• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Capacitor voltage ratings: do higher voltage ratings perform differently in circuit

monkboughtlunch

Super Member
I have read a few apparently conflicting things about increased voltage ratings when recapping. Here is a summary of things I have read. Assume for the following that the capacitance value for the recap is the same as the original cap, but the voltage rating is increased.

Conflicting statements read online...
  • 1. Increased capacitor voltage ratings just means there is higher rated insulation
  • 2. Capacitance is affected if one increases voltage ratings over the original value
  • 3. Capacitors can have trouble forming if the voltage rating is dramatically increased over the original value.
Does anyone know if any of the above are correct / incorrect?
 
Register to hide this ad
In issues like this, there are theoretical issues and practical issues. I’m not a capacitor expert. I’m sure there are certain applications where changing the voltage rating has some significant effect. But for 99.9% of vintage audio applications, I believe there will be no practical effect of upping a voltage rating by one notch. In most cases, it’s probably desirable, and only possible because modern packaging generally is smaller for the same rated cap compared to 40 years ago.
 
Capacitors are not "just" capacitors. They are a complex conglomeration of things: series resistances (relating to leakage), effective series resistance, inductance, and all kinds of various internal combinations of inductance, resistance, and capacitance. There are also various characteristics relating to common physics, such as the dielectric's resistance to changes, sort of like momentum. The higher up in working voltage you go, the larger certain aspects get, physically, which increases the other non-ideal characteristics. Aluminum electrolytic caps are among the most non-ideal caps you will find. In order to get as much of what you design in, you want to select based upon as close to what you need as possible. Otherwise, you have to consider, and deal with, the non-ideal aspects of your components. If you are not able to anticipate and deal with such things, you should go with what the designers specified and used. One notch up in working voltage should not be a big problem. Just know that most of us designers have already designed in a significant amount of various kinds of de-rating.

Here is a document that shows a bit of what I am talking about: https://hal.archives-ouvertes.fr/hal-00140547/document

Rich P
 
Last edited:
Stupid question #1. Does running an amp at high volumes increase/decrease the lifespan of a capacitor? I've wondered. Since I have recently purchase a vintage amp circa 1972. I'm trying to be careful to it while using it until I can get it checked out. Can't resist listening with it.
 
Stupid question #1. Does running an amp at high volumes increase/decrease the lifespan of a capacitor? I've wondered. Since I have recently purchase a vintage amp circa 1972. I'm trying to be careful to it while using it until I can get it checked out. Can't resist listening with it.
I would take it easy with an old amp, until it has been checked out. You would not want to take an old car around the track until you have had it checked out. It is a similar thing.

Rich P
 
For the sake of this thread, are we only discussing electrolytic caps?

Yes, this thread pertains to electrolytics.

General consensus seems to be, ok to increase voltage to the next step. But hypothetically, let's a vintage capacitor value and voltage rating from 50 years ago is no longer supported in current capacitor production lines and you are forced to double or triple the voltage rating when you select a new cap. Setting aside increased size and cost of higher rated voltage caps, is there a potential negative circuit impact with respect to the voltage rating boost? Or do you just get more robust insulation with the significantly higher voltage rating?
 
Yes, this thread pertains to electrolytics.

General consensus seems to be, ok to increase voltage to the next step. But hypothetically, let's say you a cap value and voltage rating from 50 years ago is no longer supported in current capacitor production lines and you are forced to double or triple the voltage rating when you select a new cap. Is there a negative circuit impact? Or do you just get more robust insulation with the higher rating?
Once again...the designers knew what they were doing when they designed in the values they did. Unless you know what you are doing, and what effect you will be having by changing things, then don't change things. I restore gear all the time. I NEVER double or triple the voltage ratings on caps. You are almost never forced to. I never have been. Old lines are not in production. Newer lines are in production. Just go with what was originally put in there and you will be fine. Please read my previous post.

Doubling and tripling working voltage ratings can quite possibly have negative impacts upon the circuitry.

Rich P
 
Last edited:
Yes, this thread pertains to electrolytics.
But hypothetically, let's a vintage capacitor value and voltage rating from 50 years ago is no longer supported in current capacitor production lines and you are forced to double or triple the voltage rating when you select a new cap
Increasing the voltages of electrolytic capacitors by 20% or perhaps up one or two 'clicks' to the next standard voltage is quite a normal thing to do when restoring vintage electronics, and is often prompted by size issues. It is also normal to increase a voltage rating to avoid the very low voltage values sometimes seen, as capacitors with these low voltage ratings e.g. 6.3V are somewhat unreliable so an increase to 16V or 25V rated component could be beneficial. Note this is personal choice - not a forced action, and only applies to those low voltage components. However, I can't see anyone being 'forced' to double or triple the voltage rating of an electrolytic capacitor, from say 25V to 75V (or 80V) as this would be undesirable, and just not necessary.
1. Increased capacitor voltage ratings just means there is higher rated insulation
Seldom an issue.
2. Capacitance is affected if one increases voltage ratings over the original value
No. (see answer below)
3. Capacitors can have trouble forming if the voltage rating is dramatically increased over the original value.
Used to be more of an issue, new electrolytic capacitors are much better able to tolerate this now, although this is still a consideration.

Note also that when looking for replacement capacitors for vintage gear, the original might be say 50µF, so if searching now for that value the choices will be very restricted. However if you look for 47µF or 56µF, suddenly a world of possibilities reveal themselves. ;)
 
Last edited:
Yes, this thread pertains to electrolytics.

General consensus seems to be, ok to increase voltage to the next step. But hypothetically, let's a vintage capacitor value and voltage rating from 50 years ago is no longer supported in current capacitor production lines and you are forced to double or triple the voltage rating when you select a new cap. Setting aside increased size and cost of higher rated voltage caps, is there a potential negative circuit impact with respect to the voltage rating boost? Or do you just get more robust insulation with the significantly higher voltage rating?

I've seen designs where they've run caps right against their rated voltage for various reasons (availability at time of manufacturing, cost, size) modern caps tend to be smaller for an equivalent value. Also years ago (think older in this statement) many caps were Axial. Later radial became popular. Sometimes its difficult to source the right value to the right configuration in the right size and bumping up a v rating or 2 opens up some varieties.
If you can determine at what voltage a cap operates in circuit (take in account for in rush surges) .
If I can help it I try and stay close to what I started with. I know the designers were smarter than me.
 
I just read where Martin Logan uses polypropylene capacitors when used in series and DF electrolytics when in parallel in their Vojko cross-over networks. Probably doesn't really pertain to the thread, but related. David Weems claims that in certain circuits higher then necessary capacitor voltage ratings can affect the "Q" of the circuit. In most normal (power handling wise) that a 100v cap is more then enough to handle the current.
 
Really? I've seen some very dodgy "designs" in my time, and I've seen products where the "designer" clearly looked for the cheapest possible options.
Not all speakers are created equal.
That is a pretty broad brush you are painting with.

As a design engineer with several decades of experience, I want to go on record as stating that, components are most generally selected, based upon specific requirements. We are not generally micro-managed by bean counters. We do have price points, as bounding criteria, but such are not the dominant criteria. We have design requirements specifications that take into consideration things like function, quality, de-rating, reliability, component availability and sourcing, and others.

Once a piece gets into production, a new area of the company comes into play, known as "Continuing Engineering." They support the products and address issues discovered in the field, in relation to the original requirements, as well as various components becoming obsolete or NLA (no longer available). Components used need to be qualified, as well as do the sources for them. In order to do this, internal part numbers are assigned at design time, which are also associated with required specs and qualification data and procedures. Designers also create design implementation documents, to record the whys, whats, hows, etc. Continuing Engineering departments need this information to be able to properly support the designs throughout production time and subsequent continuing support time. Restoring vintage audio gear is a kind of "Continuing Engineering," with all the trimmings.

Service manuals and schematics are not design implementation documents, nor component qualification and specification documents. Frequently, much can be reverse engineered by capable/knowledgeable/experienced individuals. Changing component specifications WILL have effects upon circuitry, some changes more than others. One should know/anticipate what those effects will be before changing things.

Personally, I want nothing to do with equipment that folks have been into, changing things. Unless the changes are fully documented, including rationales for said changes. If I take in a piece, and the design has been fiddled with, it is generally gonna cost you significantly extra. I may have to reverse engineer the changes (if I can find them all...not likely), in addition to the original designs, and, most generally, reverse the changes, to return to a clean slate. I have dealt with alot of pretty strange stuff over the years. I can design, and re-design things, and do it well. However, I am generally not going to second think the original equipment designers, especially if the gear still wants to work over 40 years after its manufacture.

I will just leave this here, and quietly withdraw...

Rich P
 
Last edited:
That is a pretty broad brush you are painting with.
There are loudspeakers on the market from manufacturers ranging from big companies like JBL, KEF, B&W, Tannoy etc. right down to one-man-bands operating from their garage. Prices range from cheap-n-cheerful to eye-wateringly expensive. In my opinion, claiming that all the designers knew what they were doing and chose specific capacitor voltage ratings based on their sonic performance is painting with a broad brush.
It all depends on the speakers in question. If someone is recapping, for example, a cheap pair of speakers bought from Radio Shack in the 1970s, I would advise them not to worry too much about using capacitors with a higher voltage rating than the originals.
 
The insulation layer in modern electrolytic capacitors is much improved n dielectric strength over that in vintage electrolytics, but it is still very lossy.
 
Back
Top Bottom