Why 48V for phantom power? Who started this convention?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

owel

Well-known member
Joined
Jun 3, 2004
Messages
1,088
Location
Nashville, USA
I'm making a preamp that I planned to be powered from a single 48V supply... powering the preamp and the phantom power.

Well, I found a problem! Problem is, I'm using dual 22V toroid transformer and the most it can do is 47VDC unregulated. (I wired the secondaries in series 22-0-22)

I checked prices of dual 24V or higher transformer and there's a big jump in price from 22VAC to 24VAC... like from $16 to $55!!!! I don't want to pay that much for a power transformer.

I've tested my preamp running on 40VDC and it works fine... mic phantom power also works on 40VDC with no problems. So I'm thinking of just using dual 22V toroids and regulating it to 44VDC. I know it will work.

Question: So why is it phantom power for mics are specified at +48V? I see some mic specs that will work for phantom power voltages less than that. Who started this 48V convention? Is this a hard rule? If your equipment doesn't supply 48V phantom power to its mic, is it considered inferior?

In my opinion, if the mic works fine, I don't need exactly +48V phantom... 40 or 44 will do. So where did this 48V phantom power convention came from?
 
good question

I have some references in my old books ... Radio Eng Handbook - Henny ... seems there were a couple of phantom standards.

An interesting parallel is that Telephone exchanges had a 48 to 50 volt rail for every thing ... including my soldering iron.

So far I can't find a definative reason or start date.
 
Typical standard values of some components involved in the first designs?
Just guessing, interesting question indeed

Tony.
 
I think Neumann set that standard, at least they invented phantom powering.

Most (I'm scared to use "all", but I don't know of any exception) mics will work fine with 44 V.

Other standards include 24 V and 12 V, mostly used by film people. And there is the Tonaderspeisung, but forget about that one...

Regarding PS: you could use a 15-0-15 V transformer and a voltage doubler - but that is not the perfect solution either...

Samuel
 
I have seen some mixer specs that run at quite weird voltages - mostly live powered mixers
15v - Mackie 808
15v - Peavey MP400
15v - Shure FP24 (mobile field user preamp/mixer)

On history - have a look here...
http://members.aol.com/mihartkopf/neumannv.htm#U_67
 
Dual 22VAC = 2 x 22VAC = 44VAC + full wave rectifer and storage capacitor = 44vac x 1.414 = 62VDC unregulated. Perfect !

Either the transformer could not supply enough current or something else is wrong.

What is the current draw of your pre and what is the VA rating of the transformer? How did you wire the bridge rectifier to the transformer?

Your 22vac - 0v - 22vac transformer should be perfect for the job you do not need dual 24vac or higher as it will just be wasted as heat.

Joe:)
 
[quote author="JLM Audio"]Dual 22VAC = 2 x 22VAC = 44VAC + full wave rectifer and storage capacitor = 44vac x 1.414 = 62VDC unregulated. Perfect !
[/quote]
Perfect ? for low noise PSU you must have one lead of transformer
grounded. You can not do it with graetz.
Better are two diodes, wounding with central tap and connect central tap to ground like with duodiode tube power supply.
Or you 22VAC, one wire {the wire near primary} connect to ground and Delon doubler. Then feedback-stabilizer {50 Hz is no problem for it}
Some problem are two elyt caps with twice capacity.
But they are low voltage, small and cheep after development
of modern caps in japan.

...and 100 nF paralel to all diodes.

xvlk
 
Joe, XV,

That's what I thought too! In my original plan, I thought it would work.

first, I use 2 diodes wired from the (22/22) windings, with the CT winding as the ground wire on the PCB. I got 37VDC after the regulator.

So i ditched that first plan and wired the 2 trafo secondaries in series giving me 44VAC. then I connected the 2 (22V) windings into a W02M bridge rectifier, and from there got the (+) and (-) voltages. This time I got 47VDC.

The trafo is 1.4+ Amps. BTW, these are unloaded voltage readings.

I do have an LED and series resistor wired on the outputs of the PCB. But I don't think this will be a big load to drain the PSU.


[quote author="JLM Audio"]Dual 22VAC = 2 x 22VAC = 44VAC + full wave rectifer and storage capacitor = 44vac x 1.414 = 62VDC unregulated. Perfect !

Either the transformer could not supply enough current or something else is wrong.

What is the current draw of your pre and what is the VA rating of the transformer? How did you wire the bridge rectifier to the transformer?

Your 22vac - 0v - 22vac transformer should be perfect for the job you do not need dual 24vac or higher as it will just be wasted as heat.

Joe:)[/quote]
 
Ahhhhh I think I found the problem... I'm using the LM317HVT regulator... and since I don't have the right parts, I just looked in my junk box and picked the closest values of resistor and trimmers I have.

Using the formula Vout = 1.25 (1+R2/R1)

I picked a too high value of R1 for my replacement. The R2 trimmer I have is supposed to be 10K, but measures only 9.9 K.

Substituting 9.9K for R2 and 270ohms for R1 gives me 47.08 volts output... which is very close to my reading of 47V.
 
Circuit should look like this roughly
48v%20power%20supply.gif


Check that you have about 60vdc at the main storage capacitor. If so you can set the voltage to 48v easy. The resistor tolerances will mean that the voltage will be not exact even when using 1% types. That is why all of our kits have 25 turn trim pots as it is the only way to trim the voltages exactly to the right voltages. To make sure the power supply is as ripple free as possible make sure to earth track is laid out just like in the schematic and take your signal ground from the point shown at the negative of C11 and do not ground any other point a long the track.

The 10ohm resistor between mains ground and signal ground limits currents in earth loops and limits earth loop humming from happening which is needed when unbalanced in & outs are being used.

www.jlmaudio.com

Joe
 
I believe this is the storie at least this applies in telco. Any electronics above 50 vdc required a high voltage cirtification from the fcc or some other entitie not sure exactly. but this is why you have most telco equiptment workling on 48v dc

Wil

Wilebee
 
as mentioned a million times, you do not always have to have 48 volts on a mic.
put a mic on a variable dc pwr supply and keep dropping the voltage til the mic craps out.
i am sure you will be very surprised at just how low you can go.

48 is a safe limit, as Jakob mentioned. That is why cars will someday (if not already) will have 48 volt batteries but nothing more.

here is a low tech pwr supply i built from scraps out of my lunch box.

only one piece of sand for low switching noise and grounded transformer, although i did not draw in the ground symbol which is obviously the bottom wire:

phantom_supply.jpg
 
Depends on how much current the head-amp draws, but usually there?s only ~34 volts at the most that gets to the mic circuit anyway. The rest is dropped across the 6.81Ks. And unless there is a voltage multiplier in there, even less gets to the capsule.

A condenser will usually ?work? down to very low voltages, but it?ll loose some headroom.
 
> If your equipment doesn't supply 48V phantom power to its mic, is it considered inferior?

There are two kinds of "48V" mikes:

Many recent mikes drop whatever they get down to regulated 9V internally. They don't care if you feed them 12V or 48V. Headroom is not lost. I feed my 414s whatever I have and they always work the same.

A few classics, and a few new/cheap mikes, use the phantom supply as capsule bias. Some of the Famous German Brand models sip very small current for the buffer, so the 3.4K Phantom resistors drop very little voltage, and 40V-42V is available to bias the capsule. If you feed them 45V instead of 48V, you get an insignificant loss of sensitivity; but if you feed them 12V the sensitivity is pathetic and the internal low-current buffer may starve to death. These mikes MUST be fed 45V-52V for proper operation, and unhappiness will be obvious around 30V-40V.

So read your mike specs. (There's a reason they shove those scraps of paper in the box.) If all your mikes indicate they eat 12V-48V, then you're cool with whatever. If any of them disclaim operation at low voltage, or only talk about 48V (45V-52V) operation, then you probably do want "full voltage".

> I connected the 2 (22V) windings into a W02M bridge rectifier, and from there got the (+) and (-) voltages. This time I got 47VDC.

You forgot the filter cap!!! Without the cap, a common DC meter will read the average voltage of a varying DC wave, which in this case should be 40V (though 47V no-load is reasonable). WITH the cap you get a steady DC voltage nearly equal to the peak of the AC wave, which computes to around 62V (as Joe says).

> Any electronics above 50 vdc required a high voltage cirtification from the fcc or some other entitie not sure exactly.

US FCC never specified studio voltages.

US NEC has special-cases for 25V and 70V power-limited. As I read it, I could develop my own 70V system for use only in the US.

Some European authorities favor lower voltages, even 24V max for household use. I think the 48V standard pre-dates the fad for 24V, and is allowed to continue because 48V Phantom is only used around professional technicians, you rarely find children chewing on Neumanns. Also 48V has been accepted for over a century as standard telco line voltage, and while it is possible to die on 48V it happens so-so rarely (if ever) that there is no pressure to change existing 48V systems.
 
Interesting and chatty skinny on phantom from friend Rick Chinn:

http://www.uneeda-audio.com/phantom/index.htm

some circuits too if you roll back to the home page. & other interesting & chatty stuff to be found on the home page germane to a couple of recent threads.
 
[quote author="PRR"]A few classics, and a few new/cheap mikes, use the phantom supply as capsule bias. Some of the Famous German Brand models sip very small current for the buffer, so the 3.4K Phantom resistors drop very little voltage, and 40V-42V is available to bias the capsule. If you feed them 45V instead of 48V, you get an insignificant loss of sensitivity; but if you feed them 12V the sensitivity is pathetic and the internal low-current buffer may starve to death. These mikes MUST be fed 45V-52V for proper operation, and unhappiness will be obvious around 30V-40V...[/quote]Thanks for the correction PRR - they loose sensitivity, not headroom (Doh! Gotta stop posting so late at night!). Also good explanation. That is exactly what I was referring to.
 
> The standard railway guage is 4ft 8 1/2 inches - another mystery!!

No mystery. It is the width of two horses' asses.

Roman chariots often used two horses. The width between the wheels is similar to the width of the two horses. The spacing was semi-standardized for convenience. (But not in English Inches!) They drove these rigs ALL over the known world, leaving ruts 4'-8.5" wide. For the next 2,000 years, anybody making wheeled vehicles ended up making them the same size, to fit in the Roman ruts. Even when railroad tracks appeared. Once a few railroads got started (on Roman gauge), all other RRs had to use the same gauge, so they could inter-operate well. (Ask the Australians, who got stuck with whatever odd gauges the English wanted to get rid of, and for many-many years could not drive a train from one state to another without stopping to change the axles to fit the different gauges in use.)

48V is enough to overcome losses in long phone lines and coupling coils, put ample current in a carbon mike, without loading the signal too much. It is also a standard voltage for stacks of lead-acid batteries, the standard way to get clean big DC before good electrolytics.
 

Latest posts

Back
Top