Author Topic: Help in replacing a high voltage capacitor.  (Read 1980 times)

0 Members and 1 Guest are viewing this topic.

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Help in replacing a high voltage capacitor.
« on: September 16, 2021, 02:32:25 pm »
Hi:

I recently became the owner of a Keithley 246 HV precision supply. I turned it on briefly to check that the tubes and transformers worked fine, but I turned it off quickly because I found a cap that was leaking quite badly in the resistor divider board. In spite of this, the supply worked perfectly when it was turned on.

I am in the process of ordering new capacitors, and I originally wanted to replace all of them. However, in the voltage doubler part there are two big 0.5uF 3kV capacitors. Those are quite expensive (by a big margin the most expensive component I need to replace). However they both have the correct capacitance. I'm aware that simply measuring the capacitance of a cap is no guarantee, so I wanted to know if this type of capacitor (picture attached) is prone to failure or if it would be safe to leave them on.

Many thanks
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #1 on: September 16, 2021, 03:08:58 pm »
Hi:

Looking a little bit more at Mouser I found this: https://www.mouser.es/ProductDetail/WIMA/SNMPW034708H4AKS00?qs=sGAEpiMZZMsh%252B1woXyUXjycuPj0va9qfWPnmcAWL6q4%3D

Could this be an appropiate replacement? As stated previously, it is used in the HV doubler right outside the transformer.
 

Offline Manul

  • Super Contributor
  • ***
  • Posts: 1105
  • Country: lt
Re: Help in replacing a high voltage capacitor.
« Reply #2 on: September 16, 2021, 03:50:04 pm »
I would not mess with non-electrolytic capacitors without explicit knowledge that they are prone to failure. Capacitors randomly picked from mouser might be even less reliable then the old ones.
 
The following users thanked this post: wraper

Online wraper

  • Supporter
  • ****
  • Posts: 16799
  • Country: lv
Re: Help in replacing a high voltage capacitor.
« Reply #3 on: September 16, 2021, 03:55:25 pm »
As I understand some rubber sealed paper cap was leaking somewhere else. Then why would you replace high voltage hermetically sealed cap? Waste of time, effort and most likely can only possibly make things worse.
« Last Edit: September 16, 2021, 03:57:26 pm by wraper »
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Help in replacing a high voltage capacitor.
« Reply #4 on: September 16, 2021, 03:58:47 pm »
The leaking 50nF capacitor clearly needs to be replaced but, unless they are actually faulty, I don't see that the 0.5uF ones do. They are hermetically sealed and from the name (Plastic Capacitors Inc) I would say that they are plastic film rather than paper in oil and probably extremely reliable. Blanket capacitor replacement is rarely a good idea, apart from obviously faulty ones (I would replace all of the 50nF ones if one has started leaking like that).
 
As this is the beginners section, I should warn you that the large high voltage capacitors could be holding a potentially lethal amount of energy, certainly enough to scare the hell out of you. They can also recover a proportion of their charge if left disconnected (dielectric absorption) so use care when handling them.

EDIT:
As I understand some rubber sealed paper cap was leaking somewhere else.

Yes, can you clarify where the leakage is coming from, it's not clear from the photos. Is it actually from the 50nF?
« Last Edit: September 16, 2021, 04:03:12 pm by Gyro »
Best Regards, Chris
 

Offline Manul

  • Super Contributor
  • ***
  • Posts: 1105
  • Country: lt
Re: Help in replacing a high voltage capacitor.
« Reply #5 on: September 16, 2021, 04:06:51 pm »
Leaking capacitor? You mean the axial one in the first picture? Is it really leaking or just glued? There is nothing inside to leak so much.
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #6 on: September 16, 2021, 04:13:46 pm »
Hi:

The capacitor in the first image is indeed leaking.

Gyro:

Thanks for the warning regarding HV. I have experience building HV supplies in various ranges and I'm well aware of the necessary precautions. The thing is that I paid quite a bit for this supplu (well, for my amateur budget) and wanted to be very sure that the caps were good (as them failing in the doubler would take out the transformer).

Thanks all for the kind and helpful responses. The 0.5uF caps will stay in their place!
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7729
  • Country: us
Re: Help in replacing a high voltage capacitor.
« Reply #7 on: September 16, 2021, 04:19:32 pm »
The Plastic Capacitors 0.5uF/3kV units are extremely unlikely to go bad and I would never replace such types without a good reason.  As for the one that appears to be leaking, that seems quite odd.  Either something has leaked on to it (is that an insect wing that I see??) or it is not constructed of the materials that I would expect it to be. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6927
  • Country: ca
Re: Help in replacing a high voltage capacitor.
« Reply #8 on: September 16, 2021, 05:29:10 pm »
I've had "Plastic Capacitors" fail short-circuit, they are just a brand name and the dielectric warps/cracks just like any other cap after many decades.
If an arc happens inside, the heat can make it spew out molten plastic/wax etc. An arc to the outside case leaves black marks I believe.
Other component failures can stress the HV multiplier caps and then cause them to fail. I would check the resistor R101 100R has not burned up and the HV rectifier diodes D102, D103 are not shorted. If one cap fails, I'd replace the second as good practice. Using a variac to test the HV multiplier is less stress on the power transformer and less blown fuses.

Service Manual here, stumbled onto a new site for service manuals https://www.opweb.de/english/company/Keithley/246
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #9 on: September 16, 2021, 05:59:27 pm »
Wow, good eye!
I hadn't noticed it, but yes, yes it is an insect wing. However it seems like a big botch for it to be the biomass of one insect.
I disordered the capacitor and took closer photos. The end cap is noticeably softer to the
touch. The grease is viscous and sticks everywhere.
The capacity was checked and is good.
« Last Edit: September 16, 2021, 06:02:24 pm by Atomillo »
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #10 on: September 16, 2021, 06:02:07 pm »
Floodybust:

I've checked all diodes as well as resistors. I've also inspected the case visually and it is clean, with no openings or burn marks. They have the correct capacity, and when the supply was on it worked perfectly.
My worries were that it would fail silently and take out the transformer along with it... That would render the unit useless.
Due to all the advice from this thread it seems like the best course of action is to just leave them where they are, but it still makes me a bit wary.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7729
  • Country: us
Re: Help in replacing a high voltage capacitor.
« Reply #11 on: September 17, 2021, 12:22:07 am »
I've had "Plastic Capacitors" fail short-circuit, they are just a brand name and the dielectric warps/cracks just like any other cap after many decades.

I was referring to that particular type, not the brand as a whole.  I see them lurking inside of various HV supplies and have never encountered a failed one of this type.  If you have, I'll keep that in mind--I haven't seen everything!  Until a few weeks ago I would have told you that polypropylene films caps (not HV) "just don't fail", and now I'm replacing a slew of them.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6927
  • Country: ca
Re: Help in replacing a high voltage capacitor.
« Reply #12 on: September 17, 2021, 03:28:27 am »
They come in many different packages, some paper, phenolic, oil-impregnated, glass tube, bathtub can etc. The construction varies widely.
In general, common Plastic Capacitors Inc. I've seen are metallized Dupont polyester (mylar) film caps going back to the 1950's as I recall. They make a bunch of good old Frankenstein Labs stuff as well.
They must have been early in the game, for mylar HV caps, they show up in the neatest equipment.

The one's I've replaced a lot of are tubular, clear epoxy with inside paper wrap label, beige in colour. They look like their "phenolic tubular" but clear body, remind me of a Rifa PME271.

OP's "LWG" series I could not find info about. It might be oil-impregnated like the old Marcon caps? This would explain the grease that spewed out.
Why you need to oil impregnate a film cap, doesn't make sense to me.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Help in replacing a high voltage capacitor.
« Reply #13 on: September 17, 2021, 04:11:51 am »
That doesn't look like a leak to me, it looks like glue. Does it feel wet?

Those rectangular oil filled capacitors almost never fail, leave them alone. I don't know why so many people think it's a good idea to replace all of the capacitors in something, unless they're electrolytic or obviously faulty leave them alone.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7729
  • Country: us
Re: Help in replacing a high voltage capacitor.
« Reply #14 on: September 17, 2021, 04:22:46 am »
Why you need to oil impregnate a film cap, doesn't make sense to me.

They could be paper and oil, although I don't recall ever seeing that type with a 4kV rating.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #15 on: September 17, 2021, 07:16:28 am »
Hi:

I don't know if wet would be the word to use, what I know is that it sticks absolutely everywhere. It also feels very liquid (i.e not solid like dried glue). It rolled a little bit by accident in my bench and I had to clean it quite thoroughly |O

I attach photos of the boards. I have decided that in the top I will replace the two beige electrolytics just left of the supply (filtering for screen supplies) and the 20uF 350Vdc electrolytic that is connected across the output of the transformer. From the other side, I will replace both film caps (this photo was taken before I desoldered the one that is "sticky"). They are doing a bit of filtering on the output.

Following the advice given, I will not touch the 0.5uF 3kV doubling capacitors. I will also change the fuse (since in came with the current rating adequate for 125V and I need one for 220V).

I would also appreciate any idea on how to protect the precious transformers. Floodybust suggested a Variac, but unfortunately I don't have one. As stated previously, the supply was briefly turned on for about 10 minutes and everything worked fine (I tested up to 2kV and the changing of polarities). The fuse is a slow fuse (as specified in the manual), does it protect the transformer in any way?

Any thoughts?

 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Help in replacing a high voltage capacitor.
« Reply #16 on: September 17, 2021, 08:18:13 am »
Why you need to oil impregnate a film cap, doesn't make sense to me.

It prevents air voids in the capacitor winding. In high voltage capacitors any air spaces can cause corona discharge and ozone production. This leads to chemical deterioration of the film.
Best Regards, Chris
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6927
  • Country: ca
Re: Help in replacing a high voltage capacitor.
« Reply #17 on: September 17, 2021, 08:27:14 am »
If you don't have a variac, a light-bulb in series with mains can work too. Something to limit the current until the power supply circuits (+/-20VDC, +160VDC) are known to be working. The fuse is 220V 0.6A or 120V 1.25A (make sure the line voltage switch is set right). Or just have lots of spare fuses.

Other capacitors I would outright replace are the electrolytics, C106 40uF 350V, C107 40uF 350V (Sprague can type) as well as C105 2uF 25V (Sprague TL-1201 transi-lytic) by Q109. None of these parts I would expect to be good. You can lift/disconnect one leg to test them but they are probably dried out.
This is a 1971 build that has 4,600VDC to the first tube and can output 3,000V so these few parts prevent future hassles for the few dollars cost. Modern EIA values are 47uF and 2.2uF

Remember the tubes are normally full-on devices, so it wants to go full output and the regulator transistors turn off the tubes to control output voltage.

Also note the +HV is the circuit common. It is not earth ground unless the Output Polarity switch is set to (-). When Output Polarity switch is set to (+) the circuit common is connected to the Output(+) level and very dangerous if you are poking around with a multimeter. Most test leads, knobs etc. not rated for 3-5kV.
Leave Output Polarity switch is set to (-) if you make any measurements on the circuit boards. You can check with an ohmmeter if circuit common = earth/chassis ground.
See Service Manual 4-5a "Always set the OUTPUT Switch to (-) when working on the Model 246."
edit: font was wonky
« Last Edit: September 17, 2021, 06:29:57 pm by floobydust »
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Help in replacing a high voltage capacitor.
« Reply #18 on: September 17, 2021, 08:41:03 am »
Hi:

I don't know if wet would be the word to use, what I know is that it sticks absolutely everywhere. It also feels very liquid (i.e not solid like dried glue). It rolled a little bit by accident in my bench and I had to clean it quite thoroughly |O

I attach photos of the boards. I have decided that in the top I will replace the two beige electrolytics just left of the supply (filtering for screen supplies) and the 20uF 350Vdc electrolytic that is connected across the output of the transformer. From the other side, I will replace both film caps (this photo was taken before I desoldered the one that is "sticky"). They are doing a bit of filtering on the output.

Following the advice given, I will not touch the 0.5uF 3kV doubling capacitors. I will also change the fuse (since in came with the current rating adequate for 125V and I need one for 220V).

I would also appreciate any idea on how to protect the precious transformers. Floodybust suggested a Variac, but unfortunately I don't have one. As stated previously, the supply was briefly turned on for about 10 minutes and everything worked fine (I tested up to 2kV and the changing of polarities). The fuse is a slow fuse (as specified in the manual), does it protect the transformer in any way?

Any thoughts?

Yes, a properly rated fuse (very wise to reduce it to the 220V value as you have done) will provide a reasonable degree of protection against large current increase due to shorted turns. Far less likely to blow on downstream faults unless they cause a low Dielectric breakdown of the transformer secondary insulation would be more of a sudden occurrence, you just have to trust to Keithley having generously specified it. Keeping the transformer nice and dry is the best preventive measure i.e. running it occasionally. It's transformers that have sat in cold damp conditions for a long time that tend to fail soon after power-up - you seem to have passed that hurdle.

Looking at the photos, you have a few electrolytic capacitors sitting very close to the hot valves. Normally I might suggest changing them, but I can't immediately think of any modern ones that I would trust to live long in such a location. You might want to check their values though. I suppose you could move the 20uF 350V one to the other side of the PCB, but at some point you have to trust Keithley to have known what they were doing.


P.S. Ha, Two viewpoints on the same subject.  :D
« Last Edit: September 17, 2021, 08:43:25 am by Gyro »
Best Regards, Chris
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #19 on: September 17, 2021, 09:16:17 am »
Hi:

Floodybust:

I've printed the manual. The caps you mentioned are the same I mentioned from the photos, with the exception of the 25V one. I hadn't really looked at it due to it being surrounded by HV giants  :D
Regarding the light bulb, what power rating should it have? If I understood correctly, it should a little bit higher than the supply consumption so that it lights up and limits the current in case it exceeds it right?

Thanks for the warning regarding polarity. I have marked a few sentences from the manual that seems to be common mistakes (like waiting for filaments to heat up before applying high voltage, to not change polarity when its on, capacitive loads, etc...). That one is getting the special treatment of both being marked and underlined! (By the way, just rechecked now: circuit common=earth/chassis ground).


Gyro:

It is both sad to hear that the fuse will not protect the transformer and great that it seems to be fine!
Regarding moving the capacitor to the other side of the board, I don't this is a good idea because it will actually be quite close to the metal cover (the space under is quite limited).

I think your temperature concerns might be somewhat justified. Since at the time all the equipment used electron valves, I would expect capacitor manufacturers to pay special attention to temperature. On the other hand, I've heard horror stories of old cap, especially electrolytics like the ones we have here...

Yesterday I browsed through Mouser and I found this subtitutes. Knowing the temperature effect and that I didnt expect the supply to be operating continuosly for long amounts of time, are they ok?

- For the 0.05uF 4kV leaked one: https://www.mouser.es/ProductDetail/505-FKP1Y024707H00MS
- For it's brother, the 0.01uF 4kV one: https://www.mouser.es/ProductDetail/505-FKP1.01-6000-10
- For the two 40uF 350V: https://www.mouser.es/ProductDetail/661-EKXL401ELL470MJ4
- For the 20uF 350V (C103): https://www.mouser.es/ProductDetail/871-B43821F5226M000
I still have no substitute in mind for the 25V yet.

I've never replaced old caps before, but because they were in filtering duties, I chose ones with a higher capacitance as well as higher voltage rating for safety. I've also tried to stick to known brands I've heard from Dave videos (WIMA, NichiCom). I would be most grateful if you could check them over to see if the were ok.

Many thanks to all for the help received.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Help in replacing a high voltage capacitor.
« Reply #20 on: September 17, 2021, 12:29:13 pm »
The 40uF ones look like a good choice, United Chemi-con is a reputable brand (Mouser don't stock disreputable ones anyway). Nice long life spec of 10k-12k hours at 105'C is good (they are probably less stressed than they would be in an SMPS type application too).

I'm less happy about the 20uF choice though. Only 85'C rated and only 2-3k hours at that temperature (which is very likely in such very close proximity to the valves). I would look for a 105'C rated part in as large a can as possible. This is actually the more difficult part to source as small can electrolytics have less electrolyte and dry out so much more easily.

« Last Edit: September 17, 2021, 12:38:44 pm by Gyro »
Best Regards, Chris
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Help in replacing a high voltage capacitor.
« Reply #21 on: September 17, 2021, 12:39:01 pm »
P.S.  Search 22uF instead, it is a standard value (unlike 20uF) and will give you much more choice.
Best Regards, Chris
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #22 on: September 17, 2021, 01:22:16 pm »
Gyro:

Thank you very much for your help!

I found this: https://www.mouser.es/ProductDetail/United-Chemi-Con/EKXL401ELL220MJ25S?qs=sGAEpiMZZMsh%252B1woXyUXj1Qqd7OcgYbMRGoZUIAOLvM%3D
Made by Chemi Con, with maximum 105C operating temperature and 12k hours. Is this a better choice?

Also, are the WIMA capacitors a good choice? They are on the other side of the board, close to the controls, and thus I would expect temperature to be a lot more tolerable.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Help in replacing a high voltage capacitor.
« Reply #23 on: September 17, 2021, 03:15:37 pm »
Much better choice.  :-+

Yes, the WIMA ones look fine - as long as you've checked and are happy with the physical dimensions.
Best Regards, Chris
 

Offline AtomilloTopic starter

  • Regular Contributor
  • *
  • Posts: 205
  • Country: es
Re: Help in replacing a high voltage capacitor.
« Reply #24 on: September 17, 2021, 04:24:10 pm »
Thanks!
I will revise everything tomorrow and place an order with Mouser. Hopefully everything will be here by Wednesday and I can finally restore it.
Again, many thanks to all!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf