Tek 'scope CRTs - how do they work?

[tggzzz]

I've found a Tek 465 scope which seems to be working well except for one feature: the intensity control is ineffective and the trace is so bright you could use the CRT as a torch.

My understanding, such as it is, is that the fault is likely to be in either HV DC restorer circuit or inside the CRT itself. My feeling, based on somewhat indirect measurements of 2.5kV circuits, is that the DC restorer is working OK and that the fault is with the CRT's grid - but my lack of experience with CRTs makes me hesitate to jump to that conclusion.

Am I right in thinking:

• the beam's intensity is controlled by the cathode-grid voltage?
• the more negative the grid relative to the cathode, the less intense the beam? I.e. the negative grid voltage repels the electrons and prevents them being accelerated by the anode?
• there is a "cutoff" voltage which will prevent any electrons hitting the phosphor?

If so, what is a likely value for that cutoff voltage?

Not having a HV differential probe, I can't measure the grid-cathode voltage (V_gk) directly, but have to resort to indirect means. Measured values:

• cathode voltage -2450V, which is correct, and the focus voltage is 20% too low
• with the CRT disconnected, the voltages on the focus plate is as expected and the cathode voltage is slightly higher. My guess is that difference is due to a high electron beam
• the low-voltage blanking signal looks sensible: 0V_pp with intensity control on minimum intensity, and up to 60V_pp as the intensity (control) increases
• V_gk = -123V_mean with minimum intensity. Is that more than the expected cutoff voltage?
• V_gk = -68V_mean with maximum intensity

Given the duty cycle of the low voltage blanking waveform, -68V_mean is what I would expect if the DC restorer capacitors and resistors are working correctly.

So, is it likely there is a fault inside the CRT?

What other tests could I make?

Might there be any way of rectifying a fault in the CRT?

Thanks for your thoughts and advice.

[PA0PBZ]

http://www.ko4bb.com/manuals/download.php?file=Tektronix/Tektronix_-_465_Oscilloscope/Tektronix-465_Service_Manual/Tek_465_Service_Section-4.pdf

[tggzzz]

http://www.ko4bb.com/manuals/download.php?file=Tektronix/Tektronix_-_465_Oscilloscope/Tektronix-465_Service_Manual/Tek_465_Service_Section-4.pdf

Thank you, but I read that and it doesn't contain the answers to my questions.

[Addicted2AnalogTek]

The following book has the answers you seek.....

http://sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf

[tggzzz]

The following book has the answers you seek.....
http://sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf

Thanks, yes.

If I understand it correctly, "most" Tek scopes will cutoff with V_kg more negative than -65V. I see -123V and it is far from cutoff! That's one of the things that leads me to believe there is a tube problem - but I know that I know very little in this area and need a sanity check!

[tggzzz]

Also, from what I remember, your specific problem is an issue with the HV power supply.  More specifically a weak or shorted diode in the control grid side.   Don't take my word for it, though. I have never had the issue with any of my scopes thus far, and am certainly still quite inexperienced.

Inexperienced like me!

The focus voltage is low (i.e. more positive), but the focus control still has an effect. If it did fully focus the beam, the phosphor wouldn't last long! The focus is a separate part of the circuit, and any problems might simply be a consequence of the beam intensity. I'll look at that after sorting out the intensity.

More importantly, the cathode and grid voltages appear correct - as best as I can judge. That's against my understanding of the "weak diode" phenomena. I've also tested the diodes in the DC restorer with a DMM, and they are fine as far as that limited test goes. I've replaced them with BAV21s, with the results given previously. (I wish I'd measured the voltages before replacement, but I was hoping for a quick fix).

I haven't replaced the HV ceramic capacitors because they are more difficult to obtain and there doesn't seem to be a problem with the voltages - but I'd like that sanity checked by some one with more experience.

Hence my questions about voltages!

[Andy Watson]

Have/can you checked the voltages on the pins of the tube?

[tggzzz]

Have/can you checked the voltages on the pins of the tube?

As best I can using a 100Mohm 100:1 potential divider (3*33M+100k)plus a DMM!

I haven't measured the anode, but many electrons are hitting the screen so I doubt that's a problem.

The cathode is adjusted to: V_k = -2450V
The grid is a bit lower: V_gk = -42V_mean to -147V_mean depending on the position of the CRT bias trimpot and the front-panel intensity control. (i.e. V_g is around -2550V)
My understanding is that the grid is physically between the cathode and everything else, and is responsible for controlling the beam current. Or not in this case!

The focus is -1060V to -1670V depending on the focus front-panel control. The manual indicates -1700V. If the CRT is disconnected, the focus is  -1600V to -1900V. With neither understanding nor evidence I presuming that is a secondary problem that might be caused by the excessive beam current.

The other voltages are all reasonable, but I'm guessing they are also unimportant at this stage.

So, what would you infer from those voltages?

[rf+tech]

As a vote of confidence in your understanding, with the cathode at -2450 Volts and the grid *more* negative (-2492 to -2597 Volts), the beam should completely cut off within the range of -2515 Volts to -2597 Volts.

It does appear that the control grid is open inside the tube, unfortunately. 

One last thing to try is carefully removing the CRT socket and visually inspecting pins in both. Look closely inside the neck of the CRT for visual clues.

RF+ Tech

[Addicted2AnalogTek]

It would help if I didn't attempt to do anything when I'm half asleep (or more...) including reading and replying on posts.  I don't remember actually reading half of your initial post!! 
Being rather inexperienced, following repair posts helps me learn, though.  It sucks that you may have a bad tube!!
I have a 465 open on my bench if you need any measurements to compare against.

[tggzzz]

Being rather inexperienced, following repair posts helps me learn, though.  It sucks that you may have a bad tube!!

Yes and yes The TekScopes Yahoo! group contained useful information.

I have a 465 open on my bench if you need any measurements to compare against.

I think the only relevant measurement is to stick a DMM across the CRT's grid and cathode, to see what the range of V_kg is. Well, not directly, in case there is a gross fault involving high voltages. I used a homebrew 1000:1 100Mohm divider chain (3 * 33Mohm + 100kohm).

With the intensity control set to minimum, there was no LT blanking waveform at TP1486. Hence V_kg should be largest, in my case ~-100V to -147V depending on the position of the CRT bias trimpot. If my understanding is correct, that should be more than adequate to blank the beam.

To find the smallest V_kg, adjust the intensity control for the maximum peak-peak voltage at TP1486 (65V in my case), and select a timebase to maximise the duty cycle to somewhere like 90% high 10% low. Then V_kg should rise by ~60V to somewhere between -40V and -87V. If my understanding is correct, that the beam should be visible for some of that range.

Looks like I need to decide whether to break it up for spares or find a new tube.

Apart from that, it is a damn sight easier to work on and repair my HP1740A: no wires soldered to boards, no pairs of components (e.g. resistor+diode) with joints in mid-air for no good reason other than they discovered a need for one of them after the board had been laid out. Yes, it is an early 465, SN106389

[Addicted2AnalogTek]

I have a 465 that was damaged in shipping - SN B308858 - which has flying lead connections as well, so I don't think they ever bothered to change the boards to accommodate. That one, being an Option 5, is even worse with the flying leads than my standard 465 B254554. 

The reason I mentioned "damaged in shipping" is, even though I got it electrically working (for the most part) that I am on the fence about parting it out. In the event I do, I would sell it's known-good CRT for $50 plus shipping.  However, if you wait, you could just get a whole parts scope for near or less than twice that.

[tggzzz]

I might conceivably be interested in that, but I'd like to know the post and packing costs first!

A suggestion on the TekScopes group is that I disconnect the CRT, apply a normal heater current, and set V_gk to a few 10s of volts with a current limiting resistor. If they are both connected they should act as a diode and conduct a few microamps. I think I'll try that tomorrow.

[Addicted2AnalogTek]

It seems I must join the Yahoo Tekscope forum asap... there must be at least a few old Tek techs over there!

That reminds me.. I need to get myself a good analog microammeter. I bought bunches of scopes, cal gear, and signal gens and didn't bother buying a single quality bench meter.  I'm going to have to try and score one or a few soon.

Thanks for taking the time to put much of your repair and troubleshooting info into the thread, though. This type of stuff is an especially helpful learning tool in addition to the books I've been reading!

As for the CRT, there would be no packing costs as I'm fully stocked on boxes and packing materials from last years purchases.  In case you want to find out, it would be approx 3lbs or a little under, and in a 20" x 9" x 9" box. (50cm x 23 x 23) and would ship out of WV USA 26501.

[tggzzz]

It seems I must join the Yahoo Tekscope forum asap... there must be at least a few old Tek techs over there!

It is an active group and the information can be very precise and detailed.

That reminds me.. I need to get myself a good analog microammeter. I bought bunches of scopes, cal gear, and signal gens and didn't bother buying a single quality bench meter.  I'm going to have to try and score one or a few soon.

If you're not careful you'll turn into a VoltNut. Dont forget that someone with two watches (voltmeters) doesn't know what the time (voltage) is.

Thanks for taking the time to put much of your repair and troubleshooting info into the thread, though. This type of stuff is an especially helpful learning tool in addition to the books I've been reading!

Ah, you lucky person. You have yet to be delighted by appreciating the much larger amount of effort visible in the TekScope group!

My small contribution has already been rewarded by the responses of others.

[tggzzz]

In my case the 465 tube was terminally damaged: the grid was internally disconnected. Rather than throw it out, I decided to chop it up and put the pretty bits in my display cabinet.

The results, showing the disconnected grid and the hidden metal mesh lens, can be seen at https://entertaininghacks.wordpress.com/2016/03/09/rescuing-a-broken-tektronix-465-crt/

Those interested in the Burr Brown equivalent of the Pirelli calendar can look at pretty scans of thick-film hybrids at https://entertaininghacks.wordpress.com/2015/07/07/images-of-late-70s-burr-brown-thick-film-hybrid-ics/ (Yes, other people have described them as "geek porn" )