Author Topic: TEK 2467B Odd failure  (Read 16341 times)

0 Members and 1 Guest are viewing this topic.

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #25 on: February 02, 2017, 02:04:28 pm »
Hello Mark: Indeed the refs at 14 and 15 of U2101 are unstable, fluctuating ~ 0.124....0.136V.

The Refs are also fluctuating, the  +1.36V on J511 pin 24, and -1.25V on J511 pin 15.

Finally there is a 10V REF  bus nearby U2101, bypassed by SMD tant, C2010.



Q+WOW! VERY unstable ~ 7.5..8.15V with dips much lower!

The plot thickens. Perhaps a failing SMD tant, overloading this critical reference?

Cheers,

Jon
Jean-Paul  the Internet Dinosaur
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #26 on: February 02, 2017, 02:53:09 pm »
Ok, let's chase that.  All bets are off if we have the power supply bouncing around.  Is it jumping on its own without turning the intensity knob?

Previously you had checked the output of the intensity pot and saw no instability there.  That is simply a wiper between the +1.36V rail and ground.  Which should have said the +1.36V supply was ok (it's the same +1.36V that you're looking at on J511 pin 24).  But I'll bet that's not stable now too.

Could be a failing cap, but let's track this backwards and see what we can find.

You had already checked the power supply voltages.  Are any of those unstable at J119 (as per table 5-1)?  It includes the +10V supply.  (Note: Table 5-1 is in the cal section.  I wouldn't try to adjust anything at this point.)
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #27 on: February 02, 2017, 03:03:41 pm »
Hello Mark: Indeed the refs at 14 and 15 of U2101 are unstable, fluctuating ~ 0.124....0.136V.

The Refs are also fluctuating, the  +1.36V on J511 pin 24, and -1.25V on J511 pin 15.

Finally there is a 10V REF  bus nearby U2101, bypassed by SMD tant, C2010.



Q+WOW! VERY unstable ~ 7.5..8.15V with dips much lower!

The plot thickens. Perhaps a failing SMD tant, overloading this critical reference? How to distinguish a problem in the 10V ref loads and bypasses vs at the 10V Ref power ckt?

Cheers,

Jon
Jean-Paul  the Internet Dinosaur
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #28 on: February 02, 2017, 04:26:38 pm »
Mark! WELL......

at J119, all bulk power seems OK, BUT -1.25 and + 1.25 are fluctuating around the normal values.

The 10.00 REF is fine there, BUT I still see it lower and fluctuation on A5 at C2010! Clews are accumuclationg! How can it be good at J119 and bad at the A5 board?

YIKES!

Enjoy,


Jon
Jean-Paul  the Internet Dinosaur
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #29 on: February 02, 2017, 04:28:01 pm »
Forgot: YES the unstable levels occur regardless of intensity control.
j
Jean-Paul  the Internet Dinosaur
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #30 on: February 02, 2017, 04:55:09 pm »
Mark! WELL......

at J119, all bulk power seems OK, BUT -1.25 and + 1.25 are fluctuating around the normal values.

The 10.00 REF is fine there, BUT I still see it lower and fluctuation on A5 at C2010! Clews are accumuclationg! How can it be good at J119 and bad at the A5 board?

YIKES!

Perhaps you have a bad connection on J251?  It's rather convoluted, but according to diagram #13 it comes right from the regulator board (also shown on schematic #12, location 4B).  The 10V ref is on pins 15 and 16 for the A5 board. 

The J119 +10V ref is it's own connection from the regulator board.

EDIT/ADD: J119 is on schematic #11.  I really hate these schematics with separate power distribution diagrams.  It's really obfuscated.

EDIT/MORE: The +10V ref makes its way onto the A1 board via J121 pin 3 (diagram #10, location 1N).  From there it goes to the J119 test connector.  (If anything, I'm posting this so I don't have to figure out the path again.)
« Last Edit: February 02, 2017, 05:16:55 pm by MarkL »
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #31 on: February 02, 2017, 05:16:53 pm »
Hi Mark: I pulled and reseated J251, no easy way to probe it. Looking at the 10V ref at C2010, and flexing the cable at both ends, no effect. The voltage fluctuates about 8.5V. Yes at J119 on main board, I get a solid 9.99V!

Really confusing.

Jon

Jean-Paul  the Internet Dinosaur
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #32 on: February 02, 2017, 05:19:38 pm »
You could try a temporary jumper between J119 where the +10V ref is good, and the purported +10V ref connection on C2010.

With the unit off you could also check the resistance between those two points.
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #33 on: February 02, 2017, 05:37:51 pm »
Hi: It seems to have a stable 1.113 K between J119 pin 4 and C2010 + side. Again very odd!

Jon
Jean-Paul  the Internet Dinosaur
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #34 on: February 02, 2017, 05:54:16 pm »
Mark!

I checked from J119 pin 4 in A1 to the pins of the ribbon cable and connector P251 pins 15, 16: Perfect connection.

BUT probing from the plug on the board at J251 pins 15 and 16, to C2010, Voila!

One is open and the other about 1.8K and unstable.

So the open is on board A5!

 Perhaps a bad solder joint or a leaky SMD tant has affected the trace.

Next step is to connect the 2 points (will use a R sub box!)

Jon
Jean-Paul  the Internet Dinosaur
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #35 on: February 02, 2017, 06:21:31 pm »
Hello again Mark! Well something is up:

Connecting a 1K...50 Ohm between J119 pin 4 and C2010 +

a, I can pullup to 9.99V as resistance lowers.

b. there is a huge effect on intensity and stability, it DOES seem to fix it!

c. BUT if I remove the jumper, the scope seems to  no longer shows the fading and instability symptom!

You see the traces,  just at a lower intensity with the 8.85V ref, but it is not fading out every few seconds.  So we are VERY close.

The resistance between the 2 points is a stable 1.275 ohms.

If I probe at J251, pins 15, 16, one is open and at the other pin I get the same R to C2010.

Time to pull out the board?

Best,

Jon
Jean-Paul  the Internet Dinosaur
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: TEK 2467B Odd failure
« Reply #36 on: February 02, 2017, 06:25:58 pm »
Hello Mark: Indeed the refs at 14 and 15 of U2101 are unstable, fluctuating ~ 0.124....0.136V.

The Refs are also fluctuating, the  +1.36V on J511 pin 24, and -1.25V on J511 pin 15.

Finally there is a 10V REF  bus nearby U2101, bypassed by SMD tant, C2010.



Q+WOW! VERY unstable ~ 7.5..8.15V with dips much lower!

The plot thickens. Perhaps a failing SMD tant, overloading this critical reference? How to distinguish a problem in the 10V ref loads and bypasses vs at the 10V Ref power ckt?

Cheers,

Jon
A known failure mode of Tantalum caps is internal shorts that self heal and they can be a nightmare to diagnose.
Some small mention of it here:
https://en.wikipedia.org/wiki/Tantalum_capacitor#Failure_modes_and_self-healing_mechanism

Google 'tantalum cap failure modes' for more.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #37 on: February 02, 2017, 06:33:19 pm »
VOILA! Mark, 

Looking very carefully , I see that there is corrosion around C2520 -15V bypass 0.1 mf and it is JUST near a round can SMD tant, C2113 15 MF SMD -15V bypass

I now think the C2113 is failing, partially, the leakage creept out towards C2520, and has eroded the +10V ref trace!

I needed very bright light and an optiovisor to spot this!

Perhaps a trace for the 10V ref is running under C2520 on the top layer!


(I am very familiar with the tant failures especially on these TEK with SMD tants. )

Time to pull A5 and recap, I think!

FINALLY the end is in sight!

Enjoy,


Jon

Jean-Paul  the Internet Dinosaur
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #38 on: February 02, 2017, 06:48:43 pm »
VOILA! Mark, 

Looking very carefully , I see that there is corrosion around C2520 -15V bypass 0.1 mf and it is JUST near a round can SMD tant, C2113 15 MF SMD -15V bypass

I now think the C2113 is failing, partially, the leakage creept out towards C2520, and has eroded the +10V ref trace!

I needed very bright light and an optiovisor to spot this!

Perhaps a trace for the 10V ref is running under C2520 on the top layer!


(I am very familiar with the tant failures especially on these TEK with SMD tants. )

Time to pull A5 and recap, I think!

FINALLY the end is in sight!

Enjoy,


Jon
Excellent!  Sounds like you got it!

Definitely time to recap if you're planning on hanging onto this unit.  But I would fix the trace and replace that one leaky one first to verify it's fixed before starting the recap operation.

Have fun and let us know the final outcome!
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #39 on: February 02, 2017, 06:53:33 pm »
I had already typed up an answer to your question just before you discovered the discrepancy between the two +10V points.

Obviously informational at this point...

...
How to distinguish a problem in the 10V ref loads and bypasses vs at the 10V Ref power ckt?
...

The usual method is to isloate (unplug) the power supply outputs, put them under load, and measure them.  For this unit it's described in the "Regulator Troubleshooting Procedure" (pdf page 474).  Since you only care about the +10V ref, you might be able to get away without loading the other outputs.

The +10V ref is driven by an LM324 (see schematic #10), so it doesn't supply much current and is short circuit protected (max current is 60mA).  One thing you could try is to connect an external power supply to the +10V reference as-is, set the current limit to say 100mA, and try to supply the reference yourself.  This isn't 100% though, since you might end up masking the problem or fighting with an issue in the output stage of the LM324.

If it turns out not to be the supply, finding an intermittent short in a bunch of parallel connected components can be difficult.  It involves isolating branches of the suspect circuit.  For the first foray, this would be disconnecting the various connectors where the +10V ref flows to at least get it down to the board.

Another approach is to use one of these PCB current probes:

  http://www.aimtti.com/product-category/current-probes/aim-i-prober-520

It's an investment if you want to expand your collection.  Another option is the old HP 547A current tracer probe.

Thermal cameras can sometimes find a short, but in this case we're talking about a small amount of power which wouldn't generate much heat.

Then there's always the shotgun approach of starting with the most likely failures (like those tantalums you mentioned), and start removing them until you find it.  More drastic measures can also include cutting traces and lifting component legs.
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #40 on: February 02, 2017, 08:07:34 pm »
Hello MarkL: I now think the 10V ref fluctuations (it was dipping to a few V at times) were causing all the symptoms including the 50 Ohm overload fault. With the jumper in place it works 100%!

What is your thought?

PIX: In these macro photos, we see  the leaky  tant looks OK,C2113  but you can see the leakage at the adjacent 0.1 MF  C2520.

I should replace ALL the SMD lytics!  I wonder if anyone has the best DK or Mouser part numbers to recap A5 in this 2467B?

Again many thanks for all your great assistance!

Enjoy!


Jon

PS: You are Engineer? Ham? Tech? In US or overseas?
 




Jean-Paul  the Internet Dinosaur
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #41 on: February 02, 2017, 09:12:04 pm »
Hello MarkL: I now think the 10V ref fluctuations (it was dipping to a few V at times) were causing all the symptoms including the 50 Ohm overload fault. With the jumper in place it works 100%!

What is your thought?
Great!  Can't argue with something that's working now.

I suppose you should do the recap since you already have PCB damage.  Perhaps someone lurking on this thread can chime in with a Digikey/Mouser/etc. parts list.  There's also this thread:

  https://www.eevblog.com/forum/testgear/tektronix-2465b-oscilloscope-teardown/

which has attracted a myriad of topics on the 246x series.  You could ask over there for a recap parts if no response here, or try the forum's search page.

If you read the thread you will also learn that the Dallas NVRAM in these scopes are at or beyond their EOL.  There's instructions there on reading out its contents and replacing that as well.

Quote
PS: You are Engineer? Ham? Tech? In US or overseas?
An engineer in the US.  I mostly do embedded HW and SW design which doesn't really intersect much with these old scopes.  I have a 2465 and I just enjoy working on them because they're so well documented and repairable.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: TEK 2467B Odd failure
« Reply #42 on: February 02, 2017, 09:45:32 pm »
Those SMD electrolytics are awful, I don't even know why they're as bad as they are but I've replaced thousands of them, usually tantalum or ceramic chip capacitors work well. It was a big problem in older camcorders back when those were worth fixing, and I've repaired a lot of vintage Mac computers that used piles of them. The electrolyte that leaks out is corrosive and damages the board if it isn't caught, as well as it's conductive so it causes malfunctions so it needs to be cleaned off. It causes a chemical change in solder it gets on and seems to raise the melting point. I find the easiest way to replace the caps is usually snip them as close to the base as you can get with a pair of diagonal cutters (watch for flying parts!) and then pick off the remains and remove the pins one at a time with a soldering iron.
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #43 on: February 03, 2017, 12:13:29 pm »

Hello Mark and James:

In general the 2465B/2467B are the ones worth keeping, the earlier versions are not nearly as desirable.
They are real workhorses and the LAST "real" analog CRT scopes from Tektronix.
The 2467B with microchannel plate intensifier can display very low rep rate or one shot fast transients. But the CRT area is a bit smaller than 2465B.

Having worked with 2465X since 1990s, I am well aware of the infamous SMD lytic leakage and Dallas NV RAM issues.
Of several here, just one has the NV RAM failure and this unit is the first I have seen affected by cap leakage.
It is only with this particular unit, that I have seen the very tiny corroded SMD cap pads, and small visible leakage.
Since most lytics have a 2000 hr life at rated  ripple current and temp, its possible they begin to fail in very high hours units.
Indeed this one shows 1000 starts and 18K hours! The recapping on A5 is easy, on PSU and other boards, its more involved.
Yes the cap replacements are on the other forums. To James, points well taken, I have the Metcal soldering iron (I designed it in 1990s!) for really good SMD work.

For NVRAM, the drill seems to be desolder the old one, add a  socket, and program a new chip. Or do a complete recalibration. Beyond my level of work, but I may find a friend who can do it.

I will post a final note after its finished.

Again many thanks for the very fine troubleshooting from Mark!

Cheers,

Jon
Jean-Paul  the Internet Dinosaur
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: TEK 2467B Odd failure
« Reply #44 on: February 03, 2017, 03:32:29 pm »
You're most welcome!  Fun for me too since there's always something to learn.  Good luck on the recap!
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #45 on: February 04, 2017, 02:07:01 pm »
Hello again: Anyone had experience with QService Greece lately for TEK parts?

Need the Dallas NV RAM and an odd knob shaft etc.
377-0512-03 INSERT,KNOB


156-2991-00 Dallas NV RAM DS-1225Y-200 64K Programmable NV-SRAM


They seem to have no stock on the NV RAM: Looking for one programmed....and with a fresh not 20yrs old date code!

Any tips?


Cheers,

Jon

Jean-Paul  the Internet Dinosaur
 

Offline cheeseit

  • Regular Contributor
  • *
  • Posts: 168
  • Country: dk
Re: TEK 2467B Odd failure
« Reply #46 on: February 04, 2017, 02:42:22 pm »
I'm sure someone in the US can help you with the NVRAM but I've used QService a few times and would definitely do so again. They/he provided excellent and helpful service and replied very quickly every time.
Ordered one screw less than I needed last time and when I asked if I could somehow order and pay for one more and have him include it in the first order so I didn't have to pay shipping for a single screw he replied that no order was necessary and he would give me one more. Only a €2-3 screw but still really nice of him. Only problem is that parcel T&T ends when it leaves Greece but he makes a note of this.
 

Offline jonpaulTopic starter

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: fr
Re: TEK 2467B Odd failure
« Reply #47 on: February 04, 2017, 03:10:59 pm »
Actually, the NVRAMS are avail in USA BUT very old date codes and unprogrammed. Seems that QService used to provide them preprogrammed at extra cost.

About the knob insert, its not the insert, the the Intensity 5K pot whouse extra long shaft is bent, that I need, it is

311-2313-01 Potentiometer 5K


With intl post, etc I prefer to find a USA source.

Anyone have a cache of parts or a 2465B/2467B junker or parts unit?


Cheers,

Jon
Jean-Paul  the Internet Dinosaur
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: TEK 2467B Odd failure
« Reply #48 on: February 04, 2017, 05:33:01 pm »
Hello again: Anyone had experience with QService Greece lately for TEK parts?

Need the Dallas NV RAM and an odd knob shaft etc.
377-0512-03 INSERT,KNOB


156-2991-00 Dallas NV RAM DS-1225Y-200 64K Programmable NV-SRAM


They seem to have no stock on the NV RAM: Looking for one programmed....and with a fresh not 20yrs old date code!

Any tips?


Cheers,

Jon
There's mention of them available from Digikey in this post:
https://www.eevblog.com/forum/testgear/tektronix-2465b-oscilloscope-teardown/msg1026013/#msg1026013
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: TEK 2467B Odd failure
« Reply #49 on: February 04, 2017, 06:59:47 pm »
Just cut the old battery out and replace it with a standard CR2032 holder. Then you won't have to source a whole new chip ever again. The battery is on the top under a thin layer of epoxy.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf