How Often to Re-Cap?

[0dBVU]

Hey guys, I just recently had an old Boss CE-2 opened up and I noticed that there is a polarized cap on the board that appears to have exploded. At what point would you guys consider it reasonable maintenance to go in a re-cap a pedal... a certain number of years? or would you wait for a failure to do it?

Also, should one replace all the film caps as well as the electrolytic caps in there?

Thanks for any advice!

[R.G.]

If you have it open, and one of the electros in there has already failed, I think it would be a good time to replace them all.

Ten years is a pretty good lifetime for an electro. Any time I have to open up any of my equipment that's more than ten years old, I'll probably recap it.

Obviously, electros can last for twenty, thirty, or even fifty years - but why wait until THEY decide you're gonna do the job?

Film and ceramic caps never need replacing. They do not have the built in wearout mechanism that electros do.

[0dBVU]

Thank you very much, that info is very helpful!

[formerMember1]

if you don't mind i would like to add a question, instead of a new thread,...

Could i just use non polar electros, so that if i ever change the polarity(say in a fuzzface) to try different transistors, or to make the pedal a different fuzzface, i would only need to swap the battery terminals, and not worry about polarized caps,..?

Also,

Mouser has a electrolytic capacitor series that says it never needs replacement, has the lowest resistance of any electros,  is this true or another gimmick,...They are called ELNA double layer capacitors: it says:
The Dz series is an entirely new type of high current capacitor that has extremely low internal resistance(only 10% of standard double layer caps), capable of undergoing virtually UNLIMITED charge/discharge cycles with no deterioration.  With capacitance ranging from 1F to 100F and internal resistance from <.08 to 1.0ohms, they are ideal for CMOS backup, or they can function as a simple power supply.

One draw back i read is that it says it could have -20% to +80% tolerance!!!

they are 1.0uf at $2.75 each ~~~MOUSER #555-DZ2R5D105
they are 3.3uf at $3.65 each ~~~MOUSER # 555-DZ2R5D335
they are 4.7uf at $4.25 each ~~~MOUSER #555-DZ2R5D475
they are 10uf at  $6.10 each ~~~MOUSER # 555-DZ2R5D106

the price would be ok, since you don't have to ever replace it, worry about replacing, but the tolerance things is wierd, i didn't know a cap could have 80% tolerance!!!

thanks 

[R.G.]

Could i just use non polar electros, so that if i ever change the polarity(say in a fuzzface) to try different transistors, or to make the pedal a different fuzzface, i would only need to swap the battery terminals, and not worry about polarized caps,..?

Yes, you could.

Note that NP electrolytics still wear out at the same rate that normal ones do, so NP is not a way around electrolytic rot.

Mouser has a electrolytic capacitor series that says it never needs replacement, has the lowest resistance of any electros,  is this true or another gimmick,

Could be true. The wear-out mechanism for normal electros is gradual thinning of the insulating oxide layer. Apparently the double layer thing is based on some surface interation with activated carbon granules, which have huge surface areas.

The ELNA site says not to use them for power supply filtering because the large internal resistance would cause internal damage from heating.

At the prices, I'm not going to start replacing wholesale yet. I'll give myself one more decade of $0.15 electros and let the double layer stuff get the kinks worked out. Maybe by then the prices will be down near electros.

[redeffect]

i'd first check out WHY did this cap "blow up". they dont often do this
red

[formerMember1]

Thanks R.G. 

That makes a lot of sense, thanks for clearing that up.   

[darkseid]

I have been thinking about recaping an old tube amp I have..... But I having a hard time finding Axial Electrolytic 40uF at 450 Volts.

I'm guessing those are old diodes close to the caps,  should I replace those?


What kind of cap is this.... it is 40uF Axial Electrolytic ....  why does it look different from the others?

high voltage caps are a little foreign looking to me...

[petemoore]

  One draw back i read is that it says it could have -20% to +80% tolerance!!!
  For power supply filtering, as long as it's value is big enough to suit the purpose I don't see it would matter if it's spot on.
  In a gain setting E bypass situation, closer tolerances or finding one that 'fits' well might be a good idea, since they supposedly don't deteriorate, I ask is the tolerance referring to 'one cap to the next' or 'this cap will stray depending on temperature or something'. If it's one cap to the next that the tolerance is related to, once you find the right value from your batch it might exhibit improved ability to 'stay at that value over a longer period of time?
   "BP" [Bipolar] 'NP' [non polarized] are the same thing and the same as film caps as far as orientation doesn't matter. They don't care which side is most positive.
   
 

[ErikMiller]

I have been thinking about recaping an old tube amp I have..... But I having a hard time finding Axial Electrolytic 40uF at 450 Volts.

You should be fine using 47uF/450V caps for this application.

[JimRayden]

I have been thinking about recaping an old tube amp I have..... But I having a hard time finding Axial Electrolytic 40uF at 450 Volts.

You should be fine using 47uF/450V caps for this application.

Or any other higher value as long as the caps fit. If you can't find any, try 33uF and see if the hum is too annoying. If not, keep 'em in. If it's bad, put the old ones back or search for higher ones.

The "wierd" cap looks wierd because it's covered in heat-shrink tubing. Underneath is a casual aluminium-covered 40uF cap. Of which voltage is not seen in the picture. You should trace a schematic to see which caps can be replaced within what tolerances which way (either smaller or bigger).

-----------
Jimbo

[R.G.]

Electrolytic caps are a study in the art of how to get enough.

The only reason electrolytic caps are commercially available at all is that they were the only way (until recently) to get capacitors bigger than a few microfarads in any reasonable size. The techniques for getting enough capacitance involve taking vanishingly thin aluminum foils then chemically etching them to get even more suface area, and then growing an oxide layer on the surface to the desired thickness to resist voltages. If the makers then had to select them for tight tolerances, the prices would be sky-high.

But the use of big capacitors is usually in places where you want a big cap and will buy the biggest capacitor you can get for your money. Power supplies are a great example. It's only rarely that you want a big capacitor that is also accurate in value. It's much more economical to do accurate stuff with small, low power capacitors then amplify it up. So the industry never developed techniques for getting big, accurate value electrolytic capacitors. The standard tolerance for electros was always +80%/-20%. I notice that some electros in smaller values have gotten to +/-10%, but big ones are always larger tolerance.

Tantalums are more accurate tolerances.

[amonte]

So when people talk about doing a cap job on an old tube amp (say 60's or 70's), only the power caps/electros need to be replaced?  I've got a Traynor YGM (70's) and a Kalamazoo Model 1 (60's) that I've been thinking about recapping for a while, but haven't gotten around to it yet because I thought I had to replace ALL of the caps.

[R.G.]

Only electrolytic (polarized, with "+" and/or "-" markings) need to be replaced. This is usually only in the power supply and on the cathodes of the triode/tubes.

[Mark Hammer]

To follow up, MOST of those will be in the power supply section, but some will be in the preamp section, usually in parallel with smallish value (1k-2.7k) resistors to ground.

One of the things that Gerald Weber (Kendrick Amps) often recommends in his monthly column in Vintage Guitar is to put two caps in series.  If I have understood him correclty, while this results in a lower capacitance, it doubles the voltage rating.  So, two 100uf/250v caps in series would function like a 50uf/500v cap.  He offers this as a way of achieving working voltage ratings that provide a wider margin of safety.  So, if your amp wants, say, 435V to the plates, and 450v rating is cutting it a little close, you can "arrange" to have 500v or 600v rated caps.

Can someone confirm or deny?  Also, does this work ONLY if they are configures back to back like a non-polarized unit, or does it also work if they are simply end to end (i.e., + -, + -)?

[JimRayden]

To follow up, MOST of those will be in the power supply section, but some will be in the preamp section, usually in parallel with smallish value (1k-2.7k) resistors to ground.

One of the things that Gerald Weber (Kendrick Amps) often recommends in his monthly column in Vintage Guitar is to put two caps in series.  If I have understood him correclty, while this results in a lower capacitance, it doubles the voltage rating.  So, two 100uf/250v caps in series would function like a 50uf/500v cap.  He offers this as a way of achieving working voltage ratings that provide a wider margin of safety.  So, if your amp wants, say, 435V to the plates, and 450v rating is cutting it a little close, you can "arrange" to have 500v or 600v rated caps.

Can someone confirm or deny?  Also, does this work ONLY if they are configures back to back like a non-polarized unit, or does it also work if they are simply end to end (i.e., + -, + -)?

I have used this method in my tube preamp and nothing has blown up yet. I connected them +-+-, if I remember correctly.

-------------
Jimbo

[R.G.]

To follow up, MOST of those will be in the power supply section, but some will be in the preamp section, usually in parallel with smallish value (1k-2.7k) resistors to ground.

Correct. Those are the ones I mentioned at the cathodes of the tubes. Physically, of course, they are paralleled with the cathode resistors. Thanks Mark!

Can someone confirm or deny?  Also, does this work ONLY if they are configures back to back like a non-polarized unit, or does it also work if they are simply end to end (i.e., + -, + -)?

I confirm it. You can series connect electros for higher voltages. For higher voltages, connect all the positive ends the same direction - i.e., + to the - of the next-higher cap. You'll wind up with one + and one - terminal to go across the DC supply.

Unless you're good at dinking with measuring leakages, only do this with two (or more!) identical caps, and parallel each cap with a resistor chosen to swamp the cap's leakage and equalize the DC voltages. I like to use about 220K resistors. Those work for up to about 300V. If you don't use the parallel swamping resistor, the  DC is distributed by the leakage currents of the caps, and one of them will get more voltage because it's lower leakage; that one may puncture.

Electrolytic caps actually conduct well when you reverse polarize them. Connecting  a cap + to +  or  - to - ensures that at least one of the caps will be correctly polarized no matter what the applied voltage. Electros hooked up this way do NOT change value by the series cap formula, so this looks like one capacitor of the indicated value except for small voltages around 0.

[troubledtom]

i've had great luck w/ old electro-caps. all my stuff is stored in the lab or studio. my 72 marshall will never be touched till
   i hear some trouble from it. but i do have a sunn beta lead amp that needs the works done to it.
       pedals?,
          if it's home made and tweaked the you like it and it can be made so again.....10-12 years
   now vintage pedals only when you have to ,if youever have one you plan to sell at sometime.
           but old or new , the pedals i use live, will always be ready for mods or retrofits. [sp?]
                       long day...........
                             - tt
             

[darkseid]

Thanks for the info, this helps alot for me.........    The reason I think I need to replace those caps on my amp is that the amp not being as loud as it was like it was pushing a few mili watts, plus that it was made in the 80's and has never been recaped.....  I checked all the in and out connections....  I replaced the tubes and it did not do the trick, also when I discharge the caps there seem to be little spark.   So hopefully it is just the caps...

[petemoore]

  I'm guessing those are old diodes close to the caps,  should I replace those?
  I found a batch of fried diodes in an EQ, right next to fried off wires and traces. Fairly new unit.
  DMM Diode checker will tell all. I'm not sure if I'm correct about my assement, as it was based on getting no response at all from the DMM when checking them in circuit...pulling one end will tell more.
  Tube amp is like Car. It may only need 1 part to be 'fixed', however it may be less expensive to buy all the parts for it than it is to have it diagnosed down to that one faulty part.
  Cars are more expensive, tube amps aren't exactly cheep...but, if you want the 'racing machine' [dependable etc.] it's probably worth it to just do a fairly complete overhaul as most of it is there and good.
  I see what looks like eyelet construction so it's almost certain to be worth your while to do so.
  I would read lots on tube amps, like I'm trying to do now, start maybe with getting your B+ voltage figured out first...
  Since this is OT for stompbox, I point out that B+ in tube amps is potentially lethal voltage, get your sure fire cap drain and 0v check method down...