Tek 5103 CRT H&V drive partially collapsing.

[basinstreetdesign]

My Tek 5103 scope shows H&V deflection only about 3/4 that of normal.  So I went looking for drive signals between the plugins and the CRT.  At first I thought it was just horizontal deflection that suffered but when I finally put in a signal to a plug-in I saw that it was both V & H.

BTW: Trying to narrow this down was hampered by the fact that I did not have a service manual for the scope.  And I haven't since several years of looking at the usual BAMA site and others like it.  That is, until I finally found a place that had scanned in a complete manual for this scope including all S/N versions of its life and you would not believe where I found it.  It was in the student-available archive of the equipment on hand at a small University in Switzerland, the University of Friburg!

https://www3.unifr.ch/med/de/assets/public/professoren/rainerg/Documents/Manuals/tektronix_5110_oscilloscope.pdf

Now that I have that, I scoped the signal chain all the way to the final drive amplifiers (runs on 200V) that feed the crt plate pins.  And all signals are as according to the manual.  I can attach waveforms and manual excerpts if anybody is interested.  But the displayed waveforms still are attenuated.  Focus adjustment is fine. Astigmatism is good.  Hor position will move the display but crush it at about 1.5 cm in from each edge.  The display has always been fairly dim to the point where many times I have to turn the room lights off to see the waveform well.

So now I'm thinking that there's something wrong with the HV.  Its about -4500V and I cant measure it.  I used to have a HV probe for my DMM but I haven't found it since we moved almost 7 years ago. 

Anyway, I am a little fuzzy on how the HV and other signals on the crt will affect the display so I am wondering if any other strategies occur to you good folks.

[fzabkar]

FWIW, if a CRT TV's raster "blooms", then that's often caused by low EHT (eg a bad tripler). OTOH, if the raster shrinks, then that could be due to excessive EHT. How does the scope regulate the EHT, if at all?

BTW, the reason that the raster blooms when the EHT is low is that the electrons take longer to travel from gun to screen, allowing the yoke more time to deflect them.

[james_s]

Small image could be either excessive EHT, or a problem with the deflection amplifiers. Did you check the power supply voltages? That's normally one of the first things you want to look at, if any of the voltage rails are not right the scope will have all sorts of issues.

[basinstreetdesign]

I checked all power supply voltages and they are all within spec +/-2%.  I know that a low HV will make the image bloom so that's not it.  Also I suspect HV is too high but cannot verify it.  I checked a waveform on the HV feedback circuit and it looks the same as sample in the manual.  One thing about the manual is that while ac Vpp are stated the dc offset or location of 0V on the screen is not so I don't know if my waveform is on a dc bias too high or low or what.

[fzabkar]

I would check the megohm resistors, R272B/C/D/E, R276 and R271.

[TerraHertz]

That is, until I finally found a place that had scanned in a complete manual for this scope including all S/N versions of its life and you would not believe where I found it.  It was in the student-available archive of the equipment on hand at a small University in Switzerland, the University of Friburg!

https://www3.unifr.ch/med/de/assets/public/professoren/rainerg/Documents/Manuals/tektronix_5110_oscilloscope.pdf

Thanks for that! But how did you find it? I can access that pdf, but link-munging fails to access anything else. I would like to see what other manuals they have.


Attached is my Tek 5000 scopes notes file. Contains a few more manuals sources.

PS I agree, shrunken trace likely means the CRT EHT is too high.

[David Hess]

I think you can use the 5110 manual.  Isn't it the same oscilloscope?

A high (more negative) cathode acceleration voltage will lower the deflections because the electrons spend less time between the deflection plates.  Conversely, a low (less positive) post deflection acceleration voltage will also lower the deflection because of how scan expansion works.

It cannot be the post deflection acceleration because the 5103 CRT does not have it so there is no high voltage multiplier to go bad.  2 MHz bandwidth oscilloscope do not need post deflection acceleration.

That leaves a high (more negative) cathode acceleration voltage.  This could be caused by a bad high voltage divider, bad control loop, or bad reference voltage.  The easiest thing to check is the voltage at the base of Q264 which should be -0.6 volts and the -30 volt supply which is used as a reference.

If you have the older 5103, then my guess is that small signal Darlington transistor Q278 is bad.  Small signal Darlingtons always seem to have poor reliability.

[basinstreetdesign]

That is, until I finally found a place that had scanned in a complete manual for this scope including all S/N versions of its life and you would not believe where I found it.  It was in the student-available archive of the equipment on hand at a small University in Switzerland, the University of Friburg!

https://www3.unifr.ch/med/de/assets/public/professoren/rainerg/Documents/Manuals/tektronix_5110_oscilloscope.pdf

Thanks for that! But how did you find it? I can access that pdf, but link-munging fails to access anything else. I would like to see what other manuals they have.


Attached is my Tek 5000 scopes notes file. Contains a few more manuals sources.

PS I agree, shrunken trace likely means the CRT EHT is too high.

I found it by googling tek 5103 display section service manual pdf and it took me to here:
https://www3.unifr.ch/med/de/research/groups/rainer
then clicking on DOCUMENTS and then MANUALS produces a short list of gear in that department.  The list has the TEK 5110 at the bottom and voila!

Update on the scope in question...
I agree that high HV seems to make sense.  Once after I cleaned some pots and was readjusting them to see what they would do (HV adjust, Brightness Range, H & V final gains) the display abruptly corrected itself.  At that moment I caught a flash of light from some protection neon lamps.  It stayed good for about a minute then went bad again, with another flash from the neons.  I realize that these neons play no role unless the cathode-grid voltage exceeds about 180V so didn't pay it much mind until I accidentally broke one off.  I looked them up in the manual and in Digi-Key and found they can still be had.  I also realized that they fire anywhere from 60-90V each and that even with only two of the three in circuit the CRT would still be protected since the usual K-G voltage is about 50V.  So, having no replacement, I just took out the broken one and shorted its position in the pcb.  The scope still acts as it did.

Thanks to all and I'll keep on keepin' on.

[basinstreetdesign]

Fixed.

Yes, the HV was out of regulation and too high.  Once when the HV was correct (lasted only about a minute) I could crank the HV adjust and make the trace zoom in or out so confirmed what effect having too high HV would have.  Probing the base of one of the feedback-path transistors showed it to be -18V when it should have been -.6V since the emitter was grounded.  It was a garden variety PNP, Q264 and had gone open.  Replaced it with a 2n3906 fixed it.  I readjusted the HV to get correctly scaled V for the test signal and then adjusting the final Hor gain gave me correct H scale.

Thanks to all who weighed in. 

Also fixed the attenuator in one of the plug-ins, a 5A12N. 
Cheers.

[james_s]

Nice job, another classic Tek scope alive again.