Tektronix 2465B oscilloscope teardown

[Howardlong]

The time out is a normal feature I think on the 2467B, just not sure if 30 secs is the default, and if that can be changed. No, it makes no difference on the intensity setting, still 30 secs  countdown, no matter what. I'm sure it saves burns, but quite annoying!

There are two timeouts, one to dimming and the other to "shutdown" (as far as I cans see this is just a sleep mode, the scope comes back immediately when you hit a button).

The first timeout to dimming on my example seems a little random, but ranges from about 75s to well over three minutes, I don't know the algorithm it uses, or maybe it's just age!

The second timer, to shutdown, is about 3 minutes.

I usually just hit Beam Find as the benign way to get the display back again, and while it's a minor irritation you do get used to it. I have the timer turned on permanently, it's the EXER 09 setting.

Here's an OCR'd copy of the manual that has the relevant sections https://onedrive.live.com/redir?resid=2EDA80327053C74D!22445&authkey=!ABOYAYHimQYnnAs&ithint=file%2cpdf

[SoundTech-LG]

Thank You Howard!     

I have been looking for this , but could not find! It's exactly what I need!

Martin

[SoundTech-LG]

Howard,

Do you know much about the differential aging feature? I was certain the CRT had light burn spots, (could see them in a flooded screen) but after reading the manual section on differential aging it sounds like I am just getting less beam amplification in those areas. Does that sound right?

Martin

[Howardlong]

Sorry, I am no expert, if Alan w2aew comes by I am sure he will be able to answer far better than I can.

What does happen is that the readout text ever so slightly dances around the CRT, a pixel or two horizontally and vertically, at certain times, and it is by design. Unless I'd read about it in the manual, I probably wouldn't have noticed it to think anything much about it.

[w2aew]

Sorry, I am no expert, if Alan w2aew comes by I am sure he will be able to answer far better than I can.

What does happen is that the readout text ever so slightly dances around the CRT, a pixel or two horizontally and vertically, at certain times, and it is by design. Unless I'd read about it in the manual, I probably wouldn't have noticed it to think anything much about it.

In my experience with the "non-B" 2467, and the 7104 mainframe that had the same microchannel plate CRT - the shutdown timer was most definitely proportional to the intensity that was dialed in.  It would range from less than 1 minute for high intensity settings, and more than 10 minutes when the intensity was dialed down to a somewhat dim value.

[SoundTech-LG]

OK. so it was really, really messed up!!!

Display kept getting dimmer, and dimmer, and then got to the point that no matter what I did I could not get ANY display at all! I started worrying the CRT was failing, and died. But no! Did NOT adjust the CRT bias or touch anything in the CRT ckt. at all.

I checked all the PS DC voltages. Spot on. Then tried to adjust the DAC per the adjustment section in the svc man. It would not adjust to the 2.5 volt total range, and kept getting worse, and worse! Studied that bridge ckt. with the adjust pot, found R2013 10k COMPLETELY open. There was a small amount of electrolyte leakage from the caps, but this area looked perfect! No idea why that resistor would open!

So to my surprise this fixed not only all the brightness issues, but also many other problems that I was certain were going to require a complete CAL. NOT SO!!!  DAC adjusted just fine. Everything seems to be working now!!! Previously amplitude, and Freq measurements were pretty far off. I'm so pleased to get this baby runnin' like it should be. About $300 and a little troubleshooting. Yippeee!

Thank you all so much for your support, and guidance. Reassuring to get informative replies. Thanks!

Martin

[tautech]

Good job Martin, your perseverance paid off.
Yep I often wonder why silly little things like a 10K resistor fails too.
Maybe its age or a crook one from new, probably why you ended up with it.

You learnt a lot and that will stand you in good stead to keep her running until you decide to get a new DSO.   

[SoundTech-LG]

Hi,

Thanks!

yeah, I use those DSO thingies at work sometimes, and I do have a Tek THS-720A here too, but I still prefer the analogs!!!
I guess you get so used to that warm green glow after so many years, that everything else seems cold by comparison

So fixing the 2467B was the good thing that happened today. The other thing that happened, was I found out I need an angiogram, maybe a stent, or even bypass surgery.
Maybe it's a really good thing... I might live to use the damn thing!

[tautech]

The other thing that happened, was I found out I need an angiogram, maybe a stent, or even bypass surgery.
Maybe it's a really good thing... I might live to use the damn thing!

Well at least the stress of the Tek has passed now.
But you need to know: old CRO's will do that to you. 

Good luck with the angiogram and stuff.

[BravoV]

Studied that bridge ckt. with the adjust pot, found R2013 10k COMPLETELY open. There was a small amount of electrolyte leakage from the caps, but this area looked perfect! No idea why that resistor would open!

Great to hear the happy ending. 

Martin, any chance you haven't throw away that resistor, please, shoot a close up and post it here.

Who knows, probably this will help someone in the future.

[SoundTech-LG]

Yes, I saved it, and taking it to work to look under the microscope at it. I'll try to get some photos of that. The only thing I thought I noticed was a bit darker colorization of the red coating on top of the resistor. Maybe there was some electrolyte that seeped in, dried, and no evidence on the outside of the part. I'll try to get photos as is, and then scrape/pop the top off, see what's going on inside...

I have read at least one post on here, and elsewhere, that describe the symptoms of this problem exactly. I never saw any resolution of the problem posted/detailed.

[SoundTech-LG]

In the article by H. Holden, he found the adjust pot destroyed. R2013 is one resistor away from it...

http://www.worldphaco.com/uploads/TEKTRONIX_2465b_OSCILLOSCOPE_A5_BOARD_REPAIR.pdf

[SoundTech-LG]

R2013 is in the DAC adjust bridge. Problems in the DAC ckt. can cause a myriad of problems. I had what looked like calibration errors, and a CRT dimming problem that eventually worsened to the point of no display at all. There were also several error messages at boot before the CRT was no longer visible. When I tried to do the DAC adjust procedure, it would not adjust properly.

Here are some photos of what I found with R2013. When I cleaned the top with IPA, the coating easily flaked off in one area. Underneath the traces were corroded apart. The resistor appeared very normal on the outside mounted in the board. Hope this helps someone.

[BravoV]

Martin, thanks for sharing those really nice macro shots of the dale smd resistor. 

I can't see clearly the resistor coating color as its quite dark, is it dark brown like below example shot of mine ?

The R2013 I have is the dark brown colored without any printed label on top of it.

[SoundTech-LG]

Yes, I just looked at the board again. Those 3 resistors in a row
(ending in R2013) are darker, and do not have values printed on them. so really, there was no outward indication this part had failed at all. The creeping electrolyte went in, and left no trace on the outside. All those resistors need to be measured if you suspect a problem (no DAC adjust/out of range) in that vicinity.

[BravoV]

Thanks, and your experience is an important lesson, that the cap's leaked electrolyte can affect the SMD's coating, not only pcb traces.

Noted ! 

Once the cap's electrolyte leaked, and even the surrounding resistors or other components that still look perfectly good, we will have to assume that they already corroded or starting to corrode the metal inside. 

[mskobier]

sparkybg,
    I have a 2445B I recently acquired that I am in the process of repairing (channel 1 trace dead). Once I get that sorted out, I plan on doing the capacitor change out on the PS and A5 boards (no indication of leakage at this time). Since I will already have the A5 board out, I plan on also removing the Dallas chip, and replacing it with the Fram FM16W08 chip and adapter board. I have made a video of the current EXER 02 settings (thanks for the idea Markl) since it is probable that I will loose the settings (25 year old battery). Back in mid Jan  in this thread you discussed doing some NVRAM programming using the FRAM chips, and I have been reading that info with much interest. I also have a programmer that will work with both the Dallas chips and the FRAM chips. I have already downloaded a copy of the bin file for a 2465B and have been examining the file structure and information contained in the file using the program that came with my programmer. Your posts back then showed the calibration info from the 2465B would work in a 2445B with the appropriate data modified. I have programmed a couple of chips in my day (20 years ago), but I am having some difficulty in understanding how the EXER 02 data correlates to the HEX data in the downloaded file. Is it possible that you could give me a brief explanation on how the two data sets compare. Thanks in advance for any assistance.

Mitch

[sparkybg]

sparkybg,
    I have a 2445B I recently acquired that I am in the process of repairing (channel 1 trace dead). Once I get that sorted out, I plan on doing the capacitor change out on the PS and A5 boards (no indication of leakage at this time). Since I will already have the A5 board out, I plan on also removing the Dallas chip, and replacing it with the Fram FM16W08 chip and adapter board. I have made a video of the current EXER 02 settings (thanks for the idea Markl) since it is probable that I will loose the settings (25 year old battery). Back in mid Jan  in this thread you discussed doing some NVRAM programming using the FRAM chips, and I have been reading that info with much interest. I also have a programmer that will work with both the Dallas chips and the FRAM chips. I have already downloaded a copy of the bin file for a 2465B and have been examining the file structure and information contained in the file using the program that came with my programmer. Your posts back then showed the calibration info from the 2465B would work in a 2445B with the appropriate data modified. I have programmed a couple of chips in my day (20 years ago), but I am having some difficulty in understanding how the EXER 02 data correlates to the HEX data in the downloaded file. Is it possible that you could give me a brief explanation on how the two data sets compare. Thanks in advance for any assistance.

Mitch

Here I already explained it:

The calibration constants from exercise 2 are located at addresses 1E00 - 1FFF in the NVRAM.

This means that if you see "12BC - 00" this means the word at address 00 is "12BC" hex. This is stored on chip at address 1E00. So on address 1E00 on the chip you will have 12 hex, than on address 1E01 you will have BC hex. Than, if you got "3456 - 01" on the screen, this is stored as 34 hex on address 1E02 and 56 hex on address 1E03. And so on...

My chip was nearly as old as yours, and in the last 6-7 years it maybe had less than a hour per month  working time. Hope for the best - maybe you will be able to read it and program it to the new chip flawlessly, just as I did.

[SoundTech-LG]

Wikipedia notes 01E is for joining two scopes with one sync pulse.

Hi, checked Wiki, but only found this...

Oscilloscope Options

The important oscilloscope options are:

01 - Digital Multimeter

03 - Word Recognizer Probe Pod (P6407)

05 - Video Waveform Measurement System

06 - Counter/Timer/Trigger (CTT)

09 - Counter/Timer/Trigger (CTT) with Word Recognizer (WR)

10 - GPIB Interface

11 - Probe Power

1E - External Clock

22 - Two additional Probes

1R - Prep for rack mounting kit

While you could connect the Ref Signal to 2 scopes, as was suggested, it's not going to do much for you.
The External Reference Option 1E only works if you have Option 06 CTT, or Option 09 CTT with Word Recognizer installed in the scope.
All it does is give you one more digit, 8, instead of 7 on the CTT display, and the accuracy is determined by the stability, and precision of your Reference signal. I took a couple shots with the EXT Ref signal, (in this case a GPSDO) and a Rubidium 15mHz source for the input, and also from a Gigatronics 12720A for the (spec limit) 150mHz signal as well. You can see the 8 digit display, and increased accuracy. Quite an easy Option to install, just a BNC, and a plug for the J5990 header. Unless you have a Pre B05xxxxx Scope, then you need to install a resistor per TimB 08/12/2015

[ronnie_murphy]

I have a tek 2445B scope that has the dreaded test 05 44 error which says a voltage is too positive.  Question 1, is this the right to ask this question and if so where or how do I find that voltage so I can understand why it is to high.  I checked all the voltages on the J119 IC socket and found all were within spec.  Also what on the main PCB would cause such an error.   

[SoundTech-LG]

According to the trouble chart 05 44 fail, you need to check U500 pin28. if you are triggering on an input signal, or doing line triggering, you should have 425 mV there. Check that first.

[mskobier]

sparkybg, I sent you a private message.

Mitch

[siggi]

I have a tek 2445B scope that has the dreaded test 05 44 error which says a voltage is too positive.  Question 1, is this the right to ask this question and if so where or how do I find that voltage so I can understand why it is to high.  I checked all the voltages on the J119 IC socket and found all were within spec.  Also what on the main PCB would cause such an error.   

This is often due to trouble on he A5 board, in the DAC machinery. The most common fault is leakage from SMD capacitors on that board.
You may want to join the TekScopes yahoo forum, there are good resources there in the archives.


Sent from my iPad using Tapatalk

[SoundTech-LG]

I'd agree with that assessment, having seen what happens with problems from unseen corrosion in the DAC adjustment bridge circuitry.
The DAC sets up quite a few voltages, and mine had several problems all exhibiting at once, then the CRT display died altogether.

[ronnie_murphy]

Thanks for your quick response.