Phantom Power Question?

[matta]

Is is okay to 'hard wire' 48V Phantom into a pre-amp circuit (i.e no switch, always on)? Or is this potentially bad? I aim to use condensers 95% of the time with the pre and so 48V is required and as far as I am aware dynamics like 57/58's don't mind receiving 48V even if they don't use it?

Thanks

Matt

 

[matta]

Okay...

Having researched this further it seems like I am in no danger of damaging dynamic mics with the phantom always being left on, but are there any other reason why this might not be a good idea? The pre in question is one of my greens, for some reason a switch doesn't work and straight wiring does, and I have tried 2 different switches. I am happy to leave the straight wire in so phantom is always one, just want to know if this is safe to do.

Thanks

Matt

 

[bcarso]

Think about what the circuit looks like in various different situations.

Phantom power provides +48V relative to your common in series with a 6.81k resistor for each of your + and - lines.

If you have a transformer in your mic, the winding is across the secondary and no d.c. current flows (same d.c. voltage on both ends of the trafo). When you swing a signal the primary of the trafo looks like it has 2 times 6.81k, divided by the square of the trafo step-up ratio, across it.

So, even before you plug into your mic pre you have a certain load on the mic, usually less than what it sees with the preamp connected. There's a little bit of thermal noise in the 6.81k R's but it is mostly "shorted out" by the mic output impedance, especially at low and mid frequencies. If there is noise in your +48V supply it is common to both sides of the trafo and nulls out to a good approximation.

Now when you connect to a preamp with an input trafo, it provides a nearly ideal floating balanced load to the mic output trafo. It loads the mic some more, depending on what is on its secondary, but still no d.c. flows.

If you have a transformerless input preamp you will typically have some input capacitors followed by transistors/op amps, and resistors. How well this rejects any noise that may be in the +48V depends on how good the preamp's common-mode-rejection is. A pretty good preamp will probably be good enough unless you have very noisy +48V and way-mismatched 6.81k resistors. The latter are standard 1% values and typically are tighter than that out of the bag---if in doubt, match a pair with a good DMM. Or buy some 0.1% ones.

Numerical example to follow.

 

[Gus]

My green is wired phantom always on to keep the electros on the input formed.

 

[PRR]

> for some reason a switch doesn't work and straight wiring does

You might want to explore that issue.

 

[bcarso]

I found a green pre schematic so, subject to some guesses and when time permits, I'll figure out what the effects of various noise sources are. Since I see there is a CMRR trim they are likely small although probably frequency-dependent. I see already that there is some lowpass filtering of the +48V, -3dB at about 160Hz, so +48V high freq noise will be reduced a bunch. Does anyone know if the R's are 5% or 1%?---could be either since the standard 5% values have been available for a while in 1% tolerance.

Also if I can dig up Cyril Bateman's piece in Electronics World (UK) a while back I'll post the information. He found bipolar 'lytics used for coupling and in feedback dividers to have much lower distortion, compared to polarized ones with voltage applied, which flys in the face of traditional thinking. I don't know if he did experiments on noise level/leakage current as a function of time the voltage has been applied, but I would definitely agree with Gus that for a preamp app that's a bigger consideration, with a.c. levels so low.

Brad

 

[SSLtech]

Just a quick interjection here...

bcarso, you're an absolute gent for your contributions to the forum. -An absolute Lord.

Keith

(right...back as you were)

 

[dale116dot7]

There's one reason I can think of. If you plug in a ribbon mic into a phantom powered channel, the small current spike is multiplied a lot (by the transformer) and can cause the ribbon to stretch when it flexes in response to that current.

That being said, all but one of my pres are wired with phantom always on. But I don't have any ribbons and I left a wire loop in each chassis so it would be easy to add a switch and a LED.

 

[dale116dot7]

The two phantom resistors should be matched to as close as you can get. I normally try to match the resistors to within 0.1%. And yes, filter out the phantom really well to get the noise down. Do all of the things you would do on phantom otherwise you'll be greeted with the standard hmmmmmmmm....

 

[bcarso]

Keith---awwh, shucks

dale116, hmmm...I guess you get the little spike with a transformerless pre that is sitting there with different voltages on the two input coupling C's? Could be, although that should be pretty small unless those input C's have very disparate leakage currents, with at least one pretty big.

OTOH, suppose the first pin to mate of your XLR was one of the two signal ones, and the preamp at that point saw a big mother electrical field pickup, to which it would respond (especially at high gain) by banging around like crazy and disturbing the equilibrium conditions on the inputs. Then you contact one or both of the other pins and maybe do get your back-propagating current spike.

Kids don't try this at home dept.: has anyone ever tried to hear the emissions from a ribbon when back-driven?


Brad

 

[bcarso]

PS: I mean a microphone, not a loudspeaker...

 

[matta]

Bcarso...

I second that you are a legend, thanks for taking the time to share your knowledge with us, but to everyone who has done the same, a big thank you.

PRR, I know why the switch doesn't work... it is wired reverse from the other channel, so it is on when the switch is in the off position and on off when in the one position. I have 2 versions of the board that were Peter's Prototypes. The switch confused me so I hardwired in the phantom, but left a wire loop if I do decide to add another switch.

Gus, nice to hear you have yours wired in as well, makes me feel a little more confident. I guess Phantom Power is just that, Dynamic mics just don't see it, Condensors do.

Dale116dot7, I don't have any ribbon mics so clear there. But if I did I guess I would just cut the mains power, plug the mic in and fire up, it would be the same as flipping the phantom switch I think?

I am adding this to the Green Meta, keep it coming, great info, invaluable.

Cheers

Matt