Tektronix 468 repair/refurb

[tautech]

You'd wonder if all the 74* IC's have been damaged by PSU problems at some stage. 
If this IS the case, then what else is hurt too? 

Getting to the bottom of all the lingering niggly faults can be very frustrating and I feel your pain.

I'll be away for a few days and mightn't be much help, although it seems you're on the right track IMO.

[Addicted2AnalogTek]

well well, another MAJOR failing on the part of Tektronix in the 468 design and manual!!!!

The manual explains how to switch the scope into test mode to test the functionality of only the basic scope functions - but they FAILED to provide ANY power to the resistor that then turns on the logic for CHOP, CH1, and CH2!!!!!!    it specifically says to UNPLUG the jumper that powers the storage mode power supply - the power supply that powers that damn resistor!!!!!!!!    GAHHHHHHH!!!

SO -   I am going to rig up a lead to put 5V to that resistor and get things moving along...

one would think that Tektronix would have noticed this dilemma in the... oh..... 30 years that this scope was in regular operation?    or AT LEAST added that info to the manual!!

[Addicted2AnalogTek]

S for STORAGE power supply!!

P3006 being the jumper from the transformer that has it's fuses based on A12.

DUM-DA-DUM-DUM-DUMBBBBBBBBBBBB

[Addicted2AnalogTek]

You'd wonder if all the 74* IC's have been damaged by PSU problems at some stage. 
If this IS the case, then what else is hurt too? 

Getting to the bottom of all the lingering niggly faults can be very frustrating and I feel your pain.

I'll be away for a few days and mightn't be much help, although it seems you're on the right track IMO.

I'm hoping that the problems I am having with the logic circuits boil down to a few 74*s and some out of tolerance resistors causing low voltages.  I think I may have located a resistor network that is out and causing oscillations to the 74*s already. 
There's just sooo many little piddly issues, and what happens is I'll be tracking one down and find another and get distracted from my initial search, and nothing gets done.  I'm really trying hard not to let myself get diverted, but make sure that I am also not ignoring little issues that may be adding up to the bigger ones.    Then there's the waiting for parts game, like my trigger JFETs - I bought 4 pairs, but the guy accidentally put all 5 remaining pairs together in one bag.  So instead of telling me that early friday to get my response, he waited until right before leaving work to tell me and now I have to wait several more days.  At least I'll have 5 pairs for 2.50 a pair, even if I have to rematch them.

Well I hope you have a good time on your trip (or whatever it is you are doing).  I'm sure I'll have a pile of interesting issues listed and hopefully sorted by the time you return.

[tautech]

Fear not, I'll be keeping an eye on you,  just not as often. 

[Addicted2AnalogTek]

it's so much easier to debug the basic circuits of this scope with all of the digital & logic completely disabled... I've already found exactly where the channel to channel crosstalk is coming from.   Both Q285 & Q245 are leaking signal into the -8V rail, and oddly enough, channel 2 is worse.  It's coming directly off of Pin 3 on both Fets. 

[Addicted2AnalogTek]

the 100x attenuator for CH2 failed.    I have a couple 100x attenuators, but they are model 307-1014-02 and it needs 307-1014-05.   anyone see a problem with using the one I have for the time being? they're both for 1M 20pf scopes, and I'm not sure what the differences are internally...

[Addicted2AnalogTek]

well this is very strange, or very simple....   It seemed like the 100x attenuator died while the scope's digital section was disabled... I thought maybe if I put it back together it would be fine, and in the meantime I opened up the 100x attenuator and cleaned it (there was actually a little bit of some kind of gunk on top of the precision resistor parts).  Now that it's back together, it works fine, but it appears that some of the finger contacts have gummed themselves up again, or the 2X attenuator is going funky on me.   I find this very odd, but only in 10mV/div does it do this - it will flatline the signal like the attenuator is dead, but there will be huge spikes with short burst ringing. 

I mean, it has to be the contacts, but can they really dirty themselves up again that freakin fast?!?!?  It was 2 weeks ago that I spent nearly 3 hours on each channel cleaning ONLY finger contacts!   
I guess this is something to be fully aware of when restoring any of these old tek scopes!

[Addicted2AnalogTek]

A quick update:

I am checking devices on the Vertical Mode Switch board since this seems to be where a number of my issues are originating.  This includes the non-working trigger view, the 25khz 80mv ripple hitting the IC's on the vertical preamp board, and others. 

Q2 (Pin 5) of U310 has a connection to pin 1, 2, 10, 11, and 20 when measured in resistance.  When measuring for V drop in diode setting, there is a 2.85V drop between pin 5 (+) to all of the other pins. This is not the case with any of the working selectors.   

This is the only IC I have fully checked out thus far.  I really don't know much about logic circuitry, but from what I do understand thus far I am quite certain that this is a problem - most likely the problem why trigger view does not work, and why the voltage at the vertical amplifier pin for trig view. 

[Addicted2AnalogTek]

I have completed the trigger view repair and made progress on the 110V unreg loading and bad PSU waveforms. 

I have discovered that the PSU waveforms are directly related to the CRT bias adjust and the majority of the CRT circuit.  I replaced Q109 and it immediately made a difference in the low voltage side of the HV CRT transformer.     I know that Q1109 is dying, but.... the person out in CA that I ordered the RCA 2N3055's from is too cheap/lazy/stupid to properly package them in anti-static bags for shipping, so they were DOA. I wanted him to send replacements, but he was too lazy to test them quickly with DMM and obtain an antistatic bag for shipping, so he decided to give me a refund instead. (NEVER buying from them again!!!!!) So now I have to wait several more days before I can get that issue completely tidied up. 

I have managed to get Trigger View functioning fully again by replacing Q114, which is what was destroying my NORM trigger signal. 

There is still plennnnty more to do on this pup, so I'll keep posting updates as I make more progress. 

[Addicted2AnalogTek]

can anyone tell me what it means when the display is not centered about the horizontal graticule line with the input coupling set to Ground?  The manual says absolutely nothing about making an adjustment to it, so I assume that means there is no adjustment and it is a fault somewhere, such as the main input capacitor being bad? 

This offset appears to be an issue throughout most of the vertical preamp and is causing many of the waveforms to have offset baselines.

Of course as soon as I switch it to AC or DC it goes to center, but that's because I adjusted it to be there.....

[Addicted2AnalogTek]

I just double checked, and now I see that it is only doing it when I have no vertical selected, but have triggered the sweep..... so that must mean there is a voltage difference between the two horizontal deflection plates - thus pulling the sweep towards one of the plates instead of it being centered.  The voltages to the plates confirm this, but I'm not 100% that I am correct in thinking this. Anyone feel like confirming or denying ?

[tautech]

I know that Q1109 is dying, but.... the person out in CA that I ordered the RCA 2N3055's from is too cheap/lazy/stupid to properly package them in anti-static bags for shipping, so they were DOA.

I'm flabbergasted, I've never considered jellybean bipolars, particularly the venerable 2N3055 to be delicate in this way. 
Sure if abused, you can bugger most things, but I reckon he was trying to rip you off with duds.

I just double checked, and now I see that it is only doing it when I have no vertical selected, but have triggered the sweep..... so that must mean there is a voltage difference between the two horizontal deflection plates - thus pulling the sweep towards one of the plates instead of it being centered.  The voltages to the plates confirm this, but I'm not 100% that I am correct in thinking this. Anyone feel like confirming or denying ?

Sweep waveform spot on to spec?
IIRC other scopes I have fixed had both a balance and gain adjustments, balance maybe  set in XY mode and gain sets trace sweep width to 10 div long/wide.

[Addicted2AnalogTek]

The A sweep start and B sweep start waveforms are off, and I am working on solving that. I am fairly certain that the issue is originating from the trigger gen circuit.   I also cannot reliably do any kind of voltage measurements or waveforms that rely on any output from the vertical preamp other than directly thru to logic then the vertical amplifier, and I am waiting for the parts to fix it. (151-0333-00) they should be here monday according to USPS, but who knows.

The rotten apples are Q154 & Q155, and are not properly amplifying the input signal. The DC voltages are spot on or very close, but once signal is put into the scope the waveforms up to TP5 & TP6 are good, but those two are 1.5V low and their waveforms are 10% to 20% low as well. If replacing those does not fix it, then it's CR258 & CR258.  But the problem is almost the exact same on both channels, so I believe it is the transistors.

Also I had mentioned that the input JFETs are likely on their way out.  I am not 100% certain on this.  I could only get one new 151-1032-00 (only one I could find--- ANYWHERE!!) from Greece.  However, according to the following site:
http://www.reprise.com/host/tektronix/reference/transistor.asp
that dual JFET is the exact same as the 151-1090-00, so I may order some of those.  I'm also looking into buying a 464, 466, or 468 parts scope.

Speaking of JFETS, I received my 151-1042-00's for  the trigger circuitry.... but....    I have 8 matched but mixed up pairs to re-match, and one proper pair but TO-106 package.  I think that their stripes are painted after they are matched, but I still think I should build a test circuit to confirm, such as this one in Post #9:
https://www.eevblog.com/forum/projects/how-to-match-a-set-of-jfets/

Then once I have the input to the trigger circuit back in order I could work on the sweep and some of the z-axis.  unfortunately the horizontal amplifiers voltages are taken in XY mode so I can't do that yet. 

At least I'm gaining momentum!!! this definitely is a beast of a repair/refurb for a beginner's learning project, but I'm gettin the hang of it! 

[Addicted2AnalogTek]

the one thing that I am confused about with the curve tracer circuit is this...

I understand that IDS goes to Channel 2 (or Y) of my scope, but where does the X signal come from?  Do I just tee off the signal from my function generator for VDS or VGS (depending on the measurement I am taking)?  The schematic and article aren't clear about this.   
I believe that I am correct in regards to teeing off the signal from the FG, but I'd like some confirmation that I am doing this right..

[tautech]

A few posts back you described problems with trace position when Gnd coupled.
In the schematic portion above there is a 20K pot for Vertical adjustment.
It's likely to be the master adjustment and the others you have adjusted just for the relevant attenuators. RTFM. 

[Addicted2AnalogTek]

In the post after I said that it was off when the channels were in the GND position I corrected myself and said that they are fine in the ground position because I adjusted them to be, but that it is off when there is no vertical selected, but I have triggered the sweep.   When doing that, the horizontal line that is displayed is about half a division above the center horizontal graticule line, and changing the timebase has no effect on this at all.   When the DC voltages were measured on the vertical plates and horizontal plates, the voltages were not the same on either side as they should be.   Also in the post where I corrected myself, I said that I thought the difference in the horizontal plate voltage would be the cause of this - and then I learned a little more about CRO CRT's.   The horizontal plates are horizontally opposing (vertically positioned).  The vertical plates are vertically opposing (horizontally positioned).  therefore, if anything is causing an offset in the horizontal positioning of the triggered sweep, it would be the difference in voltages at the vertical plates - thus pulling the electron beam more towards one of the plates.  In my case it is the top plate.    Anyway, this "issue" is likely nothing to worry about and will most likely be resolved once I finish fixing the vertical section.

On another note, I have the JFET matching circuit finished and need to know the pinout of these 151-1042's... In the picture below I labeled what I believe they are, but am unsure of whether I have the Source and Drain backwards or not.  I know technically the JFET doesn't care much and it probably doesn't matter much for the test circuit, but I would like to know anyway.   In the link below, if I am correct about the pinout, that would mean it is similar to the JFETs of this type that were manufactured by Central. Most of the others have it arranged the other way around with the Source as the middle pin. Anyone care to shed some light on this?   I've been researching JFETs a lot and think I have it right but I want to be sure.

http://alltransistors.com/mosfet/transistor.php?transistor=19124


[EDIT]  Had it backwards. the pinout on these is D S G.

[Addicted2AnalogTek]

Just replacing those trigger JFETs has the rectifiers running quite a bit cooler.  A trigger/sweep is doing great even though the waveforms are still just a little bit off.  B trigger/sweep is partially back to life but still has some big problems that need to be resolved.  The A trigger controls are affecting the B trigger/sweep, and I can only get the sweep to trigger with a certain level setting.  The problem doesn't seem to be in the adjustments so it seems I have some investigating to do. 

I've got my fingers crossed that at least my order of IC's and transistors makes it here tomorrow so I can make some progress on the vert section and open up some more areas to check/fix. However it will be another week or two before my 1032's arrive, but hopefully these originals have enough kick left before they bite it to get me til then.

Since I built those 2 JFET measuring circuits I was able to measure the old ones and see just how bad they had become..  They were still alive, but very unhealthy - especially when compared to the entire batch of TO-92 type and TO-106 type new ones.  I'm sure I'll be finding at least a few more battered and fried transistors in this pup..

[Addicted2AnalogTek]

I'm baffled.... 

I kept finding transistors that we're on their way out, so I went thru and replaced all of the 2n3906, 2n3904, 2n2222A, 2n5551, 2n2907, etc etc.  I had plenty high quality Moto & Raytheon ones I got for cheap so I figured what the hell - why not.   I also replaced several zeners - a few of which were very unstable, and some that seemed OK but too close to their 5% tolerance rating for comfort. 

What has me baffled is this -    after the repairs when I initially started it up, the PSU waveforms looked like complete hell.  2x to 3x proper amplitude and way, way off of the expected waveforms.  As it warmed up a little and I turned on the trace, the waveforms just magically decided they were ready to go to spec...    WTH is going on?!?!?!     I know these old analog scopes take a little time to warm up, but I have never heard of the PSU waveforms being crap on startup - especially not like they just were. 

The only thing I can think of is that there are still a few semiconductors somewhere that are acting up and don't like to function properly until they are warmed up...    The rectifiers are running a whole lot cooler now and the scope seems to be acting up a lot less, but there are still some bugs to be found and solved.  ( I have not yet received my new trigger IC's or vertical input FETs and diodes ).   

Anyone out there that can shed a little light on this matter for me?

[Muxr]

I'm flabbergasted, I've never considered jellybean bipolars, particularly the venerable 2N3055 to be delicate in this way. 
Sure if abused, you can bugger most things, but I reckon he was trying to rip you off with duds.

There is a special place in hell for people like that. I've bought components before where they were either obviously fake or used, pulled from gear before, and I am fine with that, goes with the territory. But knowingly selling bad non working parts and presenting them as good, is low. This seller knew it I am sure.

[Addicted2AnalogTek]

in hindsight, yeah they probably knew.    It took several messages before they finally backed off the "No one else is complaining" & "well how do you know they are bad" rhetoric and stated they would send two more - but when I demanded they test them first and properly package them in a static shielding bag they resorted to " Dear buyer, we have decided to give you a full refund to make things simpler.."

It's the only seller on ebay with a large number of RCA 2n3055s, and there's a picture of numerous ones packed in their foam packing.  If you find that seller and look at their negative feedback, there are several cases where they sent broken & misrepresented items. Then there are several replies to the neg feedback stating " Block this buyer! SCAMMER! "  etc etc.

 There should be a list of rotten ebay sellers who try to rip people off. Of course that won't happen here, though, because ebay/paypal has the backs of those type of scum.  That is why I am giving you all the info I can for you to identify them without actually identifying them.   It should be easy enough to find them now....... only a few sellers of RCA 3055's, and only one with over 70 or so sold (last I checked). 

[tautech]

I know these old analog scopes take a little time to warm up, but I have never heard of the PSU waveforms being crap on startup - especially not like they just were. 

Anyone out there that can shed a little light on this matter for me?

I wouldn't stress too much about it, unusual as it sounds. Remember there is usually a stated time in the Service manual that the scope must be running for before accurate measurements or calibration procedures be commenced. This applies to all scopes, CRO's and DSO's.

Curiosity might make me monitor power-on mains currents accurately or just with a dim bulb tester.
Are the same rogue waveforms produced at re-power after the scope is warm?
It does sound unusual. 

[Addicted2AnalogTek]

the incredibly bothersome PSU waveforms are only present when the scope is cold-started.   After about one minute they go to normal.   

After changing out all of those cheap transistors, the psu waveforms for the 110 & 55 are amazingly better. Now there is just a little "lump" on the peak at the beginning of the discharge parts of the cycle.  I think that the waveforms are good enough now. Maybe they will improve to exact spec once I've resolved the few other issues, or maybe it's due to the small amount of ESR I measured in the 55V filter cap. (It's not much at all - maybe one-quarter ohm up from the others, which are all very low.) 
Another thing I have noticed that is a direct result of the fresh semi's throughout is that it is much, much easier to pick out and locate problem areas.  I'm thinking that the crispness of the switching is making the slow & leaky guys stick out like a sore thumb.  I'm going to continue to shop around for deals on bulk of the remaining semi-parts. Soon I'll have this scope running like a champ again with many, many more years of use - and a huge stockpile of parts!

The biggest problem that still remains is the CH1 pickoff outputs to trigger, X, and storage that are still garbage. I have a couple new 151-0434-00's (2N4261) en route from Greece that will hopefully crush that problem. If not, I will need more 0434's (freakin 12 on the vert board in total - at $6 each... MINIMUM!!!) Hopefully that clamps down on the drifting horizontal (in XY mode), which travels >4 divisions as the scope warms. Then that should allow me to discern what is the cause of the bad horizontal CRT plate voltages which are high a few volts, differ by 5 volts at best, and I believe is also the cause of the 110V rail being pulled UP (oddly enough). I've changed the resistors that set the 110V, so I know it's not that...   
Annnnd last and possibly least, the only rectifier still getting really hot is the 15V.

[cd_edwards]

How did you know your jfets in your input amp were flaking out? I just purchased a used 466 scope. I've got a good trace on channel A with a proper amplitude in 5mv from the calibration source, however channel B is only about 1/10 of the amplitude. I've only moved the probe from A to B channel, so nothing different there..

[Addicted2AnalogTek]

your best bet is going to be to check the waveforms at the test points on both channels and compare.  If the waveforms for the entire channel 2 section are bad, then your problem is near the input area.   Your problem could be anywhere from directly at the input to somewhere in the vert switching logic where the signal for both channels gets combined for the vertical amplifier.    I would mess with the gain adjustment for CH2 and see if I could get it back to spec first. but if you have to max out the pot to get the correct amplitude, you've definitely got something wrong.

As for me being able to tell that my input FETs are flaky:  I took them out and tested them after noticing that the gate of the B unit was leaking oscillations into the -8 rail, and the test point waveform at the gate to A (from channel input) was horrible. 

Another thing you should do is check different V/div settings to see if any are good.  If they are all off in the same manner, then you probably don't have an attenuator problem.