Tektronix 2465B oscilloscope teardown

[tazdog]

If you can't find a free manual, http://artekmanuals.com has them. Good scan's and they are searchable.

[tazdog]

OK, google is your friend, right from the source - https://www.tek.com/manual/2445-service-manual - if you don't mind giving up your email address 

[MarkL]

OK, google is your friend, right from the source - https://www.tek.com/manual/2445-service-manual - if you don't mind giving up your email address 

Thanks, but that's for the 2445.  We need the schematics pages for the 2445B, emphasis on the B.

[BravoV]

All, just a tip for anyone working on the A5 board due to capacitor leakage, don't forget to check R2010 20K trim pot. Mine was reading close to 30K so that's a big difference!!! They are easy to get from DigiKey.

What was the problem/symptom before you fix that ?

[z01z]

Thanks, but that's for the 2445.  We need the schematics pages for the 2445B, emphasis on the B.

You could use the manual for the 2465B, the 45s are said to be mostly the same as the 65s. Search for the filename: 070-6873-00_2465bSvc_Sep89.pdf

[grumpydoc]

Thanks, but that's for the 2445.  We need the schematics pages for the 2445B, emphasis on the B.

You could use the manual for the 2465B, the 45s are said to be mostly the same as the 65s. Search for the filename: 070-6873-00_2465bSvc_Sep89.pdf

There might be other differences but, from memory, the firmware limits the fastest timebase slightly in the '45 and there is some filtering on the input to the final Y amp - after the delay line, to limit bandwidth slightly. At least some of the implementation is with PCB traces to form inductors.

Some unscrupulous vendors have, in the past, been known to bypass this and sell the result on eBay claiming it to be a 2465.

[tazdog]

The "2445" manual should be good enough for the issue he is experiencing.

[casinada]

http://www.ko4bb.com/getsimple/index.php?id=manuals&dir=Tektronix
 

[MarkL]

http://www.ko4bb.com/getsimple/index.php?id=manuals&dir=Tektronix
 

Been there.  In the 2445B directory, right?  Like I said previously, the service manual is there, and in many other places around the net, but they are all missing the schematics.

The suggestion to get a complete service manual from Artek is probably the way to go.  They have the 2445B.

Strictly speaking, I agree it's not necessary and you can probably get away using the schematics from one of the close relatives of this scope.  But if it were me and I never fixed one of these before, I'd want something accurate for now and to keep as a future reference.  It's certainly up to user mahwe.  I'm just trying to provide some troubleshooting hints.

[tazdog]

I agree. start with the free one and if that doesn't have enough info then purchase a manual.

[Orange]

If you need a free 2465B manual with schematics, have a look at the Bitsavers site. It is there. It has reasonable readable diagrams.
Link of the mirror site:
http://bitsavers.informatik.uni-stuttgart.de/test_equipment/tektronix/scope/

Follow MarkL his suggestion

[Orange]

I just uploaded a better scanned 2465B service manual to the http://www.ko4bb.com/ site

It is a 1993 manual, with schematics on one page, good quality, 35 Megs

[BravoV]

I just uploaded a better scanned 2465B service manual to the http://www.ko4bb.com/ site

It is a 1993 manual, with schematics on one page, good quality, 35 Megs

Thanks. 

Can't download it yet as its still in unchecked upload area.

[Orange]

Try to set-up an ftp to ftp.ko4bb.com. User is manualuploads, pw is also manualuploads.

There are two manuals 

[BravoV]

Thanks, got it ! 

Found an interesting date in there, highlighted in red.

[Orange]

Thanks, got it ! 

Found an interesting date in there, highlighted in red.

Even better, did not even noticed this

[BravoV]

Thanks, got it ! 

Found an interesting date in there, highlighted in red.

Even better, did not even noticed this

Yeah, cause all this times I've been wondering about my fan's date code, and was keep suspecting whether its originally installed from factory, I guess this confirmed it.

Again, thanks a lot for your contribution. 


Edit : Also already uploaded to TiN's xdevs.com documents server, hope you don't mind.

[Miti]

You guys made me curious about the numbers of hours on my Tek 2445B. Check this out.
I replaced the electrolytic caps only in the PS and HV, the FRAM is on its way, I saved the content of the NVRAM and installed a socket. Do you think I should change the caps on the main board? It works fine.

[james_s]

IMHO if it works fine leave it alone, there's always some risk inherent in taking the thing apart.

[Miti]

There is a bit of a vertical jitter in the readout on some horizontal ranges and the tube goes out of focus if I go all the way up with the intensity but I guess this is mostly related to the aging of the tube?

[Miti]

I've got my FRAM FM6W08 from ebay and it doesn't work. I can program it with Xeltek SP5000, it passes verification, but once inserted in the scope I get TEST 04 FAIL 11. It's either a faulty rejected chip, I wouldn't be surprised considering it was only $4.39, or maybe it doesn't work in 2445B. I would expect the addressing mode to be identical between 2445, 2465, 2467. Any experience?

[MarkL]

There is a bit of a vertical jitter in the readout on some horizontal ranges and the tube goes out of focus if I go all the way up with the intensity but I guess this is mostly related to the aging of the tube?

A little bit of jitter is normal.  There are adjustments to tweak the vertical and horizontal readout jitter, R805 and R618.  It's in the service manual in the "DC Balance, And X-Y Phase Differential" section.

The de-focusing with increased brightness is called "blooming" and you're right, it gets worse as the tube ages.  You may be able to partially compensate for it by going through the CRT adjustment procedures in the service manual, particularly the High Drive Focus.  Some of these adjustments may change the position of how the beam lands on the graticule and the display (and just the display) may become uncalibrated.

Hint: Mark the current position of any pots with a good Sharpie before you move them in case you need to put them back.   The CRT adjustments have a lot of interaction between them and can drive you crazy trying to get them just right.  Unless the jitter and focus really bothers you, I would probably leave everything alone.

[MarkL]

I've got my FRAM FM6W08 from ebay and it doesn't work. I can program it with Xeltek SP5000, it passes verification, but once inserted in the scope I get TEST 04 FAIL 11. It's either a faulty rejected chip, I wouldn't be surprised considering it was only $4.39, or maybe it doesn't work in 2445B. I would expect the addressing mode to be identical between 2445, 2465, 2467. Any experience?

No experience, but one thing you could do is use Exerciser 02 to scroll through the calibration data you programmed into the FRAM.  As discovered in another thread, RAM locations 0x1e00 - 0x1fff should be the 16-bit calibration constants visible through the exerciser, big endian.

Maybe the values (or lack of) would provide some clues.

Is that part number right?  Did you mean FM16W08?

[Miti]

Unless the jitter and focus really bothers you, I would probably leave everything alone.

That's exactly what I'll do, it is very usable as is. 

No experience, but one thing you could do is use Exerciser 02 to scroll through the calibration data you programmed into the FRAM.  As discovered in another thread, RAM locations 0x1e00 - 0x1fff should be the 16-bit calibration constants visible through the exerciser, big endian.

Maybe the values (or lack of) would provide some clues.

Is that part number right?  Did you mean FM16W08?

Yes, I meant FM16W08. The problem is that the scope keeps rebooting so I have a hard time running the exerciser. I managed to do it once and all the locations were 00. I removed it and read it with the programmer and other than the very last address and one somewhere in the middle of the calibration space they were all there.
What I suspect is that it is a part that failed in production. I works in the programmer, probably it is programmed at low speed, but once it is addressed at full speed, it doesn't keep up. Don't know.

[MarkL]

What I suspect is that it is a part that failed in production. I works in the programmer, probably it is programmed at low speed, but once it is addressed at full speed, it doesn't keep up. Don't know.

Seems like a good theory.  The cycle time of the 2445B and others in the series is a leisurely 800ns, so the FRAM must be in pretty bad shape to not work at that speed.  Either that or some setup/hold times are on the edge since it works some of the time.  It might be better to get a fresh, known good FRAM from Digikey or Mouser before investing more time in it.

You might already be familiar with this paper that talks about using FRAM in a 2465B successfully:

  http://worldphaco.com/uploads/TEKTRONIX_2465b_OSCILLOSCOPE_CALIBRATION___REPOWERING_THE_DS1225.pdf

The 2445B should be the same as far as the SRAM concerned.