All things "C12 output capacitor"

[soapfoot]

A lot of searching and reading (and sorting through a lot of information that's old and no longer current), I decided to start a new thread on this topic. Let's see if we can organize a place to talk about this allegedly-very-important capacitor (which is, coincidentally "C12" on the schematic).

Here are some things I feel I understand:

1. Originals were Bosch or Siemens MP (metallized paper), 0,5µF @250v
2. Paper capacitors in general have relatively high risk of leakage and shorts. A short can be deadly for the output transformer
3. However, there are a lot of claims that only a MP cap "sounds right" in this application
4. In originals, form factor is an issue--keeping it to 32mm or so in length is safest (old Soviet K40Y-9 @200V will not fit)

And here are some questions:

1. Anode voltage is about 70VDC in the mic. High voltage (as delivered by the supply) is 120VDC. Is it unwise to use a 100V capacitor? What's the safest minimum?
2. Can anyone produce an actual empirical justification for the importance of the MP dielectric? Or to turn the question on its head... does anyone dispute the claim that the MP cap's behavior is an important part of the sound in a C12?
3. For those who maintain that it is very important, do paper-in-oil types work as a good substitute, or is there something about "metallized paper" specifically that's claimed to be important?
4. What about "better" dielectrics like PTFE and polypropylene-and-film? How could we expect these to deviate from expected "vintage C12" behavior? Do any feel that changing to a better dielectric is actually a net positive?
5. Has anyone found any good substitutes for the increasingly-hard-to-find Bosch and Siemens originals? Russian MBM? Sprague Difilm? Just give up and use a nice PTFE capacitor and accept the "improvement" in performance?
6. Or for that matter... has anyone found a decent source/stash for the Bosch or Siemens originals? (PM okay!)

Would love to get a gauge on what everyone is using!

 

[LevinGuitar]

In my experience I found some quality MP to sound better (pleasant to my ears) than the best MKT/MKP or even new expensive copper foil caps. The russian MBM are ok, but not the best. I liked a lot the golden Rifa and sealed square Siemens, Koweg also sound very good. Have tried unsealed axial Bosch and Siemens, but the sealed did sound better and didn't get out of specs.

 

[Khron]

We're back to the old "capacitor sound" chestnut again, are we?

 

[kingkorg]

This is probably the easiest thing to test ever. By injecting music material into the grid of the mic and do a null test of the recorded, through the mic circuit, material. Make sure the cap values are identical.

 

[soapfoot]

This is probably the easiest thing to test ever. By injecting music material into the grid of the mic and do a null test of the recorded, through the mic circuit, material. Make sure the cap values are identical.

Would probably work best to create a test mule circuit… swapping out components in the center of a C12 isn’t the most convenient

 

[soapfoot]

We're back to the old "capacitor sound" chestnut again, are we?

Not really. It’s more of a form factor and supply chain question.

But if anyone has something helpful to share re: performance (particularly empirically measurable), then that would be welcome, too

 

[kingkorg]

If you want to make some empirical measurements you can just inject signal from REW instead of music. You get to see phase shift, whole harmonic profile, FR, basically everyting. Just use aligator clips and swap caps until you pass out
It's not sensitive part of the circuit.

I think what Khron is getting at is there won't be any audible difference as long as the cap is in it's operating range, in order, and same value. He is empirically driven, just as i am. I have made the measurements, found no difference.

 

[RuudNL]

I never heard any difference! (But maybe I only used 'good' capacitors...)

 

[Matador]

I've tested out quite a few caps in the 'plate-to-transformer' position, and once you correct for the value (and know the cap is 'good'), the differences became indistinguishable (at least to my ears).

If the tube stops conducting (for any reason), the plate voltage will rise up to the supply voltage. So the minimum 'safe' voltage rating for C12 is 120V. I consider 160V rated caps to the the lowest practical voltage rating (since I've never seen a 120V rated cap that would work in this circuit). Obviously anything above that rating works (250V is far more common) that physically fits inside the body.

 

[soapfoot]

Well-advised, thanks to everyone so far.

Seems like there’s no compelling reason to use anything more esoteric than a quality polypropylene film-and-foil, then?

 

[rock soderstrom]

Seems like there’s no compelling reason to use anything more esoteric than a quality polypropylene film-and-foil, then?

I think there is one reason to use the original Bosch/ Siemens MP capacitors.

If you are restoring vintage microphones you want the same capacitors for visual reasons alone. This also has something to do with originality and value preservation, so it definitely makes sense for me here, if you can find a good working one.

For new builds, I see it more critically. I am very sceptical that the old caps sound much better than current capacitors. I must admit, I have not yet made a comparison, so I have no experience of my own in this regard.

I have had a few of these capacitors and almost all are out of specs. Remember these caps are from the fifties of the last century.

Oddly enough, this type of Bosch MPs measure much better, they are all spot on as far as their readings go; at least with the capacitors I have/had.

 

[moamps]

Well-advised, thanks to everyone so far.

Seems like there’s no compelling reason to use anything more esoteric than a quality polypropylene film-and-foil, then?

Here I have quoted the thinking of a colleague from the forum who obviously has a lot of experience with old microphones and whose opinion I highly value and mostly share.

I have no answer to the initial question, but I have compared various old "PIO" capacitors.

Bosch and Siemens "MP" are very high resolution but pretty neutral - don´t really add coloration or "mojo", if you like that expression.
Eastgerman RFT "MP" in the same dimension have equal resolution but slightly more pleasant coloration. They add some shiney midrange richness. I actually changed the Bosch and Siemens MP´s in my microphones to RFT´s. Those are still available on ebay.
The usual Russian PIOs are a bit better than good quality modern film caps but don´t come near the German ones, at least for my personal sense of aesthetics. But cheap and available.

I am taking here about application as op coupling cap in tube microphones. Results in other apps may differ.

My experience with low value Sikatrops (beige ceramic tube) from Siemens, Hydra ect. in tube mic circuits is that they are very high resolution, very low coloration. No "mojo" whatsoever. Used them for a while but eventually threw them out. Those are sometimes referred to as "PIO" as well.

I hope that´s helpful.

In life, you probably won't need more than 10 or so good coupling capacitors for your selected, special microphones, so you can certainly find that amount. If you mean to produce and sell microphones, then that's a different story. What does "famous" KH say about that topic?

 

[soapfoot]

I'm always a little leery of listening tests (with results that include words like "resolution," which have no clear definition) as sole basis for selection.

This is only because I've fooled myself plenty of times with expectation bias.

Not looking to produce microphones for sale; only making a pair for personal use. But I want them to be good.

 

[Khron]

Then define "good"..?

 

[soapfoot]

Then define "good"..?

Ultimately, a positive subjective reaction is all I'm after--I suspect many of us are the same.

In a perfect world, I'd be able to understand more about what factors most-reliably lend themselves to that subjective positive reaction.

When looking into my options for parts, I ran across a specious-seeming statement elsewhere. The exact quotation about the MP capacitors was: "They are picked for their exceptional hysteresis properties due to paper as a dielectric."

This is almost objective-sounding, but there was unfortunately no greater detail about (much less empirical supporting evidence for) the claim.

Seems like many here wouldn't lean toward validating that, which is a result.

 

[Purplenoise]

Ultimately, a positive subjective reaction is all I'm after--I suspect many of us are the same.

In a perfect world, I'd be able to understand more about what factors most-reliably lend themselves to that subjective positive reaction.

When looking into my options for parts, I ran across a specious-seeming statement elsewhere. The exact quotation about the MP capacitors was: "They are picked for their exceptional hysteresis properties due to paper as a dielectric."

This is almost objective-sounding, but there was unfortunately no greater detail about (much less empirical supporting evidence for) the claim.

Seems like many here wouldn't lean toward validating that, which is a result.

Welcome to the schizophrenic subjectivity of the arts. I don’t think there is another field that treasure and garbage could describe the same thing haha

 

[kingkorg]

Welcome to the schizophrenic subjectivity of the arts. I don’t think there is another field that treasure and garbage could describe the same thing haha

Dont forget expensive! Garbage has to be expensive in order to become a treasure. 6s6b is too inexpensive to be concerned a treasure. For me it is though.

 

[Disco Volante]

The exact quotation about the MP capacitors was: "They are picked for their exceptional hysteresis properties due to paper as a dielectric."

This is the exact kind of pseudo-scientific BS the hi-end market is full of, aka audiophoolery.
Google says:
The magnetization of ferromagnetic substances due to a varying magnetic field lags behind the field. This effect is called hysteresis, and the term is used to describe any system in whose response depends not only on its current state, but also upon its past history.
https://scienceworld.wolfram.com/physics/HysteresisEffect.html
Why would paper be better? Does paper retain charge longer than other dielectrics? What is even "better"?

Talking about the sound of specific brands of caps is IMHO waste of time. Doubleblind tests tend to show that we are all listening with our eyes and, indeed, our entire body. Put a beautiful expensive cap in there and it will be good.

Wireless World blind-tested three very different amplifiers back in the seventies. The panel was equally undecided or plain wrong in all three cases... (Though of course we are talking about microphones, which is a whooole other ballgame *wink*)
Wireless World Jul 1978

Happy tinkering;-)

 

[Purplenoise]

Dont forget expensive! Garbage has to be expensive in order to become a treasure. 6s6b is too inexpensive to be concerned a treasure. For me it is though.

Yup haha…although in some cases I am “guilty “ of hearing differences in different capacitors. Of course I don’t have the equipment or the expertise to scientifically back it up and I don’t necessarily believe that it’s because of the dielectric. Could be different esr, Vloss and other factors that play a role in that particular position…even the condition of the cap.
Also I think that any component that is changed needs a grace period in the circuit to “brake in”. Every mic that I have built sounds slightly different after a few days of use.

 

[soapfoot]

Hysteresis is possible in capacitors in the form of dielectric absorption.

I’ve never heard it alleged elsewhere that paper is superior in that regard, which is why that claim stood out to me as potentially-specious

But I don’t like reflexively dismissing claims like that out-of-hand without doing some further investigation, because quick-to-dismiss is just as much a marker of uncritical thinking as quick-to-accept.