Phantom power blocker

[trashcanman]

I’m looking for an alternative to using expensive DIs for protecting devices from accidentally being fried by phantom power from a mixer. I’ve done some googling and found that a passive RC highpass filter seems to be the way to go, such as This.
Is this a reliable design? I don’t know much about phantom power but it seems that should work and is commonly used for removing DC offsets, which I think is what phantom power is essentially. I’m intending to build a bunch myself and I think if I can get the board size down I can cram it into a cheap XLR coupler.
Using film or ceramic capacitors instead of electrolytic would help to reduce the space needed so I was wondering if doing so would change the filter characteristics? Doing some basic simulation suggests it would perform the same but I thought I'd consult the hivemind.

 

[gyraf]

What units are you afraid of frying?

Phantom power is designed to be inherently safe

The only things that I've seen taking damage from being exposed to phantom power are the newfashioned all-in-one-chip electronically-balanced outputs, if they're left unprotected (and they often are)

/Jakob E.

 

[mad.ax]

Using film or ceramic capacitors instead of electrolytic would help to reduce the space needed so I was wondering if doing so would change the filter characteristics?

Look at the actual size of a 100uF Film cap... Unlikely you're gonna cram a couple of those into a cheap XLR coupler...

 

[gyraf]

..and caps that size don't do no good anyway, or even worse - they store (and discharge) much more energy-at-a-time

/Jakob E.

 

[abbey road d enfer]

Since the posted link refers to protecting dynamic mics, it is important to evaluate risks.
First, I've never seen a dynamic microphone damaged by receiving phantom, even in the case of one leg being grounded, which shouldn't happen, of course, but as the saying goes, sh.t happens.
It produces a rather loud click, but generally it doesn't fry the mic.
It may happen, though, it seems some have reported it.
Now ribbon mics can be damaged by inadvertently being submitted to phantom, again in case of defective wiring. The mic doesn't fry, but the ribbon gets distended and becomes floppy.
The first protection is making sure the cables are correct, that no conductor is grounded. No patchbays or jack inputs for microphones either.

Now this thing can work (basically it's what is built in most transformerless mic preamps), but the capacitor value cannot be compromised. It is tempting to use the recent ceramic caps, with values that go up to 10uF, but a) they're only rated at 50V, which is too close IMO and b) their distortion is considerable.
Film capacitors would be very bulky (and expensive too), which puts a constraint on EMI/RFI susceptibility.
Electrolytic caps seem to be the only practical solution.
But anyway, listen to what gyraf said: this solution may not be always safe. Actually it may just displace the problem and add another element of potential hazard.
I repeat, check your cables, and avoid as much as you can hot-plugging.

 

[Newmarket]

It is tempting to use the recent ceramic caps, with values that go up to 10uF, but a) they're only rated at 50V, which is too close IMO and b) their distortion is considerable.
Film capacitors would be very bulky (and expensive too), which puts a constraint on EMI/RFI susceptibility.
Electrolytic caps seem to be the only practical solution.

Additionally the ceramic caps typically have X7R dielectric. There is an associated voltage co-efficient that may not be clear from the datasheet and may not be consistent across different manufacturers and device sizes. Basically your 10uF / 50Vdc ceramic capacitor is unlikely to be giving you 10uF at 48Vdc.

Temperature and Voltage Variation Ceramic Capacitor | Maxim

 

[JohnRoberts]

Presumably this is only an issue for SKUs plugged into XLR mic inputs. I recall a spate of failures in the new inexpensive wireless mic recievers (last century). The inexperienced product designers did not size the output blocking caps to handle 48V if/when phantom is energized. Some cheap mixers don't switch phantom power on/off for individual inputs.

JR

 

[transient]

To my mind, the most serious problem is caused by users connecting (electrolytic) capacitor-coupled or DC-coupled line outputs to mic inputs with the phantom power still switched on. Apart from possible headroom problems, this may reverse-bias the output capacitors, causing eventual failure, or it may upset the DC servo on the output, or even cause it to self destruct. Some manufacturers have taken evasive action, such as Mackie with an RC network and, for a while, Avalon Design fitted a TRS jack for the line output on the VT-737sp to discourage unthinking users from making an XLR/XLR connection to something like a computer interface's mic input. I have replaced many bulging and leaking capacitors and not a few failed transistors and opamps. It seems that phantom power is still widely misunderstood.

 

[swpaskett]

Stepping into the wayback machine, I learned the hard way that some old-fangled ribbons, hybrids (ribbon/dynamic like the 639s) and even some dynamics have a centertapped transformer with the centertap of the secondary (mixer side) tied to the shield. This creates a DC path through the transformer, e.g. pins 2 and 3 to pin 1, which can magnetize the transformer and over minutes to hours seriously hamper the performance of a once good mic. But what can be magnetized can also be demagged with a little low current AC applied to the xformer, you just don't always have a chance to do that before going live.
The simple solution to this problem is to open the DC path; this guy used caps, a no-parts method is to snip that centertap loose from pin 1. Check out any old mics: Altec 639, Shure 300, RCA 44 and 77 with an ohm meter before you plug into anything with phantom. I have found a number of these mics wired this way, and others from the same era are not. The others, I presume, were previously modded.

 

[john12ax7]

Apart from possible headroom problems, this may reverse-bias the output capacitors, causing eventual failure

The prudent thing, if using output capacitors, is to size them appropriately and provide a suitable discharge path. This would allow for continuous and repeated phantom power without damage. Unfortunately this solution does not seem to be commonplace.

 

[transient]

The prudent thing, if using output capacitors, is to size them appropriately and provide a suitable discharge path. This would allow for continuous and repeated phantom power without damage. Unfortunately this solution does not seem to be commonplace.

What is needed is pointing the +ve terminals of the capacitors to the outside world (often by default the other way round because there is no obvious polarisation with bipolar supplies and we cling to the idea of a single +ve supply) and installing suitable pull-down resistors which will attenuate the 48 volts nominal (could be up to 52 volts, of course) below the capacitor voltage rating, in conjunction with the 6K8 P48 resistors, and will not excessively load the output amplifiers.

 

[transient]

some old-fangled ribbons, hybrids (ribbon/dynamic like the 639s) and even some dynamics have a centertapped transformer with the centertap of the secondary (mixer side) tied to the shield. This creates a DC path through the transformer, e.g. pins 2 and 3 to pin 1, which can magnetize the transformer and over minutes to hours seriously hamper the performance of a once good mic.

I'm curious about this. Do the magnetic fields not tend to cancel out in a centre(or center if you like)-tapped transformer? Some mics (e.g. C414 B-ULS, C414 EB) pick up their DC supply from the centre-tap of their output transformer. Also, some mic preamps send outgoing phantom power through a centre-tap on the input transformer primary.

 

[trashcanman]

Thanks for all the answers everyone.

What units are you afraid of frying?

Devices that don't have protection in their outputs to withstand ~48V, usually laptops. Well made audio gear can often handle it, if just briefly, but the people operating the gear aren't always careful to check that phantom is off before quickly plugging in a laptop that someone just realised they need and there are no DIs available. Something like this might be an inexpensive solution in these situations.
You can actually buy them pre-made, such as this, but the price is close to that of a cheap DI and the design doesn't look too far from "a couple of caps in a cheap XLR coupler", no offense to them if they hang out here.

I only actually checked mouser for film caps after posting and realised that they're probably no better than using electrolytics. I also whipped up a board and managed to get the circuit small enough to fit in a cheap XLR coupler.
The only worry is if one of the capacitors fails short, would be good to have it fail safe. I did find another schematic, attached below, which adds zener diodes to limit DC voltage to an acceptable level should one of the caps fail. I would probably put a big LED on it too to show when phantom is on, maybe a buzzer.

 

[Newmarket]

Thanks for all the answers everyone.


Devices that don't have protection in their outputs to withstand ~48V, usually laptops. Well made audio gear can often handle it, if just briefly, but the people operating the gear aren't always careful to check that phantom is off before quickly plugging in a laptop that someone just realised they need and there are no DIs available. Something like this might be an inexpensive solution in these situations.
You can actually buy them pre-made, such as this, but the price is close to that of a cheap DI and the design doesn't look too far from "a couple of caps in a cheap XLR coupler", no offense to them if they hang out here.

I only actually checked mouser for film caps after posting and realised that they're probably no better than using electrolytics. I also whipped up a board and managed to get the circuit small enough to fit in a cheap XLR coupler.
The only worry is if one of the capacitors fails short, would be good to have it fail safe. I did find another schematic, attached below, which adds zener diodes to limit DC voltage to an acceptable level should one of the caps fail. I would probably put a big LED on it too to show when phantom is on, maybe a buzzer.

Couple of points wrt the attachment there:
It states
"The zener diodes are for spike suppression - should a higher voltage than about 6V appear on either + or - leg on the amp side of the capacitors, the diodes go open circuit and the leg is grounded."
Obviously incorrect wrt Zener Diodes operation.

It's not the best idea to use "clear" ie unshielded enclosure. I doubt if it gives a real problem at low impedance line level but still...

Note wrt electrolytic cap rating (I don't see electrolytic stated but shown as polarised) is correct but does seem to treat the voltage rating as a "safety/operational" parameter. True enough but also derating the Vdc increases the expected operation lifetime so the higher the better.

 

[user 37518]

To my mind, the most serious problem is caused by users connecting (electrolytic) capacitor-coupled or DC-coupled line outputs to mic inputs with the phantom power still switched on. Apart from possible headroom problems, this may reverse-bias the output capacitors, causing eventual failure, or it may upset the DC servo on the output, or even cause it to self destruct. Some manufacturers have taken evasive action, such as Mackie with an RC network and, for a while, Avalon Design fitted a TRS jack for the line output on the VT-737sp to discourage unthinking users from making an XLR/XLR connection to something like a computer interface's mic input. I have replaced many bulging and leaking capacitors and not a few failed transistors and opamps. It seems that phantom power is still widely misunderstood.

This is very true, when it comes to output coupling caps I always try to use non-polarized electrolytics or 2 standard electrolytics back-to-back. At least 50 V rating is also very important to have, many of the prosumer interfaces use the same input for the Mic pre and line, and if phantom is turned on, you will suddenly hear a loud pop.

 

[Whoops]

What units are you afraid of frying?

Phantom power is designed to be inherently safe

People that are not familiar with Live Sound may not see a use for this, but let me tell you:
In Live sound normally all the inputs are Mic Preamp Inputs with XLR connector.
Live consoles with some Line inputs exist but are exceptions and not the norm.
So if you want to connect your laptop, iphone or mp3 player to the console so you play music on the PA you will have to connect that device to 2 Mic Inputs (Left and right channels). It's the only way normally and a risk is presented if Phantom Power is turned on by mistake, and mistakes are something that we humans often do.
Turning on phantom power will send 48V into your laptop or phone output and can permanently fry that output.

So what is done normally to protect the device from phantom power is to use 2 DI boxes between the playback device and the Live Console Mic inputs.
Probably that's what the OP was mentioning.

I personally don't like to have 2 unnecessarily DI boxes between my playback device that I use to test the PA, I find the sound is worse after going through the 2 DI boxes than going direct to the Mic Preamps., so I used in the past Phantom Power blockers for protection and can totally see a use for it.

 

[Newmarket]

People that are not familiar with Live Sound may not see a use for this, but let me tell you:
In Live sound normally all the inputs are Mic Preamp Inputs with XLR connector.
Live consoles with some Line inputs exist but are exceptions and not the norm.
So if you want to connect your laptop, iphone or mp3 player to the console so you play music on the PA you will have to connect that device to 2 Mic Inputs (Left and right channels). It's the only way normally and a risk is presented if Phantom Power is turned on by mistake, and mistakes are something that we humans often do.
Turning on phantom power will send 48V into your laptop or phone output and can permanently fry that output.

So what is done normally to protect the device from phantom power is to use 2 DI boxes between the playback device and the Live Console Mic inputs.
Probably that's what the OP was mentioning.

I personally don't like to have 2 unnecessarily DI boxes between my playback device that I use to test the PA, I find the sound is worse after going through the 2 DI boxes than going direct to the Mic Preamps., so I used in the past Phantom Power blockers for protection and can totally see a use for it.

Can't argue against that. Although ime the desks I've used usually have one or two "stereo" line input only input channels where there is no P48 .

 

[sodderboy]

A simple iso transformer that travels with the sensitive equipment could also work, and not sound too irritating, like the Radio Shack Stereo Iso Transformer, a New York Dave favourite. The last version was plastic cased and pretty light. I think it was 600:600, prolly on ebay, but there are pretenders to the throne that are irritating (but good for old Korg stuff, in the studio, not "Live Sound").
Or just keep it off the mic ins if it is such a problem. There's rev returns and the like that have no p48 anywhere near people's sausage fingers.
Mike

 

[user 37518]

A simple iso transformer that travels with the sensitive equipment could also work, and not sound too irritating, like the Radio Shack Stereo Iso Transformer, a New York Dave favourite. The last version was plastic cased and pretty light. I think it was 600:600, prolly on ebay, but there are pretenders to the throne that are irritating (but good for old Korg stuff, in the studio, not "Live Sound").
Or just keep it off the mic ins if it is such a problem. There's rev returns and the like that have no p48 anywhere near people's sausage fingers.
Mike

Those Radio Shack transformers have saved me many, many times

 

[Captngeetch]

I use these on all my ribbon mics. Work great and give me peace of mind. Seems it would be simple enough to duplicate

https://www.sweetwater.com/store/detail/PhantomBlockT--triton-audio-phantom-blocker