Author Topic: TEK 2467B Odd failure  (Read 11111 times)

0 Members and 1 Guest are viewing this topic.

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
TEK 2467B Odd failure
« on: January 29, 2017, 12:44:03 pm »
Hello all: Just got a TEK 2465B, SN>B050...., seems to be in good condition, and works for about 2 minutes.   

Then, even with no input, the relays click, it displays 50 Ohm Overload and the display becomes unstable and then fades off, regardless of intensity setting.   

It is consistent every time it is started. No errors on the startup self test.

I suspect a power supply issue or perhaps a failing tantalum cap causing power overload or short after it warms up.

Anyone ever seen such behavior? Any troubleshooting tips?

MANY THANKS

Jon
Jon Paul
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 10919
  • Country: gb
    • Having fun doing more, with less
Re: TEK 2467B Odd failure
« Reply #1 on: January 29, 2017, 01:39:28 pm »
Is it a 2465b or 2467b?

Anyway, look at the caps and other components on the control board near the DAC; there are many guides to this, so I won't repeat them poorly.

Also look at the mains X/Y rated filter caps; replace them if they are crazed before they detonate a series resistor and the PCB prepreg. Again, there is an extremely explicit pdf of how to do that.

Don't forget that if operating out of the case, you will need to arrange for external airflow over the hybrids and other ICs.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #2 on: January 29, 2017, 03:40:46 pm »
Hi there many thanks for the fast response.

Yes 2467B.

The readouts seem stable issue affects  only the trace intensity. Each time it blanks out, moving the intensity control up a bit brings the traces back for a few seconds, then they fade out again. The 50 Ohm Overload /relays clicking has not occured again.

I have opened it up and checked for the A5 SMD caps failure, they seem undamaged. The HV board visually seems OK.

I now suspect either the intensity pot/FP mux or a fault in the Z-axis circuit.

Next stop: Pull off PSU cover and see if any caps are leaking, and check all LV PSU voltages.

Any further input is appreciated.

With Kind Regards,



Jon

Jon Paul
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 10919
  • Country: gb
    • Having fun doing more, with less
Re: TEK 2467B Odd failure
« Reply #3 on: January 29, 2017, 04:19:20 pm »
Always check the voltages for value and ripple; ripple indicates higher ESR, which can be independent of capacitance.

If it is like the 2445/2465, then you can look at the output of the single DAC directly, and also the caps used to store the demultiplexed voltage.
If those voltages are stable, then consider the LT version of the Z-axis voltage waveform, and whether the DC restorer is working correctly.

Without checking, I would assume that the overload relay threshold is directly or indirectly dependent on the DAC output.

The LT Z axis waveforms are relatively "safe". The action of the DC restorer is to translate the voltage changes up to the CRT cathode's voltage, i.e. 2.5-2kV => less safe :)

I have, on a 465 and with a great deal of care, managed to use a DC voltmeter to measure the mean DC and mean AC value of  Vgk directly. First test was to check that the mean  DC Vgk was suitable for ensuring blanking and brightness.  Next I noted that changing the mean AC value of the LT Z axis waveform by 10% changed the mean AC value of Vgk by 10%, implying that the DC restorer was working OK. In my case the problem was a detached wire inside the CRT itself.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline WaveyDipole

  • Frequent Contributor
  • **
  • Posts: 470
  • Country: gb
Re: TEK 2467B Odd failure
« Reply #4 on: January 29, 2017, 06:12:41 pm »
I have something like this on my old Tek 465. It would run for a couple of minutes and then the display would fade away. Turn it off, and then back on a again and the trace would be back for a while and then disappear again.

It turned out to be a faulty metal can capacitor on on the -8v line. It would seem that the cap was shorting internally upon warming up. There was no external sign of a problem. The line would go from -8v to zero which affected all circuits dependent on it. I did also find a shorted tantalum cap, but this appeared to be unrelated. Replacement of the can resolved the problem but only for a short while until the rectifier bridge failed. Evidently the fault on the old cap had taken its toll on the rectifier as well. With both replaced, the scope has been fine so far.

I know we are dealing with a different and more modern scope here so the problem may be entirely different, but I thought I'd mention it for what its worth.
« Last Edit: January 29, 2017, 06:16:31 pm by WaveyDipole »
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #5 on: January 29, 2017, 08:19:20 pm »
Hello again:

1, PSU volatge and ripple are fine, I checked at the main board PS 14 pin header, J119, everything is within spec.

2, At the Z Azis test point, I can see the upper waveform peaks unstable and jumping as the trace brightness fluctuates.

3, Any intensity control setting lasts for just a few seconds, then the traces fade out.

4, Readout intensity is not affected.

It is either an input to U950 the Z axis hybrid, or the hybrid itself.

Again, any assistance is appreciated!

Kind Regards,

Jon
Jon Paul
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 10919
  • Country: gb
    • Having fun doing more, with less
Re: TEK 2467B Odd failure
« Reply #6 on: January 29, 2017, 09:25:49 pm »
The manual will describe how the intensity is controlled.

Look at any relevant DAC output voltages. Follow them through the circuit.

The correct readout intensity makes it likely that u950 is ok.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #7 on: January 30, 2017, 07:32:14 pm »
Does the trace appear momentarily no matter if you turn the intensity up or down?

The DAC output for the display intensity is on U2630 pin 7 ("DIA"), as tggzzz suggests.

If that isn't behaving, you could check that the display intensity ("DI") signal into the ADC is ok.  That's on U2501 pin 2 and should vary smoothly from 0 to 1.36V as you turn the intensity knob.

All the DAC output holding capacitors are being refreshed continuously, so it's odd that the intensity fades only after a change.

Maybe a 10 second video of the problem would help.
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #8 on: January 30, 2017, 07:39:52 pm »
Hi there: The DIA and DI signals are indeed affected. To clarify, the intensiy is normal for a few seconds, then either fades out OR entire display becomes unstable. After the fade out, if you move the front panel control to a new setting the sequence repeats.

With the entire display becoming unstable, and occasionally getting the 50 Ohm Input Overload message (with no inputs) I stared to check all power more carefully. Yet all power voltage and ripple seem to be normal. I inspected for any leaky tantalum SMD power bypass caps, all seem intact and none overheating as failed tants often do.



Finally I have the QService 2465B/2467B manual, for SN > B050.

BUT the A5 control board seems different so I cannot even trace the signals further.

Any assistance will be appreciated.

I will try to make a video of the issue. 


Many thanks again!

Jon
Jon Paul
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 17480
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: TEK 2467B Odd failure
« Reply #9 on: January 30, 2017, 07:46:48 pm »
You might find this can offer some guidance:
http://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf
Avid Rabid Hobbyist
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #10 on: January 30, 2017, 08:52:05 pm »
Hi there: The DIA and DI signals are indeed affected. To clarify, the intensiy is normal for a few seconds, then either fades out OR entire display becomes unstable. After the fade out, if you move the front panel control to a new setting the sequence repeats.

With the entire display becoming unstable, and occasionally getting the 50 Ohm Input Overload message (with no inputs) I stared to check all power more carefully. Yet all power voltage and ripple seem to be normal. I inspected for any leaky tantalum SMD power bypass caps, all seem intact and none overheating as failed tants often do.
The DI and 50ohm overload circuits all come into the ADC.  It could be that the input multiplexing is having issues.

We should first confirm the signals coming into the ADC are stable.

DI comes into one side of R2702 before it goes to U2501 pin 2.  Is the DI signal stable on the other side of R702?  You can also use P512 pin 29 if you can't get to R2702.

If you look at the Ch1 and Ch2 overload signal inputs on U2601 pins 15 and 12, are those unstable also?  How about on P511 pins 5 and 2 which is before their ADC input resistors?


When you say the display is "unstable", what do you mean?

I'm looking at schematics for SN > B050.  If you think there's a mismatch for your serial number, a picture of your board could be helpful.
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #11 on: January 31, 2017, 02:36:04 pm »
Hello again:

Attached are photos: The A5 board, the display first seconds, afterwards and same sequence if intensity pot is changed.

The BIG issue: The A5 boards are not as in the QService and all other 2465B/2467B manuals for SN over B050! Mine is B050878,


Mine has NO separate backup battery, the Dallas NV RAM, and uses almost all 74HC logic rather than 4051, etc! So the A5 schematics and board layout are totally different! I have another with the exact same A5 board type.

So rather hard to continuer troubleshooting.

Does anyone have the A5 sch and PCB placement for this version?

Again, any assistance is appreciated!

With Kind Regards,

Jon


Jon Paul
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #12 on: January 31, 2017, 04:09:38 pm »
Does anyone have the A5 sch and PCB placement for this version?
I've been looking at "Tek 2465B 2467B service OCR with schematics":

  http://www.ko4bb.com/getsimple/index.php?id=manuals&dir=Tektronix/Tektronix_-_2465B_Oscilloscope

There are two of them, named slightly different.  They're identical.

There are separate pages for SN >= B050000, and SN < B049999.  You're board certainly looks like SN >= B050000.  A spot check on a few chips also matches.

The scan isn't the greatest, but it's readable.  If you want a really good version, you could buy a PDF copy from artekmanuals.com.  Their manuals are professionally scanned, and inexpensive for what you get.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #13 on: January 31, 2017, 04:20:50 pm »
Hi there: The DIA and DI signals are indeed affected. To clarify, the intensiy is normal for a few seconds, then either fades out OR entire display becomes unstable. After the fade out, if you move the front panel control to a new setting the sequence repeats.
I just re-read this.

To clarify:  When you say "affected", do you mean that DI and/or DIA smoothly tracks the intensity control when you turn it, or does the voltage drop over a few seconds by itself (or possibly appear erratic)?

We should try to find the earliest point where we can observe a correlation with the fading.  And not forgetting about the overtemp and relay clicking.  It's very likely something in common to all.
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #14 on: January 31, 2017, 04:29:52 pm »
Hi there MarkL: WOW! Great link,

Perfect, YES in the REVISIONS at the end IS the exact board and schematic I have for A5.  I had somehow assume the QService Greece printed manual I had was perfect. But it has NO revisions at all!

This is a big help. The scans are fine.

Notes: 1. Check intensity pot at FP, perfect 0 - 1.36VDC. It seems to go from the FP connector on the main board A1 across to the connectors linking to A5 Control.

2, When "unstable" I get a fading trace, it moves vertically and comes back if I move the intensity pot.

I will try to take more photos, after I dig into the schematics.

Affected: The DI seem to be very odd, with pulses and the baseline jumping about. BUT I am not 100% sure I was probing the correct resistors and IC pins. I will check more carefully and try to get more photos.

The entire thing is so odd, as the PSU seems fine but 3 symptoms happen at one, always with a few seconds delay. A real mystery.

Many thanks again! Back to work!

Cheers,

Jon


Jon Paul
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #15 on: January 31, 2017, 06:15:39 pm »
Hello again MarkL!

Many thanks for these better schematics, you have linked me to,


Now I can see that intensity controld from FP,  RDI and DI are fine, exactly as expected DC controlled by the FP pots.

 These signals flow thru R2701 and R2702, and appear at the MUX, pins 2,5 of U2501.

Looking at the MUX output, again all is normal, I see the MUX signal, and the levels at the corresponding time slots vary as you move the pots.

Thus the problem lies after the MUX but before the Z axis circuity.

I think that board A5 is OK.

Again any further clews are appreciated.

Kind Regards,

Jon
Jon Paul
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #16 on: January 31, 2017, 10:06:24 pm »
...
Looking at the MUX output, again all is normal, I see the MUX signal, and the levels at the corresponding time slots vary as you move the pots.

Thus the problem lies after the MUX but before the Z axis circuity.

I think that board A5 is OK.
...
Well, I wouldn't go that far yet.  I'm assuming you're saying the problem is continuing to happen while DI looks fine.

There's a lot between the ADC MUX output and the Z-axis drive.  Such as the ADC, the processor, and the DAC, for starters.  The processor reads the position of the intensity knob, alters the value as needed, and spits it back out via the DAC.

Are you ready to take a look at DIA: U2630 pin 1?  This is the intensity output from the DAC after the processor is done with it.  (Note: I said pin 7 before; I guess I wasn't zoomed in enough.  Sorry about that...)

Actually, before you look at that, start up the Exerciser Routine #1 (on page 6-15) and poke around with it a bit.  This is reads the various pots on the front panel and prints out their value after conversion by the ADC.  It also includes the DAC output value.  See if you see anything of interest there while turning the intensity pot.
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #17 on: February 01, 2017, 03:06:04 pm »
Hello again MarkL! Many thanks again for your expert assistance!

Before I follow-up on your very fine suggestions, a few  more clews:

0. Both RO and min intensity pots have correct voltages measured at the pot terminals.

1, Readouts are rock solid and intensity is behaving normally as you turn the FP knob.
Since both readout intensity and display intensity follow parallel paths, its odd this affects one but not the other.

2, The symptom happens regardless of use of SWP A, SWP B or both.

3, There seems to be a loss of beam current control, as you can turn intensity way up before it fades out to zero.

Perhaps some clews here?

I will continue later today,

Cheers,

Jon

Jon Paul
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #18 on: February 01, 2017, 04:35:08 pm »
Dear MarkL: I just checked:

1/ Exer 01: Yes perfect in intensity, the hex codes vary as expected and are stable. All other controls operate normally in this test.

2/ DIA: U2630 pin 1: Sure, the  DAC intensity signal is bad:

About 5..6 V DC (with mux noise) when intensity is set to be visible. Varies with control setting.

After the few seconds and the trace fades out, it lowers to 3V, with no mux noise. As soon as you move the control and get a trace again, the exact same behavior at U2630 pin 1.

The similar signal for readout intensity is fine.

Best,


Jon
Jon Paul
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #19 on: February 01, 2017, 10:34:48 pm »
I'm beginning to wonder about one of the the DAC muxes, U2530, or the control inputs to it.  The reason is that you said the trace moves vertically when the fade problem is happening.  U2530, among other things, is responsible for the trace vertical position.

So, you first could take a look at U2630 pin 3 (connected to U2530 pin 2).  This is the input to the opamp from the holding cap which eventually is output on U2630 pin 1.  If U2630 is ok, you should see this input tracking the problem too.

Does the vertical movement happen at the same time as the fade on all the channels?  Or just a particular channel or subset of channels?  Can you confirm that the voltage change is happening on U2530 pins 12, 13, 14, and/or 15?  These are the vertical position outputs for Ch4, Ch1, Ch2, Ch3, respectively.

U2530 has control input signals that are in common to the other DAC mux, U2521.  Take a look at U2521 pin 2.  See if this is moving with the fade problem.  Probably not, but interesting if it is.

You could also take a look at U2530 pin 6.  This is the mux inhibit signal.  It probably has some periodicity to it, but for now just check to make sure the logic levels are sane (<0.7 V or >2.1V).  You should also check the select inputs U2530 pins 9, 10, 11 to see if they're sane too.

Also check TP2420 (+1.36V) and TP2421 (-1.25V).  These control the DAC reference voltage.  From your other observations these should be stable, but let's make sure.
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #20 on: February 01, 2017, 11:29:05 pm »
WOW! Yes I did notice the vertical trace position wandering when the traces start to fade!

I will follow up on your great suggestion later tonight or Thursday!


Again, many thanks!


Jon
Jon Paul
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 11054
  • Country: us
Re: TEK 2467B Odd failure
« Reply #21 on: February 01, 2017, 11:49:39 pm »
What about the DAC reference voltage? Is it external to the DAC?
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #22 on: February 02, 2017, 12:06:19 am »
 U2630 pin 3 (connected to U2530 pin 2).

YES I see the  input tracking the problem 


  U2530 pins 12, 13, 14, and/or 15?  vertical position outputs for Ch4, Ch1, Ch2, Ch3,

YES I see the instability on these pins

U2530 has control input signals that are in common to the other DAC mux, U2521.  Take a look at U2521 pin 2.  See if this is moving with the fade problem. 


IT IS OK

You could also take a look at U2530 pin 6.  This is the mux inhibit signal. 


OK 

 U2530 pins 9, 10, 11 to see if they're sane too.


OK

Also check TP2420 (+1.36V) and TP2421 (-1.25V).  Cannot find these! On parts list but not on schematic or PCB layout.


Whats next?


Thanks Again!

Jon
Jon Paul
 

Online MarkL

  • Supporter
  • ****
  • Posts: 1593
  • Country: us
Re: TEK 2467B Odd failure
« Reply #23 on: February 02, 2017, 01:54:32 am »
...
Also check TP2420 (+1.36V) and TP2421 (-1.25V).  Cannot find these! On parts list but not on schematic or PCB layout.
I must have flipped over to the SN < 49999 schematic for a second.  The TP's on that page.  But you can pick up +1.36V on J511 pin 24, and -1.25V on J511 pin 15.  They should be ok, but it's good to check.

If you want to check the Vref for the DAC as james_s mentions that's on pin 14 and 15 of U2101.  Just check for stability.  All other indications say they should be fine too, but I'm never against being sure.  When troubleshooting it's always an ordered list of probabilities and never absolutes.

Quote
Whats next?
We need to figure out if U2530 is working.

I assume you have another scope you've been using to look at things.  What kind of equipment do you have?

To take a first pass at this, I would trigger the scope on the rising edge of the inhibit input on U2530 pin 6.  Then probe any of the outputs DI or CHx position that are unstable during the problem and look at pin 6 and any of those outputs at the same time.

No outputs should be moving during inhibit high (all the mux output switches should be disconnected).  If any of the outputs moves during inhibit high, U2530 is dead.  I suppose you should also verify it has power on pins 8 (Gnd) and 16 (+5V).

The converse is not necessarily true.  If nothing moves, it doesn't mean it's ok.  We can still test for that, but it's going to need fancier triggering (pattern trigger).

The 50ohm overload problem is still a concern.  The DAC in this unit is also the ADC when combined with comparator U2510.  It's a very busy DAC.  If one of the DAC muxes is misbehaving it could affect the ADC.  Just a theory at this point.
 

Offline jonpaul

  • Regular Contributor
  • *
  • Posts: 156
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #24 on: February 02, 2017, 12:30:46 pm »
Hello Mark:

Yes I have a very extensive lab, TEK 2465B, 2467B, 2467, 7104, 7704, etc plus about 100 plugins including some rare ones. Also 2 pcs Yokogawa DL7704 with logic probes. ( a 40 yrs collection of stuff!)

So I can do a pattern trigger test and save it.

1, Will follow-up today on all your new suggestions.

2, The 50 Ohm OV symptom happened only occasionally, with no inputs, you get the warning as relays click. Only happens during the intensity fault.

More news soon!

Enjoy,


Jon


Jon Paul
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf