Transformers and Phantom Power

[sr1200]

Just want to see if I can get something clarified, since I'm not 100% sure I understand this. Here's how I understand it:
If we run DC through a transformer that has an end tied to ground we get current flowing through one side of the transformer. This is a magnetization current and results in the core heating and over time MAY cause issues. With phantom power, since we are applying (hopefully) an equal 48v DC to both legs of the winding, there is no current flowing through the coil, therefore no magnetization current and no heating. This is assuming everything is working properly and we dont have a bad cable with a pin connected to ground etc.

Please correct what is wrong with this statement if it is wrong. I'm trying to research some things regarding phantom power applied to certain dynamic mics, including ribbon mics, and haven't been able to find a straight forward answer as to what is happening at the transformer in the real world.

 

[Potato Cakes]

What are you trying to accomplish with your research? Are you trying to determine how to power a circuit with phantom power that uses an output transformer?

Thanks!

Paul

 

[moamps]

Just want to see if I can get something clarified, since I'm not 100% sure I understand this. Here's how I understand it:
If we run DC through a transformer that has an end tied to ground we get current flowing through one side of the transformer. This is a magnetization current and results in the core heating and over time MAY cause issues. With phantom power, since we are applying (hopefully) an equal 48v DC to both legs of the winding, there is no current flowing through the coil, therefore no magnetization current and no heating. This is assuming everything is working properly and we dont have a bad cable with a pin connected to ground etc.

Please correct what is wrong with this statement if it is wrong. I'm trying to research some things regarding phantom power applied to certain dynamic mics, including ribbon mics, and haven't been able to find a straight forward answer as to what is happening at the transformer in the real world.

Your thoughts are correct. Any properly designed dynamic or ribbon microphone connected with the correct cable must survive phantom power. If that is not true, then we would have to call that power supply by some other name. But there is a problem.
If we make a microphone patch using a TT (bantam) patch bay, in one short moment of switching on or off the connector, a short circuit of one active pin to ground occurs. Then the second active pin is loaded with a full voltage of 48V towards ground and that moment is dangerous for dynamic but more so for ribbon microphones. That is why reconnection with active phantom power is never recommended.

 

[sr1200]

What are you trying to accomplish with your research? Are you trying to determine how to power a circuit with phantom power that uses an output transformer?

Thanks!

Paul

Im more looking into the truths and myths behind some information regarding phantom power that is put forth by "experts" and manufacturers. One manufacturer claims that having phantom power on a ribbon mic will eventually burn out the transformer. From what i can tell, that would only happen if you had one side of the transformer going to ground and even then... burning out the transformer seems a little... off.
Other manufacturers say, not a problem as long as everything is working correctly. (which is what i think is more the case)
There are some folks that claim to be experts on the subject that say having phantom power on the line with ribbon or dynamic mics degrade their performance... not sure HOW that is happening if everything is working correctly.
But perhaps I'm being mislead by the idea of a perfect environment. On paper, yeah, everything "SHOULD" be ok, but what happens in the real world.

I've setup some experiments i plan on using. 3D printed a ribbon mic which, to my disappointment, actually doesn't sound half bad. Also 3D printed a dynamic mic and used a cheapo $4 chinese dynamic capsule. I plan on setting up some scenarios to test various working conditions. I have a power supply setup with 48v to try some of the hot patch scenarios so I dont wind up damaging any of my mic pres (thats another claim that is made... the preamp can be damaged by using phantom with certain mics).

I just want to get to the bottom of the reality of the situation and be able to say - "This is what happens in these situations".

 

[sr1200]

Your thoughts are correct. Any properly designed dynamic or ribbon microphone connected with the correct cable must survive phantom power. If that is not true, then we would have to call that power supply by some other name. But there is a problem.
If we make a microphone patch using a TT (bantam) patch bay, in one short moment of switching on or off the connector, a short circuit of one active pin to ground occurs. Then the second active pin is loaded with a full voltage of 48V towards ground and that moment is dangerous for dynamic but more so for ribbon microphones. That is why reconnection with active phantom power is never recommended.

Thats exactly what im after. The hot patch is one of the experiments on the table. The burst you get from having that moment of high voltage and "enough" current. In the software simulations im getting a significant spike on the back end of the transformer (motor/capsule side). Just a matter of what the damage looks like when that happens.

Im trying to find an unbalanced mic that has a transformer in it. My thoughts are, that wont go well. Esp if the cable is wired with pin 3 to ground or if the second lead of the transformer goes to ground.

 

[Potato Cakes]

Your thoughts are correct. Any properly designed dynamic or ribbon microphone connected with the correct cable must survive phantom power. If that is not true, then we would have to call that power supply by some other name. But there is a problem.
If we make a microphone patch using a TT (bantam) patch bay, in one short moment of switching on or off the connector, a short circuit of one active pin to ground occurs. Then the second active pin is loaded with a full voltage of 48V towards ground and that moment is dangerous for dynamic but more so for ribbon microphones. That is why reconnection with active phantom power is never recommended.

+1

 

[emrr]

I did indeed have a brand new Beyerdynamic M160 die with inadvertent live 48V patching.

I didn’t look for a link to the original paper, but here’s an update:

https://www.thatcorp.com/datashts/AES127-000183.pdf

Describes some large transient current faults, none of which I want passing through any transformers.

There are plenty of legacy mics and preamps with grounded transformer center taps that will draw current to ground, albeit symmetrically. 10-50 ohms resistance to ground at best sags your phantom supply.

 

[sr1200]

I did indeed have a brand new Beyerdynamic M160 die with inadvertent live 48V patching.

I didn’t look for a link to the original paper, but here’s an update:

https://www.thatcorp.com/datashts/AES127-000183.pdf

Describes some large transient current faults, none of which I want passing through any transformers.

There are plenty of legacy mics and preamps with grounded transformer center taps that will draw current to ground, albeit symmetrically. 10-50 ohms resistance to ground at best sags your phantom supply.

Thank you for the info.
So if the transformer has the center tap tied to ground we're going to have constant current running through, which can cause that mag. current and cause heating of the coil. I wasn't aware that CT's were used in microphones, good to know. Also, didn't think about it being on the pre side, which would cause the same result? Seems an odd place to put something like that since even with no mic plugged in pin 2 and 3 are going to be connected to this inside the preamp causing constant current flow.

Just read through the "THAT" document you shared. Very interesting read and the protection the suggest (on the preamp side) is pretty comprehensive. Wonder if anyone is going to implement that in the real world.

 

[emrr]

So if the transformer has the center tap tied to ground we're going to have constant current running through, which can cause that mag. current and cause heating of the coil. I wasn't aware that CT's were used in microphones, good to know. Also, didn't think about it being on the pre side, which would cause the same result? Seems an odd place to put something like that since even with no mic plugged in pin 2 and 3 are going to be connected to this inside the preamp causing constant current flow.

Just read through the "THAT" document you shared. Very interesting read and the protection the suggest (on the preamp side) is pretty comprehensive. Wonder if anyone is going to implement that in the real world.

Nothing in the phantom power era has a grounded center tap. Most RCA ribbons and tube era preamps have grounded center tap connections, for truly balanced circuits. In theory the windings in those cases should be balanced, so a gently applied small current won't disturb anything, just inadvertently drain off available current from a modern 48V approach. The original 48V approach is current through 1 resistor feeding a center tap, which confirms the theory. Many mics transfer 48V through the mic secondary center tap to the internals.

There's a fair number of RCA input transformers that don't like the CT floated, I suspect the internal shields are connected to ground through that point. Fine with phantom so long as there are blocking caps, plus a condenser into an input transformer with 20-27dB gain will need a 20dB pad in most cases, between the transformer and the phantom, so there's another 680ish in each leg, so rather than 10-50 ohms to ground it's more likely 700-720 to ground on each leg. Getting in the weeds......

 

[sr1200]

Nothing in the phantom power era has a grounded center tap. Most RCA ribbons and tube era preamps have grounded center tap connections, for truly balanced circuits. In theory the windings in those cases should be balanced, so a gently applied small current won't disturb anything, just inadvertently drain off available current from a modern 48V approach. The original 48V approach is current through 1 resistor feeding a center tap, which confirms the theory. Many mics transfer 48V through the mic secondary center tap to the internals.

There's a fair number of RCA input transformers that don't like the CT floated, I suspect the internal shields are connected to ground through that point. Fine with phantom so long as there are blocking caps, plus a condenser into an input transformer with 20-27dB gain will need a 20dB pad in most cases, between the transformer and the phantom, so there's another 680ish in each leg, so rather than 10-50 ohms to ground it's more likely 700-720 to ground on each leg. Getting in the weeds......

Thank you this is good. I have a couple RCA mics in the studio and this lines up with other things i've been finding about older vintage stuff. (im guessing the single resistor feed made things easier since tolerances in the mid 60s were still kinda rough. (now we have .1% tolerance on film resistors... i dont think ive seen a carbon comp under 5%) or perhaps it was just an easier and cheaper way to do it since you only needed a single resistor.

Most of what i've seen backs this up too. As long as everything is working properly, even this scenario shouldn't be too much bother. Its when you get a single lead to ground, or the power doesn't come up both rails at the same time that things get ugly.
Thanks!

 

[emrr]

CT power is also a descendent of telco circuits, and a callback to dual button carbon mic powering circuits.

 

[CJ]

i remember Oliver at Tab Funkenwerk talking about inductive spikes killing off V72 input transformers,
kind of like the ignition coil of a car,

 

[emrr]

i remember Oliver at Tab Funkenwerk talking about inductive spikes killing off V72 input transformers,
kind of like the ignition coil of a car,

Seems that V stuff has problems with insulation breakdown. I've never heard of phantom killing anything else.

 

[kags]

Great info- thanks!

 

[KA-Electonics.com]

A phantom fault will magnetize a preamp transformer core, mic output transformer core, or a dynamic mic element. As we have shown elsewhere even the hefty RE-111 can distort significantly with small amounts of DC current injected.

Do not assume that when applying phantom power to a ribbon mic that the preamp has a transformer input. It may be a capacitively-coupled "active" input and, on turn-on, can create differential currents that could damage the ribbon mic's transformer or element. I would trust the manufacturer's advice not to do it.

 

[gyraf]

..but before we go too hypothetical, let's dwell for a while and agree that phantom-induced errors are in fact very, very rare, contrary to common belief..

/Jakob E.

 

[KA-Electonics.com]

No I will not agree to that. Y'all have fun. I'm out.

 

[ruffrecords]

A phantom fault will magnetize a preamp transformer core, mic output transformer core, or a dynamic mic element. As we have shown elsewhere even the hefty RE-111 can distort significantly with small amounts of DC current injected.

Do not assume that when applying phantom power to a ribbon mic that the preamp has a transformer input. It may be a capacitively-coupled "active" input and, on turn-on, can create differential currents that could damage the ribbon mic's transformer or element. I would trust the manufacturer's advice not to do it.

Why would anyone apply phantom power to a ribbon mic??

Cheers

Ian

 

[kags]

Why would anyone apply phantom power to a ribbon mic??

Cheers

Ian

Why would anyone close a car door on their fingers? It happens. Not to be sarcastic, but it’s true.

There are times when phantom is only available as an “all or none” option. Also, it can happen accidentally. I can think of several times when someone accidentally presses the 48v button on a usb interface, either through user error or simply bumping the button while adjusting the device. I’ve seen this many times with the Focusrite Scarlett, where phantom is inadvertently and momentarily turned on or off by a user trying to move the interface on a desktop or “fat fingers” pressing buttons by mistake.

I understand the pitfalls of hot-patching a phantom-powered signal on a TT/TRS bay, but besides the (potentially destructive) audible thump of activating/deactivating phantom with an on/off switch or XLR patch, I’ve never seen first-hand a mic or other device adversely affected by phantom power. The exception being a T-Power mic that was damaged from phantom power on a sound stage.

I’ve personally tested applying phantom power to ribbon capsules and (modern) ribbon output transformers with no ill effects, but I wouldn’t try it dozens or hundreds of times to see if eventually some damage occurs.

-Adam

 

[sr1200]

Everyone's input kind of reinforces my desire to get to the bottom of things. lol

The general consensus is, obviously, not to apply phantom power to things that dont NEED it. My quest is to see where the dangers actually lie, and where marketing is just lying. As @kags stated, accidents happen and sometimes its out of your control. Being able to see what damage happens and how fast/slow it occurs is where I'm headed (and I actually WANT something to fail so we can point to a definitive event that caused the failure).

My experiment will be as follows. I plan on leaving the phantom power on for a period of time, not just a quick on off unless i see the mic fail.

1. Ribbon mic (no CT as i dont have a TX that has that available at the moment im willing to sacrifice)
   1. good cable
      1. phantom on/off
      2. hot patch
   2. bad cable (pin 3 unconnected / pin 3 touching ground)
      1. phantom on/off
      2. hot patch
2. Dynamic mic w/transformer
   1. good cable
      1. phantom on/off
      2. hot patch
   2. bad cable
      1. phantom on/off
      2. hot patch
3. Dynamic mic w/ no transformer
   1. Good cable (balanced)
      1. phantom on/off
      2. hot patch
   2. Good cable (unbalanced)
      1. phantom on/off (pin 3 tied to ground)
      2. hot patch

If anyone has practical suggestions that might be more useful, please share.