Author Topic: Tektronix 2465B oscilloscope teardown  (Read 460787 times)

0 Members and 2 Guests are viewing this topic.

Offline Bryan

  • Frequent Contributor
  • **
  • Posts: 602
  • Country: ca
Re: Tektronix 2465B oscilloscope teardown
« Reply #675 on: March 05, 2016, 02:36:54 am »
Another tell tale sign is the caps will give off a horrible fishy smell once unsoldered.
-=Bryan=-
 

Offline Old-E

  • Regular Contributor
  • *
  • Posts: 57
  • Country: us
Re: Tektronix 2465B oscilloscope teardown
« Reply #676 on: March 05, 2016, 05:01:42 am »
Mr. Sauce,

You wrote:  "Wouldn't the cal data be kinda useless after repairing the power supply board, it would need a recal anyway?"
I don't know of any reason the calibration would change when rebuilding the power supplies.   I counted 9 regulated voltages not counting the CRT HV.   But all of those are supposed to be at specified DC levels.   These voltages and their tolerances, including allowable ripple, are noted on page 5-3 of the "Tektronix Service Manual" and are available for measurement at J119 on the main board (bottom of scope).  Service Manual is available on line.   You may already have all this info.

I have not started on my PS boards yet.   But beware - there is an error in Service Manual regarding value & location of a couple of the caps.   Replacing them with the same ratings as is there now works ok, but removing them all and then using the Service Manual as a guide to install the new ones will be a problem.   Previously, and on page 26 MSO explained in more detail, and made a number of entries on replacing these caps.   Also, I made a recent entry on replacing the fan including the part # that may be of use for you.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 19727
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Tektronix 2465B oscilloscope teardown
« Reply #677 on: March 05, 2016, 07:03:46 am »
Baddies.
All of these probably, hairline cracks on case likely ballooned caps

Quite common in X and Y rated caps, they can survive for years in this condition in a dry climate.
When they finally let go...poof, it's more of an inconvenience than anything else plus the mess they can make.  :palm:
Avid Rabid Hobbyist
 

Offline ChunkyPastaSauce

  • Supporter
  • ****
  • Posts: 535
  • Country: 00
Re: Tektronix 2465B oscilloscope teardown
« Reply #678 on: March 05, 2016, 07:04:44 am »
Mr. Sauce,

You wrote:  "Wouldn't the cal data be kinda useless after repairing the power supply board, it would need a recal anyway?"
I don't know of any reason the calibration would change when rebuilding the power supplies.   I counted 9 regulated voltages not counting the CRT HV.   But all of those are supposed to be at specified DC levels.   These voltages and their tolerances, including allowable ripple, are noted on page 5-3 of the "Tektronix Service Manual" and are available for measurement at J119 on the main board (bottom of scope).  Service Manual is available on line.   You may already have all this info.

I have not started on my PS boards yet.   But beware - there is an error in Service Manual regarding value & location of a couple of the caps.   Replacing them with the same ratings as is there now works ok, but removing them all and then using the Service Manual as a guide to install the new ones will be a problem.   Previously, and on page 26 MSO explained in more detail, and made a number of entries on replacing these caps.   Also, I made a recent entry on replacing the fan including the part # that may be of use for you.

Thanks!  :-+ I have the service manual, but miss understood the difference between a recalibration and performance check & adjustment. Manual saya to do a performance check /adj if any repairs are made to power supply, but doesn't mention the need for a autorecalibration. Autocalibration, has to have correct equipment, performance check & adjust is a lot more flexible.....So hopefully mine still holds the calibration constants...

Thanks for letting me know about the caps, Ill check the pages you mentioned. Ive only removed the 6 silver caps so far.... so hopefully I can sort those out. I didn't take pictures or log because i was going to go via the manual ::)
 

Offline ChunkyPastaSauce

  • Supporter
  • ****
  • Posts: 535
  • Country: 00
Re: Tektronix 2465B oscilloscope teardown
« Reply #679 on: March 05, 2016, 07:07:57 am »
Another tell tale sign is the caps will give off a horrible fishy smell once unsoldered.

Didn't know that...looked it up...Dimethylamine
 

Offline ChunkyPastaSauce

  • Supporter
  • ****
  • Posts: 535
  • Country: 00
Re: Tektronix 2465B oscilloscope teardown
« Reply #680 on: March 05, 2016, 07:19:12 am »
Baddies.
All of these probably, hairline cracks on case likely ballooned caps

Quite common in X and Y rated caps, they can survive for years in this condition in a dry climate.
When they finally let go...poof, it's more of an inconvenience than anything else plus the mess they can make.  :palm:

Hmmmm... I was feeling motivated but now a laziness opportunity...  replace now or be lazy....  :-//

Most of the others have cracking but are flat on the sides at closer look, this one has a bulge on the side (hard to see in pic), so I think this one maybe ready to go if it's not already gone.

I was wondering about the design of these, the casing is to keep humidity out?
« Last Edit: March 05, 2016, 07:27:57 am by ChunkyPastaSauce »
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 19727
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Tektronix 2465B oscilloscope teardown
« Reply #681 on: March 05, 2016, 08:38:26 am »
Baddies.
All of these probably, hairline cracks on case likely ballooned caps

Quite common in X and Y rated caps, they can survive for years in this condition in a dry climate.
When they finally let go...poof, it's more of an inconvenience than anything else plus the mess they can make.  :palm:

Hmmmm... I was feeling motivated but now a laziness opportunity...  replace now or be lazy....  :-//

Most of the others have cracking but are flat on the sides at closer look, this one has a bulge on the side (hard to see in pic), so I think this one maybe ready to go if it's not already gone.

I was wondering about the design of these, the casing is to keep humidity out?
Of course and humidity/and or time spent unplugged will allow ingress, then pop.
As these X and Y caps are across mains AND when they go pop, it's not just the mess but the physical damage the might do to the PCB not to mention the chance of carbonising the top layer and resultant chance of further flash overs in the future.
C'mon do the job properly or not at all, they're easy to get, cheap and peace of mind that they've been replaced will help you sleep well.  :)
Avid Rabid Hobbyist
 

Offline Bryan

  • Frequent Contributor
  • **
  • Posts: 602
  • Country: ca
Re: Tektronix 2465B oscilloscope teardown
« Reply #682 on: March 05, 2016, 09:37:00 am »
There is a few spreadsheets floating around if you need the drop in replacements for the replacement caps. Here is one that I have and slightly modified to correct for some issues that I encountered with wrong pin spacing on some of the caps.
-=Bryan=-
 

Offline ChunkyPastaSauce

  • Supporter
  • ****
  • Posts: 535
  • Country: 00
Re: Tektronix 2465B oscilloscope teardown
« Reply #683 on: March 05, 2016, 07:29:32 pm »
C'mon do the job properly or not at all, they're easy to get, cheap and peace of mind that they've been replaced will help you sleep well.  :)


Thanks. Ill be doing it right. Laziness opportunity was more of a joke... as well as a small dream.
 

Offline ChunkyPastaSauce

  • Supporter
  • ****
  • Posts: 535
  • Country: 00
Re: Tektronix 2465B oscilloscope teardown
« Reply #684 on: March 05, 2016, 07:34:03 pm »
There is a few spreadsheets floating around if you need the drop in replacements for the replacement caps. Here is one that I have and slightly modified to correct for some issues that I encountered with wrong pin spacing on some of the caps.

Thanks a ton for this! Takes a lot of the work out.

I noticed are replaced with different values, is this an availability/convenience thing or an 'upgrade' thing? [edit] Excel states upgrade
« Last Edit: March 05, 2016, 07:55:18 pm by ChunkyPastaSauce »
 

Offline Old-E

  • Regular Contributor
  • *
  • Posts: 57
  • Country: us
Re: Tektronix 2465B oscilloscope teardown
« Reply #685 on: March 05, 2016, 08:49:42 pm »
Thanks to all those who published their parts lists for cap replacements.   I made hard copies of them (data not lost to an EMP) for the upcoming replacement project.

Mr. Sauce -

If I understood your question correctly regarding capacitor replacements; They can be replaced with higher voltage ratings.   Actually higher is better for longer life.   As technology advances, sometimes the same size capacitor is available in a higher voltage rating.   Also, if the original voltage is not available, the next size up is better, providing it will physically fit the board.   Sometimes higher voltage means larger physical size.

Capacity values (uf) should be closer to the original size.   In the case of filter cap's which I think most on the PS boards are, their capacity values can be increased to the next available size.   The larger capacity values will filter the DC line they serve to a higher degree, which is good.   But increasing the capacity too much, or too many higher values on the same PS line, could have an additive effect that could be a problem.

A capacitor tends to look like a short circuit to ground for the first small fraction of a second after power is applied.   Depending on the circuit design, if the capacitor is too large, it can stress the components feeding it, or even blow a fuse in an extreme case.   Like many things, sometime more is better, but beyond a certain point becomes detrimental.

For example; (without looking at the circuit) the first capacitor listed is .068 uf.   Increasing it to .1 uf, as shown, should be ok.   Increasing the voltage from 250 to 630 is great, providing the physical size will fit.

There are other places where capacity values are critical, like in timing applications.   In those cases, the value (uf) should be the same, even if you have to combine cap's to obtain the same value.   In the case of critical values, it should be called out on the documentation, or maybe have a much tighter tolerance.

You may already know much of this.   If not, hope this helps.   
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 19727
  • Country: nz
  • Taupaki Technologies Ltd. NZ Siglent Distributor
    • Taupaki Technologies Ltd.
Re: Tektronix 2465B oscilloscope teardown
« Reply #686 on: March 05, 2016, 09:01:33 pm »
There are other places where capacity values are critical, like in timing applications.   In those cases, the value (uf) should be the same, even if you have to combine cap's to obtain the same value.
You may already know much of this.   If not, hope this helps.   
One could add that timing caps are not normally under the same stresses as PSU, local bulk capacitance or decoupling caps and therefore the need for replacement is uncommon.
Sure, when there's trouble obtaining the range of adjustment needed in timing circuits (that can be adjusted) the timing caps then should be suspect, but not always.

Widespread replacement of caps should be avoided IMHO, excluding PSU's and local bulk capacitance of course.
Avid Rabid Hobbyist
 

Offline Bryan

  • Frequent Contributor
  • **
  • Posts: 602
  • Country: ca
Re: Tektronix 2465B oscilloscope teardown
« Reply #687 on: March 05, 2016, 09:18:05 pm »
Just a note, if you want to power the power supply boards outside the circuit. (recommended if you make a mistake) you HAVE to use the appropriate load resistors and they have to be rated correctly. Without the correct load resistors you will likely also notice Q1062 and R1071 getting hot. On mine R1071 actually burned up. This had me sidetracked thinking something was wrong in this area, but it wasn't and had me chasing ghosts. Probably just a bad design and the circuit started oscillating. Could also hear a faint whistling. Anyways point being is the service manual I believes states to start at the 87v and work downward (actually it's the 10v reference) so there is probably a short somewhere on one of the voltage lines and the current limiting is kicking in.

On the 87v I used a 2k resistor. In the theory of operation in the service manual. 87v rail should not exceed 100ma. -8v rail should not exceed 480ma.. +42v rail should not exceed 400ma, so I used a 150 ohm. Basically size your resistors accordingly so not to overload and have the current limiting kicking in, which will start part of the power supply to shut down making it difficult to diagnose.

I did not find it necessary to load the +10v reference.

As well as it has pointed out there is a error on the schematic for one of the capacitors and it's orientation so take a picture of your board caps before removing for references. On certain board revisions , and I am going by memory there is one more or one less bi-polar cap. I think it was one more.

« Last Edit: March 05, 2016, 09:21:12 pm by Bryan »
-=Bryan=-
 

Offline Old-E

  • Regular Contributor
  • *
  • Posts: 57
  • Country: us
Re: Tektronix 2465B oscilloscope teardown
« Reply #688 on: March 05, 2016, 11:30:01 pm »
It's all good advice for me too.

Tautech - I agree completely with your thoughts on caps.

MSO - Regarding the programmer, yes, most of those things are or have been done.   The two different IC's have the same memory capacity, pin outs and voltage ratings, etc.   One exception is the toggling of the address line, which is supposed to be compatible with the 2465B.   I had also read where that conversion had been done successfully, so proceeded ahead.   And we did successfully copy the data from the old DS1225Y and transfer it to the new FM16W08 without any snags.   And the scope does work as well as it did before changing IC's.   We also successfully compared Write & Read copies off the programmer as you suggested.   Problem I'm having now is getting the next (spare) FM16W08 to be programmed.   As a test, tried programming another DS1226AD (they are supposed to be the same as a Y version), and it was having problems too.   Have a message into the GQ Support office that I may get a response to on Monday.   Thanks for your thoughts that are always appreciated.

On a different issue - FYI
I'm set up with all the anti-static hardware for working on the scope.   The mat was in good condition, but used long ago.   So I cleaned it up and put it back in service.   But the question was - how can I test it?   Common ohm meters don't work, and I would hate to be adding ESD failures to the scope's existing issues.

Solution - I found a short thread on this Forum where others had the same concerns.   Then it occurred to me.   There's a really simple way to test it.   My write up is at the end of this thread.  https://www.eevblog.com/forum/beginners/measuring-surface-resistance-of-esd-mat/
 
 

Online med6753

  • Super Contributor
  • ***
  • Posts: 7332
  • Country: us
  • Tek nut
Re: Tektronix 2465B oscilloscope teardown
« Reply #689 on: March 06, 2016, 04:28:29 pm »
Mr. Sauce,

You wrote:  "Wouldn't the cal data be kinda useless after repairing the power supply board, it would need a recal anyway?"
I don't know of any reason the calibration would change when rebuilding the power supplies.   I counted 9 regulated voltages not counting the CRT HV.   But all of those are supposed to be at specified DC levels.   These voltages and their tolerances, including allowable ripple, are noted on page 5-3 of the "Tektronix Service Manual" and are available for measurement at J119 on the main board (bottom of scope).  Service Manual is available on line.   You may already have all this info.

I don't completely agree with this statement. Power supply issues can potentially affect calibration, especially if you have a 2465 series with options. When I acquired a 2465 DMS with the DMM and CTT (Counter, Trigger, Timer) options it had a power up fail stating that the Counter was out of calibration. I also discovered later on that the DMM was in need of calibration but the power up routines don't check that. (And that's as per the DMM options manual). If the cal data is stored in ROM how did these options get out of calibration? I can't say and I don't know the history of this 2465 DMS. I was able to get both options calibrated and operating properly but lately I noticed an issue that led me to believe that subtle power supply issues can possibly affect calibration and definitely will affect operation.

Lately I noticed that DMM and Counter were both drifting slightly. For example: against my 10.000 VDC standard the DMM would slowly drift to 9.997 VDC. Against the 1.000 VDC and 100.0 mv standard it was rock steady. The Counter over time against a 1.000 MHz signal would drift down about 50 Hz. Granted, that's not a lot but I still wondered why this was occurring. I suspected there was a power issue that I was missing and it turns out I was correct.

The DMM board takes a 5.5 VAC @ 25KHz from the Inverter board and steps it up to +/- 12VDC. Those voltages were in spec. But it also gets a +5.00 VDC digital voltage from the mainframe. That measured +4.78 VDC which is out of spec. But if I measure the +5.00 VDC at J119 on the main board it was exactly +5.00 VDC. Clearly something is not right and I found the cause. All the options (DMM, CTT, etc) get their +5.00 VDC directly from the Inverter board, not the Regulator board. So I guess it's time to pull the Inverter and Regulator board and at least re-cap them. I bought all the caps a few months ago for that day when I'd have to pull it apart so for the past 2 days I did it.

I don't care what anybody says. Pulling one of these puppies apart to get to the power supply boards, especially if you have full options, is a royal pain in the nutz.  :-- Not enjoyable to me at all. But I got it done. It appears that all the caps were original 1985 build vintage. But I didn't find any leaking or swelling. I didn't test them. I just did a mass replace. The result? That +5.00 VDC supply to the options boards now measures +4.90 VDC. Success!

I still have to go back and verify the calibration of the DMM and CTT but I'm pretty sure the drifting issue is fixed. Time will tell.

I have some pictures that I will share. I really wasn't concentrating on doing a “show and tell” teardown. I just wanted to get the damn thing done.   


Here is the section of the Inverter Board that supplies the +5.00 VDC to the Option boards at J303. From Section 9 of the 2465 Service Manual. I assume 2465A and 2465B are similar.


The DMM Board before removal


The top plate after removing the DMM Board. The CTT Board and Buffer Board are attached to the underside of the plate and pull up with it when removed.


The A5 Board on the older vintage 2465 (no suffix). No Dallas chip or battery to deal with. Uses EAROM for cal data.


The Regulator Board before re-cap. As near as I could tell all caps where original 1985 vintage.

« Last Edit: July 20, 2017, 01:11:53 am by med6753 »
An old gray beard with an attitude.
 

Offline FireDragon

  • Regular Contributor
  • *
  • Posts: 62
Re: Tektronix 2465B oscilloscope teardown
« Reply #690 on: March 08, 2016, 12:21:00 pm »
I finally got all of the parts I needed to install the probe power connectors on my 2465BCT. This is Option 11, but is not available from the factory for a BCT because there aren't enough allocated connector locations. Previously, I installed Option 1E, but is just removing a resistor and adding a capacitor on the CTT board (I also added a 50 ohm resistor for proper termination which Tektronix (intentionally) omitted). I found two locations that would work, and drilled the holes for the connectors while I had the board out of the chassis. The cable routing for the top connector wasn't a problem. For the bottom connector, I slightly enlarged an existing hole in the chassis (the hole was due to how the metal was folded).

The bottom connector would have been too close to the main board. So, I took a 1/2 OD, 3/8 ID steel bushing, and cut off appropriate lengths from each end (so that one end of the resulting standoff washer was rounded and the other flat). That also moved the top of the connectors to near the surface of the plastic.

I also had to make two small nibbles on the edge of the plastic shield. One at the top for the edge of the washer & nut for the top connector and one at the bottom for routing the bottom cable. If I had placed the top hole just a fraction of an inch (0.05, perhaps) to the left I could have avoided the nibble at the top.

Unfortunately, I did not have my drill press and the need to measure the placement of the holes in the plastic extremely accurately resulted in some raggedness of the holes in the plastic - and one drill mishap that nicked the nearby plastic. Still, not too bad.

One parts gotcha is that the maximum wire size for the LEMO type connector (actually, I used a knock-off from China - a good bit cheaper) is 26 ga stranded  or 24 ga solid. The only two wire sizes for the header were 22 ga and 24 ga. That meant that I had to use 24 ga, solid. Of all the wire I have sitting around, that was not one of the sizes. I had to make an extra DigiKey order to get 2$ of wire.

Here are the pictures for the modification.

Probe Power Parts                     
Top Probe Power Connector             
Bottom Probe Power Connector           
Top Probe Power Connector in Case     
Bottom Probe Power Connector in Case   
Top Probe Power Shield Nibble         
Bottom Probe Power Shield Nibble       
Probe Power Headers 1                 
Probe Power Headers 2                 
Probe Power Routing 1                 
Probe Power Routing 2                 
Bottom Probe Power Routing 1           
Bottom Probe Power Routing 2           


 

Offline z01z

  • Regular Contributor
  • *
  • Posts: 150
Re: Tektronix 2465B oscilloscope teardown
« Reply #691 on: March 13, 2016, 03:32:12 pm »
I've a 2465A, on this all replaced caps were still in specs but all the 180uF/40V and 250uF/20V caps started bulging plus the Kemets developed some cracks.
The backup battery was also replaced, the original still had some juice in it after 29 years!  :-+

Just for the fun I've also made a few thermal images, enjoy!
 

Offline FireDragon

  • Regular Contributor
  • *
  • Posts: 62
Re: Tektronix 2465B oscilloscope teardown
« Reply #692 on: March 14, 2016, 07:46:37 am »
Just for the fun I've also made a few thermal images, enjoy!

The thermal images are nice!

Note that the temperatures are low enough that heat sinking the U800 isn't needed. They aren't any worse that a hot summer's day down south (in the US) and better than the desert areas.

Of course, when I'm down there, I need a heat sink!

 

Offline z01z

  • Regular Contributor
  • *
  • Posts: 150
Re: Tektronix 2465B oscilloscope teardown
« Reply #693 on: March 14, 2016, 12:33:23 pm »
About the heatsink on U800, it did have one, sorry.
So I took it off, took some pictures, put it back on and took some more.

Without the heatsink the U800 definitely runs warmer.
Now, it can be argued how good the heat transfer to the heatsink is. From the close-up image it appears that the body underneath has about the same temperature as the heatsink.
Another thing is that these images show a state when the board is in the open and a fan blowing on it. When the case is on, it will surely change the whole situation and not for the better I think.
 
 

Offline Old-E

  • Regular Contributor
  • *
  • Posts: 57
  • Country: us
Re: Tektronix 2465B oscilloscope teardown
« Reply #694 on: March 14, 2016, 07:36:31 pm »
Should be remembered that the thermal images are without the case.   As pointed out here many times, the holes in the bottom of the case directly under U800 force moving air past U800.   That should make a considerable reduction in temperature, especially with a finned heat sink installed.   A "case on" temperature measurement would be an interesting test.

Thanks for the images.
 

Offline z01z

  • Regular Contributor
  • *
  • Posts: 150
Re: Tektronix 2465B oscilloscope teardown
« Reply #695 on: March 15, 2016, 06:23:26 pm »
OK, last test from me. The pictures were taken after the scope running ~30min both with and without heatsink on U800 with the case on. Then the scope was powered off, the case was quickly removed then the pictures taken.
The difference now is smaller than the scope was running open, U800 still appears to be the warmest spot.
 

Offline Old-E

  • Regular Contributor
  • *
  • Posts: 57
  • Country: us
Re: Tektronix 2465B oscilloscope teardown
« Reply #696 on: March 16, 2016, 04:35:56 am »
Nice work z01z!   If I understand correctly, the temperature with the heat sink in place, and in the case, is about the same as when it is out of the case and an external fan blowing on it.   That seems to make logical sense.   And the heat sink appears to be doing some good too.

I previously missed the part about an external fan blowing on U800 with no case.

Thanks
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 7161
  • Country: 00
  • +++ ATH1
Re: Tektronix 2465B oscilloscope teardown
« Reply #697 on: March 16, 2016, 04:49:01 am »
z01z, thanks for sharing those IR shots, really appreciate it.  :-+

Hope this will bring more awareness of blowing the whole big A1 with a big fan when working without the case.

If its not too troublesome, appreciate if you can post photos of your U800 mounted with heatsink, curious how does it look like.

Offline z01z

  • Regular Contributor
  • *
  • Posts: 150
Re: Tektronix 2465B oscilloscope teardown
« Reply #698 on: March 16, 2016, 08:14:54 am »
If I understand correctly, the temperature with the heat sink in place, and in the case, is about the same as when it is out of the case and an external fan blowing on it.
Exactly. Except that it took some time until the case was removed and the picture taken, so it must have been a bit warmer inside the case than on the pictures.
 

Offline z01z

  • Regular Contributor
  • *
  • Posts: 150
Re: Tektronix 2465B oscilloscope teardown
« Reply #699 on: March 16, 2016, 08:27:23 am »
If its not too troublesome, appreciate if you can post photos of your U800 mounted with heatsink, curious how does it look like.
Sure, no problem, I can take pictures when I get home.

I've used this heatsink, cut it to match the width and filed away some of the fins so that the nut can be fit. I've also put a small insert between the metal tab and the heatsink.
« Last Edit: March 16, 2016, 06:08:37 pm by z01z »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf