Tektronix 475 scope: repair required.

[Skywave]

Originally Posted by PJL

Hi Al,
How accessible is it?
Might be easiest using another scope and look for the sweep on base Q1124 then collectors Q1124, Q1134, Q1152, Q1162.

Fairly accessible. However, at the moment, looking at a.c. waveforms isn't really going to be of much use, since the d.c. biassing conditions are all awry. Obviously, these need to be correct prior to investigating the a.c. operating conditions.

Al.

[PJL]

I was assuiming that some of it would be biased OK. Can you fill us in on what you have discovered so far. Have you got the DC voltages for above? I am sure that the combined efforts of those on the forum should be able to crack this.

[mhennessy]

Hi Al,

Unfortunately I'm swimming in a sea of 1960s radios at the moment, but if you're still struggling in a couple of weeks time, let me know. I have a fair amount of luck with these sort of things ("luck" being the key work there ), and I like a challenge!

All the best,

Mark

[coil1234]

I hope I'm not telling you something you have tried already, but the standard way of tracing imbalance in these stages was to short the differential signals together, moving through the amplifier, thus forcing them to be the same potential and enabling you to find the faulty stage.

This was a Tek method and was certainly applied to the older 5XX series valved & Hybrid 'scopes. HOWEVER I'm not sure if would be safe to apply to a transistorised 'scope with presumably lower impedances and a large imbalance.

Apologies if you've considered this already, I've had my share of battles with Tek gear, great when working but the occasional fault can be a right B*****

Regards Mike A

[Skywave]

Mike -
Without going into the full-bloodied analysis, the appropriate circuitry starts with a single-ended drive source, a section of d.c.-coupled differential pre-amplifiers, ending in d.c.-coupled differential power amplifiers that drive the LH and the RH deflection plates.

The single-source drive point normally sits at 0v d.c. I isolated the drive source and joined the amp. input point to gnd. / 0v. I disconnected both deflection plates from the O/P of the finals. This procedure was duplicated on a fully-functional 475: the O/P correctly sat at half mid-rail on this 'scope, but on the faulty unit the O/P stage remained turned 'hard-on', due to a biassing fault somewhere in the very complex d.c. couplings between driver input and final O/P. I was able to get close to determining where the excess current was flowing, but was unable to determine why. All active devices had been exchanged between the two 'scopes without revealing any faulty active components. All Rs and Cs checked out as good. Therefore, all electronic devices (in the affected areas) had been eliminated. All the PSU rails were established as O.K. & present.

Obviously, I'm missing something. But my experience of fault-finding, so that I can proceed further, regretfully does not run to black magic! Hence, my 'shout' here.

Al.

[Skywave]

Originally Posted by mhennessy

Hi Al,
Unfortunately I'm swimming in a sea of 1960s radios at the moment, but if you're still struggling in a couple of weeks time, let me know. I have a fair amount of luck with these sort of things ("luck" being the key work there ), and I like a challenge!
All the best,
Mark

Thank you, Mark. I may well be contacting you later then to discuss this.
Regards,
Al.

[Patrick Dixon]

Originally Posted by Skywave

All active devices had been exchanged between the two 'scopes without revealing any faulty active components. All Rs and Cs checked out as good. Therefore, all electronic devices (in the affected areas) had been eliminated. All the PSU rails were established as O.K. & present.

I have no wish to intrude on your pain, and I don't know anything about the circuitry involved, but the logical conclusion of what you say is that a connection is broken somewhere. If you meter across groups of components maybe you will find it - it's possible that by metering each component, you are making a broken or dry joint.

Sorry if I'm teaching you to suck eggs, but sometimes saying something obvious (or stupid) helps!

[coil1234]

Hello again Al,
as previous post , trust I'm not teaching you to suck eggs (strange expression) , clutching at straws here, but I have had (Very painful) experience of switch leakage (in my case a Tek 546). Try lifting one end of R1109 (cct 9) this switches on the X10 mag neon, very slight leakage (from a 110v supply) here may upset the d.c. bias conditions. You never know.

Regards Mike A

[PJL]

I had some fun recently with dry joints on a scope tunnel diode trigger circuit. I wasted a lot of time measuring and looking at waveforms and finally realized that pressing the probe on the board was intermittently fixing the problem making the readings useless.

Anyway, how about a few voltage readings at the points suggested?

[Skywave]

Originally Posted by Patrick Dixon

I have no wish to intrude on your pain . . .

it's possible that by metering each component, you are making a broken or dry joint . . .

Sorry if I'm teaching you to suck eggs, but sometimes saying something obvious (or stupid) helps!

Sentence 1: all contributions are welcome!
Sentence 2: that, indeed, is a valid consideration.
Sentence 3: Couldn't agree more! Apologies not required.

It's a very rare human who has acquired a near-infinity of experience and that, as a consequence, can confidently pick up any item of electronics kit and bravely state:
"Step aside! I can fix that! Just watch!"

Al.

[Skywave]

Originally Posted by coil1234

Try lifting one end of R1109 (cct 9) this switches on the X10 mag neon, very slight leakage (from a 110v supply) here may upset the d.c. bias conditions. You never know.
Regards Mike A

Again, another valid contribution; thank you.

However, in my case, since I had disconnected the 'signal source' from the I/P to the diff. amp. for fault location analysis, this seems unlikely, since the switch to which you refer is prior to the point of disconnection.

Al.

[davidw]

I don't know if your scope has PCBs but I have had a fault that cleared by application of heat (hair dryer). Component changes made no difference and it turned out to be the PCB had become conductive (probably due to being kept in an outside shed ).
Just a thought ...Good luck.

[Skywave]

Yes, David - I have met that problem before & it may well explain the fault and fit the 'known history' of this item. There are points of distinctive high-impedance in this part of the circuitry which would, therefore, cause problems.

OTOH, I do recall that when it was first switched on from cold (many months ago), the trace did initially appear - then it coughed & spluttered for a bit - then the X-deflection collapsed entirely. I wonder: damp in the pcb that was then 'moving about' / drying out? However, it's been in the warm for quite a long time since, so . . . .

Al.

[dave cox]

Al, unlike some tektronix manuals this one doesn't have any expected signal annotations so I've put some estimates down for the first part of the H amp with what I would expect the quiescent / no signal conditions to be. I’ve assumed the first LTP is going to be balanced with the spot in the centre of the screen. Does this look sensible / similar to what’s happening in the real world ?

By the way, does the beam finder switch have any effect on the display and / or operating points ?

dave

[Skywave]

Dave -
The base of Q1172 & Q1182 were found to be correctly at 0v. d.c. with the fault present. In the faulty unit, I joined these points to 0v. / gnd. to investigate the subsequent stages: the fault condition (as previously described) in these subsequent stages remained the same.

Under 'normal conditions' - i.e. that hook-up, as above, not fitted - the beam finder had no effect.

Al.

[dave cox]

I guess that leaves the two deflection plate drivers.
Don't you just knew the fault will be in the most complex part ?

Q1182/1184/1188/1254/1212/1214 look like an X20 inverting amplifier with an adjustable offset controlled by R1256 feeding some current into the 'virtual earth' between R1253 and R1258 (which seems to control the other driver as well)

I was guessing the upper driver is similar but how it works has me stumped at the moment. There doesn't appear to be any drive to the output stage ...

I would be looking very carefully at that biasing / cross feed circuit around R1256, especially the current in R1258.

dave

[glowinganode]

Originally Posted by Skywave

Dave -
The base of Q1172 & Q1182 were found to be correctly at 0v. d.c. with the fault present. In the faulty unit, I joined these points to 0v. / gnd. to investigate the subsequent stages: the fault condition (as previously described) in these subsequent stages remained the same.

Under 'normal conditions' - i.e. that hook-up, as above, not fitted - the beam finder had no effect.

Al.

That would suggest the fault is beyond the base of Q1172/82.
If you're not happy to let us look at it, then can you give us the electrode voltages for Q1172/74/82/84/88/1202/04/12/14/54/94 under the fault condition. I know this is a lot of readings to take, but without a feel for how the circuit is sitting, then any guesses are going to be stabs in the dark.
Rob.

[Skywave]

Originally Posted by dave cox

I guess that leaves the two deflection plate drivers.
dave

Quite: with regards to the layout on the cct. diag., it is the O/P stage to the upper plate that is turned hard on. The lower plate is biassed correctly. The actual deflection plates were temporarily disconnected for fault investigation.

Al.

[dave cox]

Nope. I can't figure that upper driver.

Am I missing something or is the schematic incomplete ?
It really needs something driving the bases of the OP devices to set the dc operating point or it will just flop-over to one of the rails.

dave

edit.
OTOH, bleeding current into R1209 could alter the balance in the OP stage with the feedback loop operating as normal ...

[Skywave]

Originally Posted by glowinganode

That would suggest the fault is beyond the base of Q1172/82.
If you're not happy to let us look at it, then can you give us the electrode voltages . . . .
Rob.

Hi Rob,

I'm quite happy for anyone who wants to 'have a look at it'; hence my OP! .
No, seriously though, I've boxed it back up for now (having taken a deep sigh & made that OP of mine ). Right now, the bench is fully occupied whilst I complete an overhaul of a Bush DAC11 wireless; hopefully, not much more to do on that.
Then I'll get back to this 'scope and make measurements / re-read my notes, etc. - come back here and reveal all - well, almost all!

Then the fun can begin.

Al.