Date   

Re: Tektronix Type 109 Pulse Generator

 

Hi David,
If I recall correctly the service manual recommended turning the instrument
off when not in use to prolong the life of the reed relay. Reed relays used
in the Telephone company were not being vibrated steadily as they are in the
109. The continuous vibration of the reed adds up pretty quickly at 720Hz.
After 400 hours of use you have vibrated that reed 1 billion times. I' don't
know at what point metal fatigue sets in but I suspect it will be before
then.

Dennis Tillman W7PF

-----Original Message-----
From: TekScopes@yahoogroups.com [mailto:TekScopes@yahoogroups.com]
Sent: Thursday, May 26, 2016 1:59 PM
To: TekScopes@yahoogroups.com
Subject: Re: [TekScopes] Re: Tektronix Type 109 Pulse Generator

For a mercury wetted reed relay though, it still seems odd that it would
ever wear out.

The reed itself is under insignificant strain because the displacement in
bending is so small and so evenly distributed. The mercury prevents arcing
from damaging the contacts. That just leaves mechanical impact at the
contact.

On 26 May 2016 11:51:59 -0700, you wrote:

Yes, they do wear out eventually. As I recall, they (and all
electromechanical relays) have lifetimes in the millions of operations at
rated load, so for general switching they can last a very long time. For
high speed like in a pulse generator running at hundreds of Hz though, the
operations count goes up pretty fast.

The wearout has two parts - the contact surfaces and the rest of the moving
elements - that stuff can only bang between the states so many times. The
load of course also affects the contact wear. Even though the Hg film makes
and breaks the continuity, the hard contact parts still hit each other and
wear out, even with no load.

------------------------------------
Posted by: David <davidwhess@gmail.com>
------------------------------------


Re: Tek 2465: readout unstable after PSU recap

 

I have done complete recaps on two 246x scopes using an Aoyue 474 vacuum de-soldering tool with n damage to any through-holes.  However, one through-hole had been damaged by leaking electrolyte.  It was simple to repair by threading a piece of 26 gauge stranded wire through to both sides and tacking it with solder.
BTW, I always check voltages and ripple before and after a re-cap.
In this case (horizontal jitter), I would check voltages and ripple, then check voltages on U800, then try the jitter adjustments.

From: "laser92awd@yahoo.com [TekScopes]" <TekScopes@yahoogroups.com>
To: TekScopes@yahoogroups.com
Sent: Sunday, May 29, 2016 12:58 PM
Subject: [TekScopes] Re: Tek 2465: readout unstable after PSU recap

  I guess I have not soldered/desoldered my scopes enough yet, but how strong are the plated through vias on the Tek boards? DId they tend to choose good PCB fabs that have very strong plated through vias that can withstand desolder cycles?

I did notice a couple blind pads (meaning, neither top or bottom layer has the pad) on the 4-layer boards. This was on my 2440, where I removed the battery - I didn't see any traces on the top or bottom layer connected to the battery pads. I hate these kind of things but I suppose as long as the plated through via holds up, it should be stronger than surface pads, but if the via is pulled out, that board is ruined and it could cause bad behavior after a recap. Then again if you got a bad egg(capacitor) in the dozen, that's also a possibility - check to make sure all your new caps are good.

Despite the issues on my 2440, I really want to avoid doing a full recap on it until I'm certain it's a bad capacitor that's causing issues. What would be best is to pinpoint a specific bad capacitor, then do a many-rats "if there's one bad cap, then look for others of the same type" kind of deal instead of swapping all of them from the get go to reduce the amount of solder/desolder cycles on the board. I am, however, tempted to swap the X2 caps as they look cracked and indeed are a sign of impending failure. Unfortunately I highly doubt they're causing the scope's operational problems.



#yiv4448377844 #yiv4448377844 -- #yiv4448377844ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv4448377844 #yiv4448377844ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv4448377844 #yiv4448377844ygrp-mkp #yiv4448377844hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv4448377844 #yiv4448377844ygrp-mkp #yiv4448377844ads {margin-bottom:10px;}#yiv4448377844 #yiv4448377844ygrp-mkp .yiv4448377844ad {padding:0 0;}#yiv4448377844 #yiv4448377844ygrp-mkp .yiv4448377844ad p {margin:0;}#yiv4448377844 #yiv4448377844ygrp-mkp .yiv4448377844ad a {color:#0000ff;text-decoration:none;}#yiv4448377844 #yiv4448377844ygrp-sponsor #yiv4448377844ygrp-lc {font-family:Arial;}#yiv4448377844 #yiv4448377844ygrp-sponsor #yiv4448377844ygrp-lc #yiv4448377844hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv4448377844 #yiv4448377844ygrp-sponsor #yiv4448377844ygrp-lc .yiv4448377844ad {margin-bottom:10px;padding:0 0;}#yiv4448377844 #yiv4448377844actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv4448377844 #yiv4448377844activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv4448377844 #yiv4448377844activity span {font-weight:700;}#yiv4448377844 #yiv4448377844activity span:first-child {text-transform:uppercase;}#yiv4448377844 #yiv4448377844activity span a {color:#5085b6;text-decoration:none;}#yiv4448377844 #yiv4448377844activity span span {color:#ff7900;}#yiv4448377844 #yiv4448377844activity span .yiv4448377844underline {text-decoration:underline;}#yiv4448377844 .yiv4448377844attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv4448377844 .yiv4448377844attach div a {text-decoration:none;}#yiv4448377844 .yiv4448377844attach img {border:none;padding-right:5px;}#yiv4448377844 .yiv4448377844attach label {display:block;margin-bottom:5px;}#yiv4448377844 .yiv4448377844attach label a {text-decoration:none;}#yiv4448377844 blockquote {margin:0 0 0 4px;}#yiv4448377844 .yiv4448377844bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv4448377844 .yiv4448377844bold a {text-decoration:none;}#yiv4448377844 dd.yiv4448377844last p a {font-family:Verdana;font-weight:700;}#yiv4448377844 dd.yiv4448377844last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv4448377844 dd.yiv4448377844last p span.yiv4448377844yshortcuts {margin-right:0;}#yiv4448377844 div.yiv4448377844attach-table div div a {text-decoration:none;}#yiv4448377844 div.yiv4448377844attach-table {width:400px;}#yiv4448377844 div.yiv4448377844file-title a, #yiv4448377844 div.yiv4448377844file-title a:active, #yiv4448377844 div.yiv4448377844file-title a:hover, #yiv4448377844 div.yiv4448377844file-title a:visited {text-decoration:none;}#yiv4448377844 div.yiv4448377844photo-title a, #yiv4448377844 div.yiv4448377844photo-title a:active, #yiv4448377844 div.yiv4448377844photo-title a:hover, #yiv4448377844 div.yiv4448377844photo-title a:visited {text-decoration:none;}#yiv4448377844 div#yiv4448377844ygrp-mlmsg #yiv4448377844ygrp-msg p a span.yiv4448377844yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv4448377844 .yiv4448377844green {color:#628c2a;}#yiv4448377844 .yiv4448377844MsoNormal {margin:0 0 0 0;}#yiv4448377844 o {font-size:0;}#yiv4448377844 #yiv4448377844photos div {float:left;width:72px;}#yiv4448377844 #yiv4448377844photos div div {border:1px solid #666666;height:62px;overflow:hidden;width:62px;}#yiv4448377844 #yiv4448377844photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv4448377844 #yiv4448377844reco-category {font-size:77%;}#yiv4448377844 #yiv4448377844reco-desc {font-size:77%;}#yiv4448377844 .yiv4448377844replbq {margin:4px;}#yiv4448377844 #yiv4448377844ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv4448377844 #yiv4448377844ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv4448377844 #yiv4448377844ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv4448377844 #yiv4448377844ygrp-mlmsg select, #yiv4448377844 input, #yiv4448377844 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv4448377844 #yiv4448377844ygrp-mlmsg pre, #yiv4448377844 code {font:115% monospace;}#yiv4448377844 #yiv4448377844ygrp-mlmsg * {line-height:1.22em;}#yiv4448377844 #yiv4448377844ygrp-mlmsg #yiv4448377844logo {padding-bottom:10px;}#yiv4448377844 #yiv4448377844ygrp-msg p a {font-family:Verdana;}#yiv4448377844 #yiv4448377844ygrp-msg p#yiv4448377844attach-count span {color:#1E66AE;font-weight:700;}#yiv4448377844 #yiv4448377844ygrp-reco #yiv4448377844reco-head {color:#ff7900;font-weight:700;}#yiv4448377844 #yiv4448377844ygrp-reco {margin-bottom:20px;padding:0px;}#yiv4448377844 #yiv4448377844ygrp-sponsor #yiv4448377844ov li a {font-size:130%;text-decoration:none;}#yiv4448377844 #yiv4448377844ygrp-sponsor #yiv4448377844ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv4448377844 #yiv4448377844ygrp-sponsor #yiv4448377844ov ul {margin:0;padding:0 0 0 8px;}#yiv4448377844 #yiv4448377844ygrp-text {font-family:Georgia;}#yiv4448377844 #yiv4448377844ygrp-text p {margin:0 0 1em 0;}#yiv4448377844 #yiv4448377844ygrp-text tt {font-size:120%;}#yiv4448377844 #yiv4448377844ygrp-vital ul li:last-child {border-right:none !important;}#yiv4448377844



[Non-text portions of this message have been removed]


Re: 2440 seems to crash when switched less than 200ns/div?

bc
 

Looking on pg 6-10 of the service manual it says the +5 (analog) should be 4.91 to 5.17V - 5.04V+/-130mV with 10mV ripple... These big triangular sawtooths are 400mV and (1/(3.5*500uS)=~600Hz) if it were repetitive! (since they're random/single shot I'm not sure how to translate this.) The fuzzy area of the trace is probably that 10mV p-p ripple and probably fine.

I didn't see these huge sawtooths in other supplies yet though, only in +5 analog. +5 digital, -5, +/- 15, +/-8 on the side board looks pretty clean.

I'm taking these when the scope is in acquire mode, the strange behavior only happens in this mode.

I tried this again with 20MHz BWL turned on my 2465: Though unrelated to the 2465, now my 2440 no longer get the warmup grace time, it goes bonkers right away when switched to 100ns/div. At 200mV/div and a x10 probe I get low frequency glitches of 0.125V (measured by cursors, they happen too fast but using phosphor persistence I moved the cursors on where I saw the trace last)... This would mean 1.25V glitches on +5V, a 25% glitch!!!

My current guess at what the problem is.... A16Q879 has issues. Or perhaps one of the sense resistors related to U870. Or something is intermittently drawing one heck of a lot of current that the LVPSU was never designed to handle despite 100ns/div and lower are legitimate modes?

What is the biggest power hogs on +5V analog? Looks like all all of the ICs on the analog board including the CCDs (though the CCD apparently is also hooked up to -5, -8, +8, and switched with other voltages)... that thing must sink quite a few amps? Or would the other rails be sinking most of the power? (Looks like the ADC is part of it, and it's even filtered off from the rest of the +5analog sinks...

-Ben

---In TekScopes@yahoogroups.com, <szigszabolcs@...> wrote :

Hi,

The service manual defines that the power ripple should be measured using 20MHz bandwidth limit, and the scope under measurement should be in Acquire Save mode. I did try, and if the scope is acquiring, you do get additional periodic ripple on some of the power rails, that look similar to yours. If you freeze the screen in save mode, this disappears.


Szabolcs


Re: 2440 seems to crash when switched less than 200ns/div?

 

I just accidentally ran across a photograph I took comparing two 2440s
of which one I returned because of what I suspected was high power
supply noise:

http://s33.postimg.org/v0rh7i33j/DSC00011.jpg

Note that the bad 2440 could trigger on the noise so averaging allowed
500uV/div on channel 1. The working 2440 could not do that because
there was nothing to trigger on.

I do not remember finding any other differences between the two but it
is likely that I did not test for problems at faster time/div settings
after noticing the low frequency noise issue.

On 28 May 2016 18:10:00 -0700, you wrote:

After some more testing, it looks like there's a reverse temperature to working condition correlation: it seems slightly better when cold and freshly powered up. This unfortunately means semiconductors? :( Now I need to get a bottle of freeze spray...

However the behavior is still the same:

- In fast sweep times, the screen tends to jitter more
- hitting 100ns/div it gets really bad. Screen can shrink to 1/2 its original size.
- Usually trace, when it's doing this, is really noisy. When the scope is cold and freshly turned on, it's not so bad. and somewhat usable though I've yet to see a trace with the sweep this fast, need to get a setup going right away when I catch it working.
- As time goes on, screen can flicker on and off. When it does flicker on and off, the input coupling sometimes suddenly thinks it's in 50 ohm termination and overloaded, and switches to GND coupling. The input attenuators click during this time.
- and eventually it screen will go out and blank. Usually the four trigger LEDs will start blinking with whatever it was triggering on

This still seems to have power implications because so many subsystems seem to be affected at the same time, but capacitors seems less likely as usually they work better when slightly warmed up...


Re: Tek 2465: readout unstable after PSU recap

 

On 29 May 2016 10:58:03 -0700, you wrote:

I guess I have not soldered/desoldered my scopes enough yet, but how strong are the plated through vias on the Tek boards? DId they tend to choose good PCB fabs that have very strong plated through vias that can withstand desolder cycles?
In my experience they are very strong but you can still damage them by
applying high temperature for too long. I prefer to do rework at 600F
instead of 700F for this reason; the lower temperature more than makes
up for the slightly longer application time.

I did notice a couple blind pads (meaning, neither top or bottom layer has the pad) on the 4-layer boards. This was on my 2440, where I removed the battery - I didn't see any traces on the top or bottom layer connected to the battery pads. I hate these kind of things but I suppose as long as the plated through via holds up, it should be stronger than surface pads, but if the via is pulled out, that board is ruined and it could cause bad behavior after a recap. Then again if you got a bad egg(capacitor) in the dozen, that's also a possibility - check to make sure all your new caps are good.

Despite the issues on my 2440, I really want to avoid doing a full recap on it until I'm certain it's a bad capacitor that's causing issues. What would be best is to pinpoint a specific bad capacitor, then do a many-rats "if there's one bad cap, then look for others of the same type" kind of deal instead of swapping all of them from the get go to reduce the amount of solder/desolder cycles on the board. I am, however, tempted to swap the X2 caps as they look cracked and indeed are a sign of impending failure. Unfortunately I highly doubt they're causing the scope's operational problems.
I tend to replace all of the roughly similar aluminum electrolytic
capacitors at the same time once one is bad. So I may change only all
of the switching power supply output capacitors and leave the
switching power supply input capacitors. In this case you have three
sets because the linear post-regulators have their own output
capacitors. The switching power supply output capacitors will be the
ones which wear out first.

You can just remove the X and Y capacitors and leave them out for now.
They are needed for EMC and not operational reasons.


Re: Tek 2465: readout unstable after PSU recap

 

On 29 May 2016 09:15:00 -0700, you wrote:

...

Honestly I would have prefered that somebody supported my horizontal readout jitter theory that simply needs readjusting, but it might be (again) more work than anticipated...
Adjust it and see if it improves the situation. Otherwise something
went wrong during the capacitor replacement procedure which would be
my first guess.

Although I don't see how recapping of the PSU might require a new calibration. I recapped many, many pieces of gear (including instrumentation stuff) and all was perfectly fine with the original calibration as far as I could tell.
I do not think it does unless it was calibrated while the capacitors
were marginal.


Re: 2440 seems to crash when switched less than 200ns/div?

Szabolcs Szigeti
 

Hi,

The service manual defines that the power ripple should be measured using 20MHz bandwidth limit, and the scope under measurement should be in Acquire Save mode. I did try, and if the scope is acquiring, you do get additional periodic ripple on some of the power rails, that look similar to yours. If you freeze the screen in save mode, this disappears.


Szabolcs


Re: 585A Power Supply Time-delay Relay Does Not Switch

Albert Otten
 

Hi Alex,

You wrote
"
I do have a parts donor (a 545A) but I'm not sure if the relay is the same. In the parts list for the 545A, it says the relay is a "DC Relay Coil, 2000 Ohm". However, in the 585A parts list, it lists "6 v 45 Sec. Delay". Furthermore, the part numbers are different. I don't think I could make the replacement.
"
You probably noted already that you mixed up K600 and K601 here.
The relay K601 in the 585A is (at least) 6 pole, in the 545A 4 pole, so not interchangeable.

Albert


Re: Tek 2465: readout unstable after PSU recap

bc
 

I guess I have not soldered/desoldered my scopes enough yet, but how strong are the plated through vias on the Tek boards? DId they tend to choose good PCB fabs that have very strong plated through vias that can withstand desolder cycles?

I did notice a couple blind pads (meaning, neither top or bottom layer has the pad) on the 4-layer boards. This was on my 2440, where I removed the battery - I didn't see any traces on the top or bottom layer connected to the battery pads. I hate these kind of things but I suppose as long as the plated through via holds up, it should be stronger than surface pads, but if the via is pulled out, that board is ruined and it could cause bad behavior after a recap. Then again if you got a bad egg(capacitor) in the dozen, that's also a possibility - check to make sure all your new caps are good.

Despite the issues on my 2440, I really want to avoid doing a full recap on it until I'm certain it's a bad capacitor that's causing issues. What would be best is to pinpoint a specific bad capacitor, then do a many-rats "if there's one bad cap, then look for others of the same type" kind of deal instead of swapping all of them from the get go to reduce the amount of solder/desolder cycles on the board. I am, however, tempted to swap the X2 caps as they look cracked and indeed are a sign of impending failure. Unfortunately I highly doubt they're causing the scope's operational problems.


Re: Tek 2465: readout unstable after PSU recap

Siggi
 

On Sun, 29 May 2016 at 12:15 ha.allm@yahoo.com [TekScopes] <
TekScopes@yahoogroups.com> wrote:

I triple checked all cap orientations before power on, as well as all
connections (cables etc.). I will check for ripple as suggested.

Honestly I would have prefered that somebody supported my horizontal
readout jitter theory that simply needs readjusting, but it might be
(again) more work than anticipated...
It's not crazy to check the ripple after a repair like this.
However, I see no possible harm in adjusting the horizontal readout jitter
trimmer - it's not going to capsize any other part of the scope's
calibration. Maybe the trimmer was jolted during repair, or maybe it
coincidentally chose this time to become tired and intermittent.


Re: 585A Power Supply Time-delay Relay Does Not Switch

Ed Breya
 

You should check the relay coil resistance to see if it's open or intact. If the coil and applied voltage are OK, it should operate. As I recall, you can remove its cover and inspect it for mechanical problems.

Ed


Re: 585A Power Supply Time-delay Relay Does Not Switch

bc
 

I guess I need to do more proofreading, and noticed another error in what I wrote, hopefully it was caught, sorry about the mistakes. The K601(relay) has a coil, not K600 (thermal cutoff, which has a heater). Need to measure the voltage across the K601 coil to see if it has the voltage to energize indicating warmup completed, and then check K600's heater to make sure it is shut off after the warmup. If K600's heater is shut off, we can assume that warmup completed, at least according to both K600 and K601.

There is also the possibility that the relay (K601) coil has a lower resistance in the 585A. If the thermal cutoff is weak it might not be strong enough to energize the relay causing the no-warmup condition.

If you can pull the relay, check the resistance between the coil pins and make sure it's not open, also compare the resistance to the working scope. I also just saw on page 7-7 that K601 should be an 18V 150 ohm relay on the 585A, not the 24V I thought... I've never seen a 18V relay before, and 150 ohms is pretty low - burning 2 watts in the relay after warming up! You may be able to swap it out for a relay with a higher ohm coil to see if it will work.



---In TekScopes@yahoogroups.com, <alex_brinister@...> wrote :

Hi Ben,

Thank you for your response. Even after about 5 minutes of warm up, the relay does not switch. The warm up time should ideally be around 30-45 seconds, as with my working scopes (545A and 585A). I measured the voltage across the relay coil of the tube (pins 4 and 9) is about 14.4 V. The voltage across pins 1 and 6 is about -2 V. I suppose I should look for a new relay.


Thank you for your help,
Alex Brinister


Re: 585A Power Supply Time-delay Relay Does Not Switch

Alex Brinister
 

Hi Ben,

Thank you for your response. Even after about 5 minutes of warm up, the relay does not switch. The warm up time should ideally be around 30-45 seconds, as with my working scopes (545A and 585A). I measured the voltage across the relay coil of the tube (pins 4 and 9) is about 14.4 V. The voltage across pins 1 and 6 is about -2 V. I suppose I should look for a new relay.


Thank you for your help,
Alex Brinister


Re: Tek 2465: readout unstable after PSU recap

Hannes Allmaier
 

Hi all,

many thanks for your input!

I triple checked all cap orientations before power on, as well as all connections (cables etc.). I will check for ripple as suggested.

Honestly I would have prefered that somebody supported my horizontal readout jitter theory that simply needs readjusting, but it might be (again) more work than anticipated...

Although I don't see how recapping of the PSU might require a new calibration. I recapped many, many pieces of gear (including instrumentation stuff) and all was perfectly fine with the original calibration as far as I could tell.

All the best, Hannes


Re: Tek 2465: readout unstable after PSU recap

Jay
 

The problem with any kind of repair or recap is that unit should be recalibrated after upgrade and the manual should mention it as my Tek 2215 does..


Re: Tek 2465: readout unstable after PSU recap

Jerry Massengale <jmassen418@...>
 

Hannes,

Murphy's law states that anything that can go wrong will go wrong and
recapping power supplies is more temptation than Murphy can stand. You can
check for ripple on a pin at J119? but you most likely has a swapped value
or a reversed cap. Make sure you have correctly restored the voltage
selection jumper.

jerry

On Sun, May 29, 2016 at 7:12 AM, ha.allm@yahoo.com [TekScopes] <
TekScopes@yahoogroups.com> wrote:



Dear all,

I just joined this excellent forum and already got me into new troubles
with my nice 2465.

The 2465 was fully fine and I decided to recap the entire PSU (high
voltage and regulator boards) as this was done before by a previous owner
using apparently low quality parts. I think this nice scope deserves
industrial quality Panasonic 105°C long life caps and this is what I used.

After putting everything back together, I noticed that the readout (both
letters and delta_v traces) is unstable and shakes back and forth
horizontally (which it did not before). Maybe the display is a tad brighter
and a slightly bit shifted upwards, but I'm not really sure about that.


My search points to readout jitter and to readjust it by R805 - I'm not
sure how a recapped psu might require that?



Is that really most likely the cause or shall I be prepared for more work?


I will try to readjust R805, but not today as one failure is enough for me
today.


Thanks for your inputs!


Hannes


[Non-text portions of this message have been removed]



[Non-text portions of this message have been removed]


Re: 475A focus problems

Artek Manuals <manuals@...>
 

On 5/29/2016 10:23 AM, damianocirri@yahoo.it [TekScopes] wrote:
Hi guys, where I can find the adjustment locations for this scope? In
the service manual are not reported...

If your manual is complete, the adjustment locations are at the end of the Schematics section

Dave ArtekManuals.com

--
Dave
Manuals@ArtekManuals.com
www.ArtekManuals.com


Re: 475A focus problems

Damiano Cirri
 

Hi guys, where I can find the adjustment locations for this scope? In the service manual are not reported...


Tek 2465: readout unstable after PSU recap

Hannes Allmaier
 

Dear all,

I just joined this excellent forum and already got me into new troubles with my nice 2465.

The 2465 was fully fine and I decided to recap the entire PSU (high voltage and regulator boards) as this was done before by a previous owner using apparently low quality parts. I think this nice scope deserves industrial quality Panasonic 105°C long life caps and this is what I used.

After putting everything back together, I noticed that the readout (both letters and delta_v traces) is unstable and shakes back and forth horizontally (which it did not before). Maybe the display is a tad brighter and a slightly bit shifted upwards, but I'm not really sure about that.


My search points to readout jitter and to readjust it by R805 - I'm not sure how a recapped psu might require that?



Is that really most likely the cause or shall I be prepared for more work?


I will try to readjust R805, but not today as one failure is enough for me today.


Thanks for your inputs!


Hannes


Re: 2440 seems to crash when switched less than 200ns/div?

 

Hello,

This is worth a recap before looking for semiconductor issues.

Paul

Le dimanche 29 mai 2016, laser92awd@yahoo.com [TekScopes] <
TekScopes@yahoogroups.com> a écrit :
After some more testing, it looks like there's a reverse temperature to
working condition correlation: it seems slightly better when cold and
freshly powered up. This unfortunately means semiconductors? :( Now I
need to get a bottle of freeze spray...

However the behavior is still the same:

- In fast sweep times, the screen tends to jitter more
- hitting 100ns/div it gets really bad. Screen can shrink to 1/2 its
original size.
- Usually trace, when it's doing this, is really noisy. When the scope
is cold and freshly turned on, it's not so bad. and somewhat usable though
I've yet to see a trace with the sweep this fast, need to get a setup going
right away when I catch it working.
- As time goes on, screen can flicker on and off. When it does flicker
on and off, the input coupling sometimes suddenly thinks it's in 50 ohm
termination and overloaded, and switches to GND coupling. The input
attenuators click during this time.
- and eventually it screen will go out and blank. Usually the four
trigger LEDs will start blinking with whatever it was triggering on

This still seems to have power implications because so many subsystems
seem to be affected at the same time, but capacitors seems less likely as
usually they work better when slightly warmed up...

[Non-text portions of this message have been removed]



------------------------------------
Posted by: laser92awd@yahoo.com
------------------------------------


------------------------------------

Yahoo Groups Links



48221 - 48240 of 176691