to calibrate against. 

I quite like my 465M, the crt draws a nice sharp trace. No fans, little weight, it just works.

So it was the cap I suggested you look at?

Yep, the cap is the only thing I've changed, hasn't blown a fuse yet.

It needs a bit more work though, still have to replace that burnt resistor. Also, the red 'VAR' on the right side has been pushed in (the metal plate it attaches to inside is bent), it must have been dropped at some point. It's missing a bunch of little things too, rubber foot, plastic cover on handle pivot and even some screws that hold the emi shield. 

The horizontal TIME/DIV knob is a little intermittent too, sometimes when turning it the waveform is shown at the wrong frequency. For example I think it selects 5uS divisions when the knob is set to 1uS, but then you turn it again and it goes back to being correct. I need more time to test it further though.

Also, it's got 'VP-24' written on the top, could mean anything, but being the military model I like to think it was used by the US Navy VP-24.