/NEWBIE/: +/- 18V and 48V power supply for a preamplifier (schematic)

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

keap

Member
Joined
Apr 11, 2013
Messages
6
Hi all.
I need to build a power supply for my upcoming preamp - looks like it is much harder than putting together the op-amp based amplifier itself.

I used the schematic from http://audio.gotroot.ca/minipow/ - just changed the rectifier circuit from discrete diodes to the IC, fixed resistors for potentiometers, and then added the voltage doubler and LM317 for the phantom power (gospel says it's all about the input-output voltage differential, and "true" voltage ceiling does not exist in this regulator).

So here is my circuit. Could someone take a look on it and tell me if there are any obvious errors made? I'm pretty sure the symmetrical supply will work as it should, but i'm not sure about the +48V circuit - especially the GND bus. The schematic is right from the LM317 datasheet.

Thank you in advance,
Krzysztof
 

Attachments

  • MiniPow v0.1 .png
    MiniPow v0.1 .png
    26.2 KB · Views: 144
keap said:
I used the schematic from http://audio.gotroot.ca/minipow/ - just changed the rectifier circuit from discrete diodes to the IC, fixed resistors for potentiometers, and then added the voltage doubler and LM317 for the phantom power (gospel says it's all about the input-output voltage differential, and "true" voltage ceiling does not exist in this regulator).
Essential info is missing. What is your DC current demand ?
You changed the single diodes rectifier to what type of bridge rectifier ?
For your 'fixed resistors for potentiometers' you might connect between trimmer pins1/3 for R6 for a clockwise increase in voltage.

I'm pretty sure the symmetrical supply will work as it should,
I'm pretty sure the reversed polarized caps (C5,6,8,10) at the negative voltage rail will explode.
Voltage rating of these parts is missing as well. With 24VAC +/-10% in, 35V rated caps for C3,4,5,6,15 are at least close, maybe already over the edge.
Depending on connected load, the Vregs up to 17V differential between Vreg-in/out times current will need huge heatsinking.

but i'm not sure about the +48V circuit - especially the GND bus.
C15 will be a polarized cap, value of cap is load dependant. C16 seems not needed with C11 present. Depending on connected load, 1800uF for C11 might be over the top, and with your 24VAC voltage doubler you probably want a 100V rated part for the up to 75V raw DC pre regulator. Using a LM317-HV or TL783 instead of a standard 317 Vreg will be the safer bet from gospel (or was it Murphy?) saying, if there is a shorted mic cable, someone will connect it where it does the greatest harm. 
With R7 120ohm, R8 set output voltage to 46V and would have a 1W rating. Maybe try 8K2 / 220 ohm for close to 48V out at half the heat generated in R8. Allowing 16.5mA for LED3 might build you a headlight. I'd try R9 4K3/1W for 10mA. 

The schematic is right from the LM317 datasheet.
...but you added some errors ...
Its just a start... good luck
 
Thanks for the great reply, I've corrected most of the things you've pointed out.

Harpo said:
Essential info is missing. What is your DC current demand ?
You changed the single diodes rectifier to what type of bridge rectifier ?

The rectifier is an IC - rated 100V/1.5A. Current? At most 4 IC-based preamplifiers, so I assume 50mA per channel (10mA for the op-amp, one or two leds)  - up to 200mA on the symmetric bus and up to 50mA (4x10mA phantom power) on the +48 bus.

I'm pretty sure the reversed polarized caps (C5,6,8,10) at the negative voltage rail will explode.
Voltage rating of these parts is missing as well. With 24VAC +/-10% in, 35V rated caps for C3,4,5,6,15 are at least close, maybe already over the edge.
Depending on connected load, the Vregs up to 17V differential between Vreg-in/out times current will need huge heatsinking.

Caps are reversed, it was a stupid mistake. I've set the voltage ratings 50V for the caps before the regulator, and 35V after.
I'm not sure why the regulators would see a 17V difference between the input and output - the simulator shows 23 +/- 1V between the V in and GND bus.

C11 will be a polarized cap, value of cap is load dependant.

Simulation in Falstad Circuit shows 1mF will keep the voltage within 1V. I used a current source as a load, don't know if it's a correct way to simulate phantom power current draw.

There's the new circuit diagram. I renamed some of the elements, to make it more logical.
Does it make sense now?
 

Attachments

  • minipow v0.2.png
    minipow v0.2.png
    24.4 KB · Views: 97
In the TL783 and LM317 datasheets, under the "recommended operating conditions" is mentioned that the minimum output current for the regulators to keep regulating the voltage is 15ma and 10mA respectively.

Since most microphones consume less than 10mA from the phantom supply (2-5mA for the most mics), wouldn't that be an issue for the operation of the regulator and/or even for the microphone?
 
C11 looks backwards...

It is generally not necessary to tightly regulate the +48V supply, for only one mic is is common to use a simple zener shunt regulator.

JR
 
Philip_BlueFX said:
In the TL783 and LM317 datasheets, under the "recommended operating conditions" is mentioned that the minimum output current for the regulators to keep regulating the voltage is 15ma and 10mA respectively.

Since most microphones consume less than 10mA from the phantom supply (2-5mA for the most mics), wouldn't that be an issue for the operation of the regulator and/or even for the microphone?
Between R7, R8 and R9 that's over 15mA of load by itself.
 

Latest posts

Back
Top