Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Can I derive 48V phantom power from a H.O. power supply?
#1
Hi,

Long story short, is there a way to source a stable 48 volts from a typical amplifier power supply without using another transformer?
Could I wire up a TL783 circuit to a voltage divider in the power supply?

I took my old H.O. turret board and chassis and turned it into a microphone preamp for recording purposes.
I don't think it matters much, but here's the preamp circuit I used.
https://tapeop.com/tutorials/30/build-tube-mic-pre/


I attached a photo of how I modded the power supply to work with the preamp circuit. 
It works great with dynamic mics. So then I tried to add phantom power and there's something that I'm not understanding about doing that.

The roughed out phantom switch+LED is in the article I linked above. I did not make the pad. In the phantom switch circuit I substituted a 270r resistor for R13 and a 100uf 100V cap for C7.

Anyway, my idea was to take that 220K bleed-off resistor in the H.O.'s power supply, R1, and make a voltage divider out of it to derive the 48 volts. 220K and 47K worked perfectly with my voltages. So then I had a switch that turned on an LED when phantom power was engaged. The current limiting resistor for the LED was 10K. When I flipped the switch, all of a sudden my phantom supply dropped below 10 volts. I realized that I was basically creating a new voltage divider since the DC saw that 10K limiting resistor as a path to ground, parallel to the 47K resistor.

OK, I compensated and figured/mathed out a combination of resistors for the bottom of the voltage divider and the current limiting resistor for the LED so thata when the switch was turned on, and the LED lit up, I'd be back to my 220K + 47K divider. I got 48 Volts.

But then when I plugged a condenser mic in, my phantom power voltage sagged back to 15 volts. It worked, but the signal from the condenser was very weak. 

So, what's the easiest way to give my build its own, stable for all microphone loads, 48 volts of phantom power?

Thanks!


Attached Files Image(s)
   
Reply
#2
How about a cathode follower, or a source follower?
Reply
#3
(06-05-2020, 01:29 AM)dalepsmith Wrote: How about a cathode follower, or a source follower?

all tubed out now, but thanks!
Reply
#4
I mean a FET source follower. As in http://www.geofex.com/Article_Folders/mo...tfolly.htm
That first diagram.
Reply
#5
There really isn't a good way to add a 48V phantom supply to a High Octane without significant modification.
Lets start with the requirements for a phantom supply, which is 48V at 15mA short-circuit current (48V through parallel 6k8 resistors).
In tube amp terms, that's a LOT of current, the HO's specified power transformer 269EX only has 71mA AC current available (about ~60mA DC current after rectification), most of which is already accounted for (53mA in the EL84, about 3mA for the preamp tubes).
We also don't have any voltage nodes very close to our target of 48V, closest being the B+3 node at 233V, but that node is too high impedance anyway, just like you found in your voltage divider experiments, so realistically you'd need to take your supply from B+1 at 250v. But that still leaves you with 202V to drop, an extremely wasteful excerise.

Similarly, trying to use the heater winding with a voltage multipilier is a fruitless excercise too, as 6.3VAC is too low to be practical, diode forward voltage drops would have you reach, by stage roughly ~8.2VDC, 15.7V, 22.5V, 28.6V, 34V, 38.7V, 42.7V, 46V, 48.6V, 50.6V. That's 9 stages of multiplication, so requires more than 9 times the input current than we want to sink, or ~135mADC (more like 500mA AC accounting for the terrible resulting power factor), the filter caps would need to be enormous (back of the envelope at least 10000uF each to avoid dropping out of regulation), and again we're limited by 2.5A available from the winding, of which 1A is already used for the EZ81 rectifier, 760mA for the EL84, and 300mA each for the 2 12AX7's, a total of 2.36A already, there's just not enough to spare.

The only practical option is to choose a different power transformer with an extra winding that is a good match, or a second transformer, something like 2x12V or 24VCT at 100mA (2VA+) with a voltage doubler.

Here's a good place to start for phantom supplies. https://sound-au.com/project96.htm

Hope that helps clear some things up.
Reply
#6
(06-06-2020, 12:14 AM)dalepsmith Wrote: I mean a FET source follower.  As in http://www.geofex.com/Article_Folders/mo...tfolly.htm
That first diagram.
Quality stuff in there, and I may just try a mosfet buffer in my next build.
Don't know if I would be able to get the current I need out of it though for phantom power.


(06-06-2020, 02:22 AM)Greg Robinson Wrote: There really isn't a good way to add a 48V phantom supply to a High Octane without significant modification.
Lets start with the requirements for a phantom supply, which is 48V at 15mA short-circuit current (48V through parallel 6k8 resistors).
In tube amp terms, that's a LOT of current, the HO's specified power transformer 269EX only has 71mA AC current available (about ~60mA DC current after rectification), most of which is already accounted for (53mA in the EL84, about 3mA for the preamp tubes).
We also don't have any voltage nodes very close to our target of 48V, closest being the B+3 node at 233V, but that node is too high impedance anyway, just like you found in your voltage divider experiments, so realistically you'd need to take your supply from B+1 at 250v. But that still leaves you with 202V to drop, an extremely wasteful excerise.

Similarly, trying to use the heater winding with a voltage multipilier is a fruitless excercise too, as 6.3VAC is too low to be practical, diode forward voltage drops would have you reach, by stage roughly ~8.2VDC, 15.7V, 22.5V, 28.6V, 34V, 38.7V, 42.7V, 46V, 48.6V, 50.6V. That's 9 stages of multiplication, so requires more than 9 times the input current than we want to sink, or ~135mADC (more like 500mA AC accounting for the terrible resulting power factor), the filter caps would need to be enormous (back of the envelope at least 10000uF each to avoid dropping out of regulation), and again we're limited by 2.5A available from the winding, of which 1A is already used for the EZ81 rectifier, 760mA for the EL84, and 300mA each for the 2 12AX7's, a total of 2.36A already, there's just not enough to spare.

The only practical option is to choose a different power transformer with an extra winding that is a good match, or a second transformer, something like 2x12V or 24VCT at 100mA (2VA+) with a voltage doubler.

Here's a good place to start for phantom supplies. https://sound-au.com/project96.htm

Hope that helps clear some things up.

That definitely does help!

I'm going to go with a transformer and just make a "phantom box." 

Is this transformer up to the task?
http://catalog.triadmagnetics.com/Asset/FS48-125-C2.pdf

run the secondary in series by jumping taps 6 and 7, get 48 volts AC, rectify and get something in the 60V range? Then regulate it and have 125ma that i could power maybe 4 mics from?
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)