Phantom Power Capacitors

[john12ax7]

Interested in people's thoughts on what to use for phantom power blocking capacitors in the audio path. There are basically 3 options

1) Polar electrolytic
2) Bi-polar electrolytic
3) Film, if allowed by space and cost

I'm well aware of things like Cyril Bateman's articles. Looking more for any other measurements or listening tests people have done and specific cap suggestions to try.

Thought the Nicicon MUSE ES might be a good choice, but they only go up to 50V. In the Forssell pre, which sounds amazing, seems to be polar Nichicon "audio" electrolytics in parallel with film caps. (The Nichicon have been discontinued).

So, what caps do you use?

 

[merlin]

If the capacitance is big enough, there is no difference to hear. Could you be more specific about where in the signal path the caps really are?

 

[john12ax7]

If the capacitance is big enough, there is no difference to hear.

Theoretically yes, but I haven't found that to always be the case in practice.

Could you be more specific about where in the signal path the caps really are?

It's the usual caps that connect to the 6.8k resistors in a transformerless pre, the 47 uF in the attached generic datasheet.

 

[JohnRoberts]

Over the years I have used truckloads of 22uF aluminum electrolytic capacitors in that circuit node. Way back in the day (70s) I used some 22uF tantalums but since then very many aluminum electrolytic capacitors.

Some argue for using > 22 uF, I don't. Larger capacitors increase the magnitude of current spikes if a mic line is transiently shorted to ground, driving the DC blocking capacitor charged up to 48V to 0v, forcing its negative lead to -48V. Mic preamps can get killed by that current spike if not anticipated (mostly an issue when using console patch bays).

Over the decades I have only encountered one sonic issue related to those DC blocking capacitors. It was while working at Peavey last century when one of our capacitor vendors introduced a new series of capacitors (the cap manufacturers were always doing something, smaller, cheaper, better?). I was not involved in approving that new part, but after I got complaints from the QA inspectors in production working on my mixers, I black balled the new part. The problem went away and I never bothered to investigate why. In hindsight, it appears the caps had excessive leakage current expressing as noise. One (after the fact) theory is that the caps were not properly formed in by the manufacturer, but I repeat I never dug into exactly why they were noisy, so IDK.

[TMI] a perhaps interesting coincidence I joined this forum many years ago because of my interest in DC coupling the mic input with the input stage floated up to phantom voltage. Wayne Kirkland was working on such a design and discussed it here. I made lots of circuit scribbles on the subject but never reduced anything to practice. Wayne has made his design work. I tried to interest a friend (now RIP) into considering this for his console designs but he was not convinced that there was enough there there. /TMI]

JR

 

[john12ax7]

Is that just regular old polar 22uF or did you also try bi-polar?

 

[JohnRoberts]

Is that just regular old polar 22uF or did you also try bi-polar?

Polar, they operate with plenty of DC bias.

JR

 

[Matt Syson]

Whichever caps you use they should be of equal value (matched) to preserve LF impedance balance so the usual method of 'swamping' (using a large value for the imp/freq required is easiest. Also they should retain the equqlity for many years if you have pretentions of quality. Of course EXACT circuit details are all important as having a pad circuit after the caps will or may affect performance. life is never simple!

 

[kampys]

I personally prefer using film capacitors in the audio path for phantom power blocking. They offer better performance in terms of sound quality compared to electrolytic capacitors. However, if space and cost are constraints, I would consider using bipolar electrolytic capacitors as an alternative.

 

[Glenn Wardle]

My AKG C414 TL version (from around 1990 I guess) users 330uF so they just went for the biggest possible electrolytic they could fit in the case. With such a vast value C, any imbalance issues will be minimal even if the C values are very different. I'm just in the process of building an opamp based mic and I've gone for 270uF Aluminium polymer capacitors as this offers pretty low leakage and a miniscule ESR of less than 20mOhm. I will let you know how I get on. If the design is 'transformer-less', and you have opamps directly driving the XLR plug, then the loading into which the capacitors has to drive could be as low as 600 Ohms and this would not really be viable to drive with film capacitors IMHO.

 

[thor.zmt]

So, what caps do you use?

Capacitorless. No Capacitor is the best capacitor.

The coupling capacitors additionally create the legendary "Phantom Menace":

https://thatcorp.com/the-48-volt-phantom-menace/

https://thatcorp.com/the-48-volt-phantom-menace-returns/

Considering how trivial it is to dispose of these evil Phantom Menace Capacitors, I am staggered we don't see Capacitorless input by far more often.

Those who must be boringly conventional, use 47uF...100uF / 50V Bipolar (Nichicon ES would be my choice).

Thor

 

[thor.zmt]

Polar, they operate with plenty of DC bias.

And? Does that help with anything? Only in old wives tales.

Cyril Bateman demonstrated the exact opposite.

Polar Electrolytic capacitors have no place in high quality audio equipment, unless enclosed in a feedback loop.

Where there is no appreciable DC and Bipolar capacitors cannot be found, antiparallel polar types may be used (Samuel Groner).

Thor

 

[Glenn Wardle]

Possibly cross purposes here... Output capacitors within the mic, or input capacitors in the preamp?

 

[JohnRoberts]

And? Does that help with anything? Only in old wives tales.

Cyril Bateman demonstrated the exact opposite.

Polar Electrolytic capacitors have no place in high quality audio equipment, unless enclosed in a feedback loop.

Where there is no appreciable DC and Bipolar capacitors cannot be found, antiparallel polar types may be used (Samuel Groner).

Thor

Sam Groner did some work teasing out very tiny distorion components in electrolytic capacitor paths. IIRC he had to craft some custom instrumentation (low distortion sine wave oscillator) with unusually low residual to measure it.

Lower is always better if cost is no object.

JR

 

[My3gger]

I'm just in the process of building an opamp based mic and I've gone for 270uF Aluminium polymer capacitors as this offers pretty low leakage and a miniscule ESR of less than 20mOhm.

Do you think using such low ESR lytics are fine for apps like psu reservoirs, support circuitry in linear regulators (low and HV), decoupling and similar? I'm asking because many good quality caps like Nichicon now have low ESR, it seems most of them will have it in the near future.

 

[JohnRoberts]

Do you think using such low ESR lytics are fine for apps like psu reservoirs, support circuitry in linear regulators (low and HV), decoupling and similar? I'm asking because many good quality caps like Nichicon now have low ESR, it seems most of them will have it in the near future.

Low ESR capacitors were developed for use in switching power supplies. They are probably innocuous or a slight benefit in typical DC blocking applications. I would be careful about using them in reservoir applications as that will increase ripple/inrush current, so simple answer is it depends.

JR

 

[k brown]

My AKG C414 TL version (from around 1990 I guess) users 330uF so they just went for the biggest possible electrolytic they could fit in the case. With such a vast value C, any imbalance issues will be minimal even if the C values are very different. I'm just in the process of building an opamp based mic and I've gone for 270uF Aluminium polymer capacitors as this offers pretty low leakage and a miniscule ESR of less than 20mOhm. I will let you know how I get on. If the design is 'transformer-less', and you have opamps directly driving the XLR plug, then the loading into which the capacitors has to drive could be as low as 600 Ohms and this would not really be viable to drive with film capacitors IMHO.

Apples and oranges - the schem that was posted was for the caps on the input of a mic pre, not those inside a mic.

 

[k brown]

And? Does that help with anything? Only in old wives tales.

Cyril Bateman demonstrated the exact opposite.

Polar Electrolytic capacitors have no place in high quality audio equipment, unless enclosed in a feedback loop.

Where there is no appreciable DC and Bipolar capacitors cannot be found, antiparallel polar types may be used (Samuel Groner).

Thor

This is most interesting - and counter to everything I've read eveywhere.

 

[thor.zmt]

Do you think using such low ESR lytics are fine for apps like psu reservoirs, support circuitry in linear regulators (low and HV), decoupling and similar?

It will commonly be fine for linear Power supplies.

In Switching power supplies nowadays the switching frequencies are from multiple 100kHz into the MHz, where common electrolytic capacitors do jack.

The actual power supply capacitors are small ceramics that are good to several to several 10's of MHz and have extremely low ESR, single digit milliohms. And there usually are inductors with also milliohm DCR.

These make great resonant tank circuits in the audio range. Adding a suitable high ESR electrolytic can convert all that resonance into a critically damped filter. Use a low ESR Cap and you you need to add a series resistor for correct damping.

I'm asking because many good quality caps like Nichicon now have low ESR, it seems most of them will have it in the near future.

Most electrolytics (outside our niche) are now for switching power systems, be they off mains power supplies, solar inverters to charge batteries or inverters to produce mains voltages from batteries.

Audio parts are dying out and are getting EOL'ed rapidly. Soon we all will need to think more how to use what we can get to make what we want (a genuine equivalent copy of Neumann U47 plus a TABV72/76 on a 20 USD budget, including metalwork) much more than now.

Thor

 

[thor.zmt]

This is most interesting - and counter to everything I've read eveywhere.

What did you read? Like this:

https://linearaudio.net/sites/linearaudio.net/files/Bateman EW 01 2003 mar2003 10 to 100uF caps and 100 Hz measurements_0.pdf

That is pretty much the standard work, originally published in wireless world (UK) two decades ago.

All articles here:

https://linearaudio.net/cyril-batemans-capacitor-sound-articles

Thor

PS, now don't bl00dy get me started on resistors.

 

[k brown]

Just so I understand what's being said here, that the phantom voltage blocking caps on the input of a mic pre (where there's 48v on one side of the cap, and very few on the other) don't have to be polarized ones?