External Phantom Power Box

[freshtapescent]

Gentlemen-

I have some vintage preamps that don't and won't have phantom power locally enabled. Thusly I'm building a 12 channel external phantom supply that will sit next to my snake, and only condensers and mics that need phantom will have to go through it. That way my tube mics, dynamics and ribbons don't have to see any superfluous circuitry. I've got a Power One linear +48 supply, and will complete the rest on perf/vero. I've poured over the net for a few days trying to figure out the best sounding but safest way to protect my mic preamps. Please have a look and call BS if you see anything you don't like. Electrolytics in the signal path will be Nichicon MUSE KZ. Switch provides a path back through 6k8 to drain caps before disconnecting mics. R4/R5 limit current through  Zeners in the case of a mic disconnecting while phantom is still on. R6/R7 drain off leakage current. Back to back zeners on both legs to save preamp input circuitry. R1/C1 Local decoupling will be on each channel to soften the snap to a thump when the global switch is thrown.

 

[moamps]

Welcome on board.
IMO, you don't need 10R resistors, the inrush current thru Zeners is already limited  with 6.8k resistors.
Everything else looks fine to me.

 

[JohnRoberts]

Welcome on board.
IMO, you don't need 10R resistors, the inrush current thru Zeners is already limited  with 6.8k resistors.
Everything else looks fine to me.

The current limiting is for the 100uF caps that could dump amps if the + side gets shorted to ground, after being charged to 48V (like by a patch bay)

100uF looks a little large to me, perhaps a more is better exercise. (I've used 22uF for many preamps).

Pretty much every resistor but the 100ohm needs to be 1% or better

It wouldn't hurt to select the caps to match each other as well as possible from parts available.

JR

 

[Audio1Man]

Be careful about your type of  R1. No SMD unless it is >1206 as the mass of the part is to small and the inrush power will destroy the resistor.  Axial 1/4 watt has a big thermal mass and the peak inrush power will not open the resistor.
Duke

 

[freshtapescent]

The current limiting is for the 100uF caps that could dump amps if the + side gets shorted to ground, after being charged to 48V (like by a patch bay)

I'm a lot less concerned about a short on either leg of the +48 side of the caps and they wont be going through a patchbay. However, I can pretty much guarantee a mic will be accidentally connected or disconnected with phantom still on...do I still need those 10R in this case? Or the Zeners? I don't want to take a hit to CMR and noise if I don't have to.

 

[abbey road d enfer]

100uF looks a little large to me, perhaps a more is better exercise. (I've used 22uF for many preamps).

That's what I thought, until the day I had to build a similar box, that connected with a broadcast mixer with input transformers. 5dB bump @16Hz ! Increasing to 100uF put it 1dB @10 Hz.

 

[abbey road d enfer]

I'm a lot less concerned about a short on either leg of the +48 side of the caps and they wont be going through a patchbay.

You never had a shorted cable? 

However, I can pretty much guarantee a mic will be accidentally connected or disconnected with phantom still on...do I still need those 10R in this case? Or the Zeners? I don't want to take a hit to CMR and noise if I don't have to.

As always, that depends on your equipment. Most of commercial equipment will have these protections built-in. Are you a belt-and-suspenders type, or can you afford a little risk?
One thing is that very often protection devices, once they have done their duty, are not intact, the resistors may have drift, the zeners may be leaky, and you don't know you have a problem; you figure it out later.

 

[JohnRoberts]

100uF looks a little large to me, perhaps a more is better exercise. (I've used 22uF for many preamps).

That's what I thought, until the day I had to build a similar box, that connected with a broadcast mixer with input transformers. 5dB bump @16Hz ! Increasing to 100uF put it 1dB @10 Hz.

If you are adding phantom in front of real transformers you don't need capacitors at all.  ;D

I was talking about in front of solid state preamps... I stopped making transformer mic preamps in the '70s.

JR

{edit] I meant to say I stopped using transformers for "premium" audio paths. I used truck loads of cheap chinese transformers in fixed install (background music) mic preamps. They still cling to their transformers and resist change in that market. FWIW the transformers I used were about the size of my pinky, so not great LF performance. Who knows the phools might like the extra distortion..(just kidding not serious do not try them). [/edit]

 

[abbey road d enfer]

100uF looks a little large to me, perhaps a more is better exercise. (I've used 22uF for many preamps).

That's what I thought, until the day I had to build a similar box, that connected with a broadcast mixer with input transformers. 5dB bump @16Hz ! Increasing to 100uF put it 1dB @10 Hz.

If you are adding phantom in front of real transformers you don't need capacitors at all.  ;D

I was talking about in front of solid state preamps... I stopped making transformer mic preamps in the '70s.

JR

Broadcast people want transformers (at least they did in the 90's); as long as they're ready to pay... :

 

[moamps]

.....I have some vintage preamps that don't and won't have phantom power locally enabled....

Also you can implement a slow ramp-on phantom power.

 

[mrclunk]

Is there any advantage to using bi-polar electrolytics in this circuit?

We have phantom power built into our wall boxes, its very handy.

 

[freshtapescent]

You never had a shorted cable?  ..... Are you a belt-and-suspenders type, or can you afford a little risk?

I've seen a shorted cable, but IMO they're very rare. What I was getting to is that the failure mode for a short (100uF dumps amps and hopefully the protection zeners save my inputs...and it probably happens once and the zeners are cooked = BAD) is different then just unplugging a mic with the phantom on. In this case, what is the failure mode? Caps stay charged...but don't dump amps, so zeners don't do anything and input circuitry is fine. Human says "oops" and flips the phantom switch off, draining caps back through 6.8k resistors to ground and in less than a second, and everyone goes on with their lives. The zeners and 100R are ONLY useful in a short-circuit scenario? To answer your question...neither a belt-and-suspenders nor a risk taker...but It would be nice to keep sh*t out of the signal path if its only there for the once in several years where not only has a solder joint in my cable has opened up, but has also managed to short itself out, and I was using phantom power that day. 

Is there any advantage to using bi-polar electrolytics in this circuit?

AFAIK, no. I've got 48V biasing those capacitors, and mic signals riding along on the order of less than 2V at max, so they won't ever be reverse-biased. Also, when phantom is not needed, this circuit will be 100% out of the path, so I don't have to worry about the use-case of this circuit in use with phantom OFF. (FWIW, this is why I'm not building this circuit into all of my old preamps and building it externally.) The hit would be double the ESR in a place where noise matters most.

Also you can implement a slow ramp-on phantom power.

Yeah, I've thought about it. People seem split as to whether it is necessary or cosmetic.  I do have some preamps with input transformers with really high turns ratio...so I read they're much happier without a couple kV shot across the secondary. I wish it worked with a RC network and loooong time constant, but it doesn't. It involves transistors and then I'm instantly in over my head... I did see this attached photo floating around. Can anyone verify it?

 

[Audio1Man]

I don't think I would use this circuit as it has about 2.5 volts drop and it will make a screaming hf oscillator. It could take out TV and the FM stations. The base of the MPSA06 needs a few hundred ohms to DQ the oscillator mode.

Duke