Planning 2467B LVPS re-cap
vdonisa
Hi All,
I'm planning a capacitor replacement for a 2467B low voltage power supply. The scope is working fine so nothing to troubleshoot, just a plain replacement. My understanding so far is that the LVPS is similar to the one in the 2465B so I could use the information on that one too. So far I have selected and ordered the parts. My approach was to use digikey's selector and: - select the required capacity/voltage - list the results in decreasing order of maximum ripple current (so the beefier caps would come up first) - from the first 10 or so results, pick the part with the longest life / highest rated temp I ended up with a mix of Nichicon PW, BT, GXL, CA, Rubycon LLE and Panasonic FR. They're all long life, on average 10000h / 105C, with some of them being 125 Celsius rated. They come in all colors from orange to black, so it will look quite garish :-) however I was not able to find a full set from the same manufacturer / series with consistent "best in class" specs. So I'll just live with the rainbow colors :-) I also ordered fresh equivalents of the X2 and Y2 capacitors there. I have a few questions before proceeding: - there's a guide on the net where the gentleman also replaced 2 resistors of 15 Ohm plus one of 270k with higher power ratings equivalents: http://www.bestelectronicarticles.com/testing-equipment/analogue-oscilloscopes/tektronix-2465b-oscilloscope-restoration-repair.html my question is - would this mod be of any solid benefit? - what is you take on tantalum capacitors? the gentleman above also replaced one with an aluminum equivalent. however by looking on the schematic, the function of that capacitor is "timing" and in my opinion tantalum was correctly specified by Tek. So if I'm going to replace it, I would chose either tantalum or polyester film. question - should I replace it? if yes, should it be tantalum or poly film? And the last and most important question - anything not obvious but of which I should be aware before proceeding? Thanks, Valentin VE3VDO
|
|
A comment about the article:
The two "15 ohm resistors" are NTC power thermistors used to limit the
current during power up of the AC line. They start out at about 15 ohms cold
(room temp) and when power is applied, they warm up. When that happens, the
resistance drops to about 1 ohm. This is to limit the line current when the two
bulk capacitors are first charged. It keeps from blowing the line fuse.
It is important to remember that these resistors are supposed to get hot.
That's how they work.
Regards,
Tom
|
|
stefan_trethan
As I have said before it can be detrimental to use the highest ripple/low esr capacitors where not necessary.
toggle quoted messageShow quoted text
To achieve such high performance they use a different electrolyte (water based) which entails a different failure mechanism (short). You will be fine for 10000h at the rated temperature, accordingly longer at lower temperatures, but after that the high performance caps _will_ fail. A conventional electrolyte would just dry out very slowly and lead to an open failure with high ESR as you are used to. ST
On Wed, Apr 3, 2013 at 4:39 PM, vdonisa <vdonisa@...> wrote: Hi All,
|
|
vdonisa
Hi Tom,
toggle quoted messageShow quoted text
I just checked on the schematic - there are indeed thermistors there, however looking at the group R1010 and RT1010, we have a 15 Ohms resistor (R1010) connected in parallel to a 7.5 Ohm thermistor (RT1010). Also the group R1019 (15 Ohm resistor) in parallel to RT1016 5 Ohm thermistor. Looking at the pics and text of the article I believe he was talking about R1010 and R1019 (the resistors). Thanks, Valentin
--- In TekScopes@yahoogroups.com, "Tom Miller" <tmiller11147@...> wrote:
|
|
vdonisa
Hi Stefan,
toggle quoted messageShow quoted text
I won't argue with this pov (although the whole discussion on badcaps.org is about low ESR ones, and most were seen failing in the traditional way - gradually drying out to high ESR). I have considered it too, as I searched the forum and found similar opinions. My idea is that the oscilloscope has seen some 8500 hours on it and is low serial nr - so that time was accumulated over some 20 years. If I can make the caps last another 8500 hours, and assuming that I'll give it similar usage per year, it means that those caps will be good for another 20 years and outlive me. I'm rather old you know.... Then some lucky guy will get it from the garage sale of a "ham radio estate" and re-cap it again for the next 20 years :-) Thanks, Valentin
--- In TekScopes@yahoogroups.com, Stefan Trethan <stefan_trethan@...> wrote:
|
|
Ed Breya
I'd recommend not doing anything to the scope except for looking into the archives about particular trouble spots, and an issue on the 2465B regarding some surface mount aluminum electrolytics on the display board that are definitely known to leak and damage the board (I experienced this one). I don't know if this applies to the 2467B.
toggle quoted messageShow quoted text
Regarding the example rebuild: The thing about the resistors may be valid - there are many pieces of equipment with hot spots due to marginal resistor power rating, evidenced by discolored parts and boards. The things about replacing the X and Y caps because they're slightly off tolerance, and about a certain electrolytic filter cap having to be "exactly 100 uF" are unnecessary. Most power supply parts are fairly loose in tolerance requirements - what matters most is ESR of the filter caps. A Ta cap used in a timing circuit will not be stressed as in power supply decoupling applications, so it is unlikely to fail. So, other than addressing definite known problems, I think that there's more risk in damaging something while doing extensive parts replacement on a perfectly working unit, compared to any benefit in reducing the chance of "possible" future failure. What you could do instead, is inspect it closely to look for anything suspicious like discolorations, puffiness or leakage in caps, and marginal connections. Fix what needs fixing, and save all those new spare parts for future use "just in case." Ed
--- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote:
|
|
Sorry, you are correct. I was just looking at the pictures and it appeared
he had the thermistors circled.
Still, I would be cautious modifying that area as there are thermal
considerations involved.
Back to your normal programming :>)
Regards,
Tom
|
|
vdonisa
Hi Ed,
toggle quoted messageShow quoted text
I mainly agree with you except I would replace the X2 and Y2 capacitors if they show any sign of damage. Please see here an X2 extracted from a 2235A, side by side with a new one: http://groups.yahoo.com/group/TekScopes/photos/album/1016612934/pic/42988326/view The old one works but the cracks in the case got me concerned. I've seen Y2 capacitors with the same degradation too, and it's not Tek specific (i.e. I've seen it in HP equipment too, including some recent one). Since these caps, beyond functional issues, can also cause safety problems, I'd rather replace them based on the "better safe than sorry" saying :-) Thanks, Valentin
--- In TekScopes@..., "Ed Breya" wrote: > > I'd recommend not doing anything to the scope except for looking into the archives about particular trouble spots, and an issue on the 2465B regarding some surface mount aluminum electrolytics on the display board that are definitely known to leak and damage the board (I experienced this one). I don't know if this applies to the 2467B. > > Regarding the example rebuild: The thing about the resistors may be valid - there are many pieces of equipment with hot spots due to marginal resistor power rating, evidenced by discolored parts and boards. The things about replacing the X and Y caps because they're slightly off tolerance, and about a certain electrolytic filter cap having to be "exactly 100 uF" are unnecessary. Most power supply parts are fairly loose in tolerance requirements - what matters most is ESR of the filter caps. > > A Ta cap used in a timing circuit will not be stressed as in power supply decoupling applications, so it is unlikely to fail. > > So, other than addressing definite known problems, I think that there's more risk in damaging something while doing extensive parts replacement on a perfectly working unit, compared to any benefit in reducing the chance of "possible" future failure. What you could do instead, is inspect it closely to look for anything suspicious like discolorations, puffiness or leakage in caps, and marginal connections. Fix what needs fixing, and save all those new spare parts for future use "just in case." > > Ed > > --- In TekScopes@..., "vdonisa" vdonisa@ wrote: > > > > Hi All, > > > > I'm planning a capacitor replacement for a 2467B low voltage power supply. The scope is working fine so nothing to troubleshoot, just a plain replacement. > > > > My understanding so far is that the LVPS is similar to the one in the 2465B so I could use the information on that one too. > > > > So far I have selected and ordered the parts. My approach was to use digikey's selector and: > > > > - select the required capacity/voltage > > - list the results in decreasing order of maximum ripple current (so the beefier caps would come up first) > > - from the first 10 or so results, pick the part with the longest life / highest rated temp > > > > I ended up with a mix of Nichicon PW, BT, GXL, CA, Rubycon LLE and Panasonic FR. They're all long life, on average 10000h / 105C, with some of them being 125 Celsius rated. They come in all colors from orange to black, so it will look quite garish :-) however I was not able to find a full set from the same manufacturer / series with consistent "best in class" specs. So I'll just live with the rainbow colors :-) > > > > I also ordered fresh equivalents of the X2 and Y2 capacitors there. > > > > I have a few questions before proceeding: > > > > - there's a guide on the net where the gentleman also replaced 2 resistors of 15 Ohm plus one of 270k with higher power ratings equivalents: > > > > http://www.bestelectronicarticles.com/testing-equipment/analogue-oscilloscopes/tektronix-2465b-oscilloscope-restoration-repair.html > > > > my question is - would this mod be of any solid benefit? > > > > - what is you take on tantalum capacitors? the gentleman above also replaced one with an aluminum equivalent. however by looking on the schematic, the function of that capacitor is "timing" and in my opinion tantalum was correctly specified by Tek. So if I'm going to replace it, I would chose either tantalum or polyester film. > > > > question - should I replace it? if yes, should it be tantalum or poly film? > > > > And the last and most important question - anything not obvious but of which I should be aware before proceeding? > > > > Thanks, > > Valentin VE3VDO > > >
|
|
Ed Breya
Yes, if anything looks suspicious, then investigate and fix it. I was only referring to it not being necessary to change the part because it measured slighly off-tolerance. The line X and Y caps do have a lot of stress, but they're supposedly built for it. It's also more likely to have problems with those in "high voltage" countries where the line is 240V.
toggle quoted messageShow quoted text
Ed
--- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote:
|
|
Tymm Twillman
To add to this -- on a 2465 I rehabbed, I had one of these burn after replacing the electrolytics on the low voltage board (scope was working fine after the recapping up until the acrid smell & white smoke).
toggle quoted messageShow quoted text
Luckily nothing too major -- just took a couple of minutes with IPA and some scrubbing to get the goo off the board, and after replacement things were fine. I honestly don't remember which of the X2/Y2 caps it was that went… but since then have been adding the replacement of any cracked-looking X/Y caps to the list on similar projects.
On Apr 4, 2013, at 7:35 AM, "vdonisa" <vdonisa@yahoo.com> wrote:
Hi Ed,
|
|
vdonisa
Hi All,
The re-cap is a go. I managed to open the scope this evening and get to the power supply for a visual inspection. Findings so far: - the supply has been repaired previously; there were 2 missing screws; one of the four translucent green capacitors (Nichicon PL) used by Tek was replaced with a black one of brand "Chong"; a second "Chong" was planted nearby; on the other board some of the 270u caps were replaced, at least this time it was with Nichicon; - both 68nF X2 capacitors were showing case cracks like in the pics I posted recently - no other issues noted, however half of the PS was not visible, as I did this inspection without removing it from the case so the covers were in place While I could have lived with the "Chong" thing, I believe the X2 caps should be replaced (safety issue) so I decided to remove the power supply and do the re-cap. Will keep you informed (with photos) and ask for help if I run into trouble :-) Bed time now! Cheers, Valentin
|
|
Tom Jobe <tomjobe@...>
Hi Valentin,
If you look in the message archive, there is some talk about
how two of the capacitors on the 2265/7 A3 board are shown reversed on the board
layout in the service manual for some serial numbers. I think it is C1132 and
C1115 that are reversed. The way they are on your A3 board now, is the way
they should be if there is any question. If either of those capacitors have been
changed by a previous owner, then maybe take a closer look at what you
have.
tom jobe...
|
|
vdonisa
Hi Tom,
toggle quoted messageShow quoted text
Thanks for the warning, I will definitely check everything twice before doing any replacement. BTW I just checked in my spare parts boxes and found a low profile heatsink that would fit nicely on top of the U800 IC (the unobtainium one attached with 2 screws to the mainboard). If anyone is looking for such, here's the part number: AAVID THERMALLOY 508500B00000G (digikey has it in stock, mouser lists it but no stock) it is the exact size of the plastic top of the IC. However since one of the screws eats space from that top, I can see about 2 options to install it: - cut and discard a short piece from one end of the heatsink then glue it with thermally conductive epoxy or - drill a hole at one end of the heatsink to make place for the screw, then attach it with a clip or glue Anyway, this brought me to a quick question - is it benefic to do such mod? Thanks, Valentin
--- In TekScopes@yahoogroups.com, "Tom Jobe" <tomjobe@...> wrote:
|
|
|
|
honolulusnowwhite
Hi Valentin,
toggle quoted messageShow quoted text
Considering the scope is ~25 years old and the low voltage power supply already had rework of dubious quality performed, I agree that it would be a good idea to replace all electrolytics and the AC line filtering plastic caps in the LV power supply. I suggest you remove and replace them one at a time, and carefully compare existing to new before installing the new. You may find that the actual parts needed may vary slightly from whatever list you consulted, whether it came from the Tektronix repair manual or other source. When you install the very small electrolytic caps, note that the terminals are closely spaced on the PCB and look out for a solder bridge across the terminals after the new cap is installed, using a magnifying glass if needed. Good luck, Patrick Wong AK6C
--- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote:
|
|
honolulusnowwhite
Since U800 runs quite hot, it won't *hurt* to install the heat sink. I haven't noticed any posts where someone has installed a U800 heat sink, only to have U800 fail subsequently. However that doesn't prove that the heat sink *actually* improves U800 lifetime.
If you don't install the heat sink and later suffer a U800 failure, you may kick yourself for being aware of that mod but ignoring it. If you install the heat sink and later suffer a failure, at least you took all reasonable preventive measures. Since it is not easy to get a replacement U800 and the scope will become a doorstop without it, I would install the heat sink. Patrick Wong AK6C --- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote: ... Anyway, this brought me to a quick question - is it benefic to do such mod?
|
|
vdonisa
LOL I knew I had to run into trouble....
toggle quoted messageShow quoted text
Here's a pic of the caps as they currently are on the board: http://groups.yahoo.com/group/TekScopes/photos/album/1314606778/pic/122314164/view (look for the black ones). Judging by the size of the white circles on the board, they're in correct position (the 270u ones have larger circles traced on the board). However, it it were to judge by the schematic manual, the tall black 10u/160V should be swapped with the smaller 270u to the left. Since previous repairs are involved and I can't trust anything, can someone kindly confirm that the 3 black capacitors are in the correct position in the photo? For fun the next picture in the album is one of the famous "Chong" brand capacitors I found from previous repair :-) Thanks, Valentin VE3VDO
--- In TekScopes@..., "Tom Jobe" wrote: > > Hi Valentin, > If you look in the message archive, there is some talk about how two of the capacitors on the 2265/7 A3 board are shown reversed on the board layout in the service manual for some serial numbers. I think it is C1132 and C1115 that are reversed. The way they are on your A3 board now, is the way they should be if there is any question. If either of those capacitors have been changed by a previous owner, then maybe take a closer look at what you have. > tom jobe... > > > > ----- Original Message ----- > From: vdonisa > To: TekScopes@... > Sent: Thursday, April 04, 2013 7:59 AM > Subject: [TekScopes] Re: Planning 2467B LVPS re-cap > > > > Hi Ed, > > I mainly agree with you except I would replace the X2 and Y2 capacitors if they show any sign of damage. > > Please see here an X2 extracted from a 2235A, side by side with a new one: > > http://groups.yahoo.com/group/TekScopes/photos/album/1016612934/pic/42988326/view > > The old one works but the cracks in the case got me concerned. I've seen Y2 capacitors with the same degradation too, and it's not Tek specific (i.e. I've seen it in HP equipment too, including some recent one). Since these caps, beyond functional issues, can also cause safety problems, I'd rather replace them based on the "better safe than sorry" saying :-) > > Thanks, > Valentin > > --- In TekScopes@..., "Ed Breya" edbreya@ wrote: > > > > I'd recommend not doing anything to the scope except for looking into the archives about particular trouble spots, and an issue on the 2465B regarding some surface mount aluminum electrolytics on the display board that are definitely known to leak and damage the board (I experienced this one). I don't know if this applies to the 2467B. > > > > Regarding the example rebuild: The thing about the resistors may be valid - there are many pieces of equipment with hot spots due to marginal resistor power rating, evidenced by discolored parts and boards. The things about replacing the X and Y caps because they're slightly off tolerance, and about a certain electrolytic filter cap having to be "exactly 100 uF" are unnecessary. Most power supply parts are fairly loose in tolerance requirements - what matters most is ESR of the filter caps. > > > > A Ta cap used in a timing circuit will not be stressed as in power supply decoupling applications, so it is unlikely to fail. > > > > So, other than addressing definite known problems, I think that there's more risk in damaging something while doing extensive parts replacement on a perfectly working unit, compared to any benefit in reducing the chance of "possible" future failure. What you could do instead, is inspect it closely to look for anything suspicious like discolorations, puffiness or leakage in caps, and marginal connections. Fix what needs fixing, and save all those new spare parts for future use "just in case." > > > > Ed > > > > --- In TekScopes@..., "vdonisa" vdonisa@ wrote: > > > > > > Hi All, > > > > > > I'm planning a capacitor replacement for a 2467B low voltage power supply. The scope is working fine so nothing to troubleshoot, just a plain replacement. > > > > > > My understanding so far is that the LVPS is similar to the one in the 2465B so I could use the information on that one too. > > > > > > So far I have selected and ordered the parts. My approach was to use digikey's selector and: > > > > > > - select the required capacity/voltage > > > - list the results in decreasing order of maximum ripple current (so the beefier caps would come up first) > > > - from the first 10 or so results, pick the part with the longest life / highest rated temp > > > > > > I ended up with a mix of Nichicon PW, BT, GXL, CA, Rubycon LLE and Panasonic FR. They're all long life, on average 10000h / 105C, with some of them being 125 Celsius rated. They come in all colors from orange to black, so it will look quite garish :-) however I was not able to find a full set from the same manufacturer / series with consistent "best in class" specs. So I'll just live with the rainbow colors :-) > > > > > > I also ordered fresh equivalents of the X2 and Y2 capacitors there. > > > > > > I have a few questions before proceeding: > > > > > > - there's a guide on the net where the gentleman also replaced 2 resistors of 15 Ohm plus one of 270k with higher power ratings equivalents: > > > > > > http://www.bestelectronicarticles.com/testing-equipment/analogue-oscilloscopes/tektronix-2465b-oscilloscope-restoration-repair.html > > > > > > my question is - would this mod be of any solid benefit? > > > > > > - what is you take on tantalum capacitors? the gentleman above also replaced one with an aluminum equivalent. however by looking on the schematic, the function of that capacitor is "timing" and in my opinion tantalum was correctly specified by Tek. So if I'm going to replace it, I would chose either tantalum or polyester film. > > > > > > question - should I replace it? if yes, should it be tantalum or poly film? > > > > > > And the last and most important question - anything not obvious but of which I should be aware before proceeding? > > > > > > Thanks, > > > Valentin VE3VDO > > > > > >
|
|
honolulusnowwhite
Hi Valentin,
toggle quoted messageShow quoted text
I checked my 2465A/2467 service manual. I believe your photo is of the A3 Inverter board, and the black caps pictured are: C1114 250 uF, 20V C1115 250 uF, 20V C1132 10 uF, 160V If I have correctly identified those parts, then they are correctly installed in your photo. Referring to schematic #9, C1132 filters the +87V unregulated power supply. If a 20V rated capacitor was installed there, it wouldn't last very long. The 250 uF caps filter the +5V and -5V unregulated power supplies. I found no error regarding those three caps when looking at the parts list and the schematic diagram for the 2465A/2467 manual. Patrick Wong AK6C
--- In TekScopes@yahoogroups.com, "vdonisa" <vdonisa@...> wrote:
|
|
vdonisa
Hi Patrick,
toggle quoted messageShow quoted text
Thanks a lot! Yes those are the capacitors, and the scope was working before disassembly, so there was a very high chance the the position on the board was correct. Thank you for confirming it! I have a version of the manual that shows them in the wrong position, as in the picture here: http://groups.yahoo.com/group/TekScopes/photos/album/1314606778/pic/1536262390/view Also thanks to Tom for warning me about this! Valentin VE3VDO
--- In TekScopes@..., "Patrick Wong" wrote: > > Hi Valentin, > > I checked my 2465A/2467 service manual. I believe your photo is of the A3 Inverter board, and the black caps pictured are: > > C1114 250 uF, 20V > C1115 250 uF, 20V > C1132 10 uF, 160V > > If I have correctly identified those parts, then they are correctly installed in your photo. > > Referring to schematic #9, C1132 filters the +87V unregulated power supply. If a 20V rated capacitor was installed there, it wouldn't last very long. > > The 250 uF caps filter the +5V and -5V unregulated power supplies. > > I found no error regarding those three caps when looking at the parts list and the schematic diagram for the 2465A/2467 manual. > > Patrick Wong AK6C > > --- In TekScopes@..., "vdonisa" vdonisa@ wrote: > > > > LOL I knew I had to run into trouble.... > > > > Here's a pic of the caps as they currently are on the board: > > > > http://groups.yahoo.com/group/TekScopes/photos/album/1314606778/pic/1223\ > > 14164/view > > > > 314164/view> > > > > (look for the black ones). > > > > Judging by the size of the white circles on the board, they're in > > correct position (the 270u ones have larger circles traced on the > > board). > > > > However, it it were to judge by the schematic manual, the tall black > > 10u/160V should be swapped with the smaller 270u to the left. > > > > Since previous repairs are involved and I can't trust anything, can > > someone kindly confirm that the 3 black capacitors are in the correct > > position in the photo? >
|
|
Tom Jobe <tomjobe@...>
Hi Valentin,
As a final check on this capacitor dilemma, follow the
connections from each capacitor to the next component and see how that all looks
on the power supply schematic to convince yourself that you are doing the right
thing.
On the schematic, the minus terminal of C1115 goes to ground.
Neither terminal of C1132 goes to ground. The positive terminals of those
capacitors go to different connectors, etc, etc.
It is very unlike Tektronix to have such an error in their
service manuals, so I think we all take the manual to be the final word and
distrust what the hardware is telling us. I think that sometimes the person
is fixing a dead scope and figures that they have found the reason why
it had died.
This same problem comes up every so often on Tekscopes, but
sometimes it is after the owner switches the capacitor locations to match the
service manual. I don't remember what the result was from switching the
capacitors, but it wasn't good!
tom jobe...
|
|