Tektronix 2467B - Display Issues

[mtchastain]

Backstory: I had seen this on the shelf over the last 2 years or so at a local recycle shop chain here where I live. It's not the normal kinda thing they have and the fact it was listed at almost 300 USD in their junk section almost guaranteed it wasn't going anywhere. Not knowing much about it other than its display was all distorted and no amount of fiddling made much of a difference, I passed it by without much thought.

Well, last week while visiting the shop, I noticed it was missing from it's usual spot and so decided to take a look. After a bit of a search, I eventually found it relegated to a lower shelf behind some old projectors and a sad looking fan. Not expecting much, I rechecked the tag and got a shock.

I guess they must have finally given up on it after all this time, since it was now listed at less than 10 USD (1000 yen)
Needless to say, you know what happened, and so here I am.

Here are my observations I have so far on the scope, if anyone has any suggestions, I am all ears. I have the service manual and am currently scouring this and other sites for others who have shared their experiences. As my testing progresses, I plan to add my findings to this thread.

Currently known issues:

• Display is off past the right side of the screen, by default something can be seen pressing Find Trace.
  - After performing part of the display diagnostic, part of the display will appear on the right but slowly move off to the right of the screen.
• When pressing any buttons, all LEDs on the panel briefly flash.
  - Seems this is a normal effect of the design
• When rotating the readout intensity over the center point, all LEDs briefly flash.
  - Seems this is a normal effect of the design
• Intensity controls past ¼ CW rotation are too bright

Confirmed OK:

• All power supply voltages are good (within the range listed under the troubleshooting procedure)

Observations and Tests:

• Power and initialization appears normal (but not 100% certain)
• A visual inspection of all boards finds no leaking caps or battery, or obviously damaged parts or any corrosion.
• Confirmed U800 (Horizontal Amp) is bad. Recovers for a short time when cooled. Part is known for failure.

Plans:

• Given the age, I expect a complete re-capping may be in order.
• Possibly add a heatsink to U800 as recommended by BravoV in his 2465B teardown
• Find a suitable lithium battery replacement
• Replace U800 with drop-in replacement made by Ramkumar (when available)

[jxjbsd]

All 2465B and 2467B users need to look at this page:
https://www.eevblog.com/forum/testgear/tektronix-2465b-oscilloscope-teardown/


Looking at the picture, the oscilloscope does not have much problems. Try rotating READOUT INTENSITY to see if there are characters displayed. If the operation problem can be ruled out, I think the capacitor on the A5 control board is leaking.

[jxjbsd]

After watching the video, it is normal when the computer is turned on. After a few seconds, the characters disappear and the track shifts. I judge that it is the problem of the analog control part, the core of this part is a DAC.
Check the voltage of these three points

[jxjbsd]

Check R2013, it should be broken

[jxjbsd]

Hi jxjbsd,
I pulled the resistor and checked, it was dead on 10K. 

What is the result after the replacement?

[jxjbsd]

 
V REF, V REF-The voltage at these two locations is wrong. Looking at the circuit diagram, it is clear that their voltage to GND should not be so low.  Is the 10V REF voltage normal?

[Kwakerman]

I would suspect that U800 has lost its left deflection plate output and so the right plate voltage s pulling the beam far right.  I have a 2445A with the same issue and just waiting for a new chip.  Try disconnecting both the left and right deflection plate connectors on the pcb, the beam should now be a vertical line with top and bottom 'blobs' caused by the readout text and a spot that can be moved up and down the screen using the channel 1 position control.

[bdunham7]

Get some spray freeze and freeze U800 to see if that keeps it going longer on startup.

[Kwakerman]

This is the normal issue with U800, heat sinking it won't help, it needs replacing.  there is a potential option to 're-bake' it, see https://www.sphere.bc.ca/test/tek-parts/tek-info.html and scroll down.  If buying a replacement try to get a Tek one and not Maxim

[bdunham7]

Hi bdunham7,
Thank you, that suggestion had the best effect so far! 
I blasted U800 each time the trace and readout started to distort and shift to the right and it recovered each time for 4 or 5 seconds. Odd thing is that chip is not that hot, it was around 25C before the issue start to happen, and was in full distortion at about 30C or so....

Sorry to hear that, your U800 is toast.  I  think this was actually a more common problem on the 2467 series because they were operated at very high sweep speeds, which stresses U800 more. 

Check this guy out:

[med6753]

Sorry to hear that, your U800 is toast.  I  think this was actually a more common problem on the 2467 series because they were operated at very high sweep speeds, which stresses U800 more. 

No, it's because the vast majority of 2465B/2467B have Maxim die for U800 which is much more failure prone than the original Tek die.

[bdunham7]

Sorry to hear that, your U800 is toast.  I  think this was actually a more common problem on the 2467 series because they were operated at very high sweep speeds, which stresses U800 more. 

No, it's because the vast majority of 2465B/2467B have Maxim die for U800 which is much more failure prone than the original Tek die.

Was the Maxim die used in all later models, including the 2445B, or just the faster models?  How can I tell the difference?

[Kwakerman]

The Maxim one has 'Maxim' printed on it and curiously the Tek one has 'Tek' printed on it!   

[bdunham7]

The Maxim one has 'Maxim' printed on it and curiously the Tek one has 'Tek' printed on it!   

OK, I guess I'm lucky.  I've never seen a Maxim one then!  I guess I'm good to go.

[Dr_Ram]

Hi Brian, Thank you for your Youtube comment on my video and for posting it in EEVBlog. To your question about whether I produce these or want to -- I am indeed planning to launch this on ePray.com in a couple of months. The PCB layout and design are in the third iteration and this video was essentially the result of my attempts to fine-tune the performance to an acceptable level before finalizing the design and layout.
Best - Ram

[BravoV]

Sorry to hear that, your U800 is toast.  I  think this was actually a more common problem on the 2467 series because they were operated at very high sweep speeds, which stresses U800 more. 

No, it's because the vast majority of 2465B/2467B have Maxim die for U800 which is much more failure prone than the original Tek die.

Was the Maxim die used in all later models, including the 2445B, or just the faster models?  How can I tell the difference?

A pic speaks thousands words. 

[BravoV]

Regarding alternative of U800, someone did it years ago -> Tektronix 2465/2467 series U800 IC replacement design

[Kwakerman]

I'd certainly be interested in a couple as I have three 2445A/B and one needs a U800.  Have you tried this design in a 2465B as ultimately that is the next scope I'm on the lookout for?

[Dr_Ram]

Thanks for the interest. I have tried this only on a 2465A so far, but since the U800 is the same chip on both I suspect it should work on the B too. If I get hold of a B however I will certainly reconfirm here,
Best - Ram

I'd certainly be interested in a couple as I have three 2445A/B and one needs a U800.  Have you tried this design in a 2465B as ultimately that is the next scope I'm on the lookout for?

[Dr_Ram]

Not quite actually. I did try implementing this (Thomas Lafay's) design to see how it worked, and I faced some issues. As posters on Tekscopes@groups.io have noted, it does not perform very well at the higher speeds. It also has some functionality missing such as the X-Y mode. However it was a splendid effort and many thanks to Mr. Thomas Lafay for doing it and inspiring me in the process to build the proto linked to to above.
Best - Ram

Regarding alternative of U800, someone did it years ago -> Tektronix 2465/2467 series U800 IC replacement design

[Galen]

Thanks for the interest. I have tried this only on a 2465A so far, but since the U800 is the same chip on both I suspect it should work on the B too. If I get hold of a B however I will certainly reconfirm here,
Best - Ram

Hi Ram, it will be very helpful for many Tek24xxA/B owners, if your design can replace the U800.
Another very fragile part is U400 for Tek24xx A/B. So when you put your U800 replacement in production, do you have plan to work on the U400 replacement design?
Anyway, thank you, good job.

[Dr_Ram]

Your comment on the U400 is very useful, as I had already designed a schematic and layout for it some time ago but shifted attention to the U800 which seemed somewhat more pressing. Motivates me to return to the U400 when I'm done with the final version of the U800.
Best - Ram

[/quote]
Hi Ram, it will be very helpful for many Tek24xxA/B owners, if your design can replace the U800.
Another very fragile part is U400 for Tek24xx A/B. So when you put your U800 replacement in production, do you have plan to work on the U400 replacement design?
Anyway, thank you, good job.
[/quote]

[Galen]

Ram, if you can produce the U400 replacement, I think there will be many demanding for it. my self have two scopes with U400 broken.