Tektronix 468 repair/refurb

[Addicted2AnalogTek]

I've definitely isolated the majority of the loading (besides the 110/55 from the CRT) to the regulated supplies to the vertical preamp.  I pulled up the jumpers for the 5V and -8V to the switching logic, and nothing changed. Isolated the Vertical Output Amp, no change... Vertical Mode Board... no change...  I also isolated the 5, 15, and -8V on the trigger board as well (in case it was related to the signal interconnection) and nothing changed.   Nothing got any worse when I changed all the 2N3565's.  I had overlooked some negative offset in several waveforms across channel 1 before, so I am going thru the schematics before I start testing again.   

My initial observations regarding what to look for:
1)  The one power supply rail that is capable of loading down all of the others is the 5V rail, and it's pass transistor's Vbe is the highest at 1.34V.  Something directly off of the 5V rail is likely what I should look for.
2)The Vbe of all of the other pass transistors is almost the same at about 0.62V to 0.64V, except for the 110reg which is close to folding. It's Vbe is 0.54V.  From what I understand, these transistors should all have a Vbe in the low mV - such as 3.6mV, which is the value when the Vert Preamp board is isolated.
3) Previously (before parts swaps) the worst part of the channel seemed to be closest to the output signal to trigger. I think I should work my way back from there. 

Now I have something else to fix while I wait for the CRT replacement parts.  YAY!   

[Addicted2AnalogTek]

now that I've had some time to think, maybe this problem is why the modifications were added....   the majority of them are to the vertical preamp board..      I need to do the mods anyway. I found and bought the exact specified zeners, so I'm going to go order the resistors now.   the other mods don't sound as important, so I will deal with those a little later.  There's also a full page of potentiometers that get changed as well........ I really hope that those have already been dealt with!!!!!

[Addicted2AnalogTek]

did old school zeners like the original ones for these tek scopes have a diode drop around .6-.7 like a regular diode??

Im testing out the trigger board, and there is a +8V rail that gets split off the +15 via a 6.2V zener.  Thus giving me 8.6V due to the .2V drop of a modern zener.  this +8V is used to feed the Fet pairs and the trigger level pots, and I am wondering if I should find a diode that will get me much closer to an actual 8V or not.   (I'm having some issues with the trigger circuitry, obviously.)

The original zener was only clipping off 5.9V so I installed an actual 6.2 on point zener.  It seemed to make a very little difference in the waveforms and voltages. 

[Addicted2AnalogTek]

well, only 1 out of 4 of the trigger FETs are good.   I've been playing with them a bit, and I am able to turn them off and on with my finger, and they also turn themselves on sometimes.  When they are turned on, the voltage drop across Source and Drain starts at almost 2V and steadily drops to about 0.2V and stays.  Sometimes the V drop will climb a few mV and then drop a few - back and forth.    Only the 1 JFET on the B trigger seems to function "properly".

I'm pretty sure that these aren't supposed to function this way if I am reading the datasheet correctly..  Their part number is 151-1042-00 and generic part number is 2N5454.  I found some for pretty cheap so I bought 4 pairs.  If I happen to be wrong and these are OK, at least I'll have some of these anyway since quite a few scopes use them.

[Addicted2AnalogTek]

I found the bastard that was dumping SMPS frequencies into the 5V rail on the vertical board.  Good Ol' 10 cent 2N3906 being used as an amplifier.  It's emitter was leaking the chop freq back into the 5V.   Of course the DMM reads it as good still, even with the resistance measurements.  I really need a real transistor tester.....

That output goes directly into the CRT circuit and is at the same frequency as the "double waveform" I was seeing at TP86, so I have a feeling I am about to solve that issue.   I remember saying " the second waveform is the CRT blanking waveform, and it stays there when the CRT unblanks. " I just didn't put it together that CRT blanking is CHOPPED blanking! (sad, because it's right there on the schematic!!!!!)     

Now that I know what I am looking at and looking for, I should be able to identify more leaky junction transistors that have decided it's time to break down.   

[Addicted2AnalogTek]

I think I may have found a way to tell when a diode is leaky using only a dmm.   While I was messing around testing some different NPN and PNP ones with hfe, diode, and ohms settings, I noticed that when I have it in Ohms connected across a junction, the resistance increases when the transistor is cooled and decreases when it's warmed up.   I noticed that the resistance did not drop when my leads were connected across the base/emitter.  It stayed almost completely stable while I held it between my fingers and when I blew on it.     I guess I could have gone a step further and did some tests with freeze spray, but that would be a waste. 

I have no idea how valid the results of this test would be on other leaky junction transistors, but I thought it might be worth sharing my observations in case it might help someone else out who doesn't have any other way to test transistors either.

[tautech]

I found the bastard that was dumping SMPS frequencies into the 5V rail on the vertical board.  Good Ol' 10 cent 2N3906 being used as an amplifier.  It's emitter was leaking the chop freq back into the 5V.   

Thats when you wonder if the 10M resistor has been doing it's job. 

What is the voltage past the 1.2K ? In excess of 2N3906's capabilities?

[Addicted2AnalogTek]

I'll do some more testing and measurements in the switching logic circuit in the morning and get back to you on that.   it's already 2AM... time flies when you're havin fun!

[Addicted2AnalogTek]

Finally just got back to testing this thing, and right off the bat I believe I may have found another rotten apple.   Switching Logic as well, but it seems to be originating from the 74LS08N  -  U508.

It is on the 5V input at pin 14.  It's also on all of the 5V inputs, but the magnitude of the oscillations are much, much higher on U508 and U409.  (Their V inputs are tied together: U508 Pin1 & U409 Pin 1&16) The amplitude is a few mV higher at U508 pin 14. The main frequency is 125Khz @ 60mV p-p. As shown, there is also some 3mV p-p 250Khz.
Also, the 5V rail is lowest at U508 Pin 1.  I believe this to be the lowest point on the entire vertical board. 

Picture 1:  20mV/DIV @ 1uS/DIV.   
Picture 2: 10mV/DIV @ 1uS/DIV & B timebase @ 100nS/DIV.

468 setup:   
CH1, CH2, CHOP
20 mV/DIV
A HORIZONTAL
.5 mS/DIV
AUTO, AC, NORM Trigger

No Input signal. CH1 & CH2 separated by 2 DIV.
Also CH2 has some small oscillations riding it, making it a bit bigger (was not able to adjust out w/ HF & LF reject pots/caps).  Not too worried about that right now though.

Since these IC's are <$1 each from Digikey, I may as well replace all 3.

It's starting to make some sense now why I had to disconnect the entire vertical section to unload the PSU rails...

[tautech]

Just check the logic levels are as expected in case some erroneous levels are causing the additional current draw. 

[Addicted2AnalogTek]

there's definitely some variance in the High voltages and low voltages, and some of the waveforms are off (acc to the manual's specified waveforms).  What worries me is that U408 gets fairly hot when the scope is on long enough.  I'm really hoping that has something to do with U409/U508/U509 and there's not something wrong with it.    I found one just in case, but it's $40.  I'm really wishing that I would have bought that other $100 468 so I could make 1 out of 2.....     

I could have bought a fully working 468 for what I've spent, but then I wouldn't have a pile of extra new parts, the experience & gained knowledge... and of course the fun of figuring this all out and fixing it.

[tautech]

I could have bought a fully working 468 for what I've spent, but then I wouldn't have a pile of extra new parts, the experience & gained knowledge... and of course the fun of figuring this all out and fixing it.

Or more importantly, you won't find one in as good condition and as reliable as when you have it finished.
BER is the term commonly used, but who gives a shit when the knowledge gained and gratefully shared is so valuable. 

[Addicted2AnalogTek]

Or more importantly, you won't find one in as good condition and as reliable as when you have it finished.

That's definitely true.  Is BER an acronym for something?  Can't think what it would be..


I don't know if/how the 3rd unit of the 74LS175 (U409) is terminated, but the voltages on those pins are allllll over the place!! 

U408 seems to be good still - possibly more resilient since it's a Cerdip and made by TEK, not Signetics like the others. I'm pretty sure U409 is bad though.  the 3 digital channels between U408 & U409 have erratically varying voltages acc to my Fluke 87III, and D2 channel has noise on it.  What makes me think that it's not U408 is that the voltages are much more stable on it's digital pins, and a bit higher as well.  Also less noise.
There's quite a bit of noise/variance on U508D, which I have no way of telling if it's that unit itself or the outputs of U409.

There's also quite a bit of noise/varying voltages on various pins throughout the chopped clock, and at the input to the first NAND gates of the chopped clock - the ALT/CHOP Select from the vertical mode switch.  This makes me think that not only are U508 & U509 bad, but one or more ICs in the mode switch circuit are likely bad.  Fortunately I left the shield off of it so I could poke around in there while it's running.

Another thing is that the external triggering and trigger view circuits don't work. I'm pretty sure I've already got that sorted for the most part though - just waiting on parts.  I'll add to that info once I have them.  For now I'm going to try and confirm my suspicions about the vert mode board and chopped clock and see where I land.

[tautech]

Or more importantly, you won't find one in as good condition and as reliable as when you have it finished.

That's definitely true.  Is BER an acronym for something?  Can't think what it would be..

Beyond Economic Repair.

There's also quite a bit of noise/varying voltages on various pins throughout the chopped clock, and at the input to the first NAND gates of the chopped clock - the ALT/CHOP Select from the vertical mode switch.  This makes me think that not only are U508 & U509 bad, but one or more ICs in the mode switch circuit are likely bad.  Fortunately I left the shield off of it so I could poke around in there while it's running.

Decoupling caps?

[Addicted2AnalogTek]

no, can't be decoupling caps.  I've replaced almost every single one.  there's 2 51pf and a 47pf 100V left in the entire vertical section that I haven't replaced, and they are glass encapsulated. I checked them, but my meter cant accurately measure anything under 100pf. All of the ones in the logic circuit and mode switch are new.

BUTTTTTTT.....    I was just reading a little into the new 87III thread and saw your post about the cracked traces/components and decided to look over it all again.  I found one cracked resistor in CH1 so far.  It was really hard to spot with 2.5X mag.  It's a really, really small crack and it's diagonal off of one corner. It wasn't bad enough for me to find with the pencil-eraser-pushing test, but probably bad enough to make some noise.... so I'm going to keep looking.    I took a new xacto blade and gently ran the back edge of the tip over the crack to confirm it was really a crack and not just paint.  it definitely catches on it.    I'm sure I'll be finding more cracks around.   hope I can find the values I'm going to need!!!

[tautech]

I was just reading a little into the new 87III thread and saw your post about the cracked traces/components and decided to look over it all again.

Wise IMHO.
I'll spend some real time inspecting a PCB, get a comfy chair and some magnification and you'll be surprised what you can find, partly dry and cracked joints, solder balls, cracked or improperly soldered component mountings, cracked joints on headers(not uncommon) etc.
Often identified deficiencies don't matter much in the scheme of things, it's just good to tidy things up and know they're right.

[Addicted2AnalogTek]

I'm definitely going to go over it again.  2.5X mag will have to do. I wish I had 5X.   
I don't think that resistor had and effect on CH1's performance.  The crack seems to be more of a superficial thing.  I did turn it into a chip, though.... it's going to get replaced anyway, but I'm reinstalling it for now.

HOWEVER.............  I did find a real cracked resistor right there, SMACK next to the position op-amp that I replaced!!!!  bad part about it is that I had already pulled up one of it's leads to work around it and never even noticed it!   I bet this is the oscillating chump that is causing all the fuzziness of my CH2 trace!  of course it's an 8.06K 1%!!   haha   at least it's film.... I'll just buy a bunch of some value close to it and measure them until I find a match... unless I can get a few for cheap off Digi or maybe ebay.... 

I'd like to try and put together as much of a comprehensive list as possible by tomorrow night so I can get the majority of my parts early next week.  Of course there will be a few parts here and there left to get.   I've already been searching for and buying piles of the heavily used transistors in these scopes just so I will have them.

[Addicted2AnalogTek]

This is what the display looks like on the 468 with 1Mhz  20mV  positive going square wave input.   The output of the Wavetek 145 is clean acc to my 465.  Adjustments don't do much, and it's there no matter what vertical setting I use or whether I use A or B triggering.    I just noticed it and thought I'd post a couple pics before I start looking for the source.   
I should have checked this thing out more thoroughly than I did, but hey.... always learning. 

Anyone have the 'light bulb' going off?   have a good idea of what's going on hither?

[Addicted2AnalogTek]

raising the input frequency made the issues in the vertical section much, much more visible and easy to pick out.  I'm sure there's a few, but now I am 100% certain that the oscillations I saw on the Vcc pin of U508 are the cause of a large number of problems on the vertical board.   The oscillations are the exact same frequency as the oscillations measured directly before the amplifier stage (?) before the delay line.  I was able to match up every detail using points on the screen of my 465, and if I had a spectrum analyzer I'm certain that they would match up on that as well.

Since this noise has flooded the 5V rail on this board, I am going to replace U509 & U409 and pray that U408 hasn't been damaged. If it has, I'll be searching for a quick parts scope!    Oh, and I'll certainly be flush-mount socketing any IC's that I have to replace.  I'm a bit peeved that they took out almost all of the sockets on the 468 model to 'improve reliability'. I can understand the reasoning, but it's a PITA for the troubleshooter!

[Addicted2AnalogTek]

It helps a lot that I am actually starting to understand how transistors/diodes function and how their characteristics are supposed to look.   I remember when I would look at a datasheet and be completely clueless as to what any of it means.... It's been a big weekend.  hah

[tautech]

Wondering out loud.......snubbers used to be added around zeners to reduce noise, just a thought....

[Addicted2AnalogTek]

not sure I understand what you're getting at.... I have no idea what a snubber would be.  I'll look it up...   but there are no measurable oscillations coming from any of the zeners on the vert board.  That's not saying anything about the vert amp, which I already know is at least half of the problem with the trace I took pics of.   I'm about to poke around there right now, actually.

Also, there is cross-talk between channels. Presumably coming from the 5V rail and the logic section, but I'm not certain.  It's also measurable on the 15V and -8V rails as well, but it's worst on the 5V near the logic stuff.

   I also know that the thermistor for CH1 (100kohm 4mW) is bad.  figured this out by making use of the CH2 signal bleeding into the rails.    I was able to measure the little bit of 1mhz signal coming thru it and noticed that the trace on the 465 floats up and down (ac coupled).  I already replaced the one on CH2 (should have ordered 2 but they were $5 each for the good ones I picked..)    I tried to get the closest dissipation that I could, (5.4mW) and that's what the price ended up being.   and I could only find it at Mouser.... horray for more shipping charges..

[Addicted2AnalogTek]

ohhhhh  boy... this pup has some ills in the logic, that's for sure.   nothing on the mode switchboard so far is a clean looking signal.  there's logic signal's piled on top of each other and offset on several pins of several chips.   also trigger view enable doesn't work at all - and it's not in the discrete components.    this should be quite an interesting continuation of the troubleshooting considering I haven't even begun to teach myself how logic circuits work.   all I know thus far is what a clean signal looks like (and obviously it's inverse), and what high voltages and low voltages are.  I don't know what frequency to look for, how NAND/AND gates work, etc etc etc.     

anyone care to point me towards some tutorials that are fairly simplistic, yet thorough enough to understand?  I'd like to try and weed out the videos/pages that are useless for a beginner in logic like myself.  Just for reference, I have an easy time understanding this guys vids: 
https://www.youtube.com/channel/UCdhGKS28LbzxebigAlA2rDg

but he doesn't have any vids on logic circuitry.

[Addicted2AnalogTek]

My apologies that I forgot to mention this, but this whole time I have been running the scope without the storage mode isolated - but for a very good reason.  The scope will not operate unless everything is connected and not in test mode!!    The pictures below are what happens when I put it into test mode:

I am not able to do anything or get a semi proper trace displayed on the scope when I have it in test mode.  This is likely due to the faulty logic IC's on the vertical board that I have been testing.    The manual says that the scope is supposed to operate in CHOP only when it is put into TEST MODE, since what I am basically doing when I put it into test mode is breaking the connection between the vertical mode switchboard and the vertical switching logic.  It powers on and everything when in test mode, but does not respond to any CH1 controls or signal, and CH2 will only show two fuzzy horizontal lines depicting the amplitude of the input signal to channel 2. The only thing that changing the timebase does is dim the trace and shift it to the right as I turn it to shorter time settings.   I am not sure what is going on, so I am going to leave it in test mode for right now and poke around to see if I can figure out what is happening.  After I figure out some more about what is happening I will likely make a video to show it.

[Addicted2AnalogTek]

Alrighty......

In the test mode, U408 and U509 are not getting the correct voltage sent to them in order to activate CH1, CH2, and CHOP.   The only settings getting any voltage at all are CH1, CH2, and XY mode, but only 0.73V.   depressing the CH1, CH2, and CHOP switches only increases the voltage to these by 2mV.    There is 50Khz riding those and several other inputs/outputs of the other 74 series ICs, and also some 5.6Khz (very low amplitude) as well. 

The 5V rail still has the same noise (5.6Khz) and still seems to be originating from U508.  I'm replacing them regardless, but I would still like to know what exactly has gone wrong.   I really feel like replacing all of the 74 series ICs and any others I can get quality for cheap, but even if I do so I still want to find out what the actual problems are.  It's not a learning experience if I just keep pulling and replacing everything that I can!! 

I have arranged some jumper links and 90 degree header connectors so that I can get better access to the mode switchboard.  I just wish I had extensions for the coax interconnection cables...  I will be on the lookout for a deal on those females and cables for future repairs - and also a longer ribbon connector cable.

any tips on what I should be looking for based on the info I've provided? or is there anything specific you want me to check/measure so you can get a better idea of what the deal is?