Analog Dialog Phantom Supply --Tiny DC Boost

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

PRR

Well-known member
Joined
Jan 30, 2010
Messages
11,143
Location
Maine USA
From the Analog Devices newsletter:

Ultralow Noise, 48 V, Phantom Microphone Power Supply Using a Tiny DC-to-DC Boost Converter
Professional condenser microphones require a 48 V supply to charge the internal capacitive transducer and power the internal buffer for the high impedance transducer output. This supply is low current, and the power supply must be very low noise because the microphone’s output levels are quite low in order to transmit the captured signal in high quality.
https://www.analog.com/en/analog-dialogue/raqs/raq-issue-176.html

Seems awful obsessed about "noise" (Phantom should reject noise), and the tutorial beats around a few bushes. You have to dig to find a nominal current; max current will need external datasheet shoveling.  They test with 1k load which is 48 (they say 50) mA which is 3 to 12 microphones (4 seems safe). I think there is a disable pin but not explained and turn-on/off thumping not discussed?

Multiple units in one system will want syncing, there's a pin for that.

Clearly they want to sell chips. The LT8362 is $5 at the usual distributors; they won't get rich from the audio market. The final circuit adds a SBCP56 transistor not made by Analog, 35 cents  (8 cents by the truckload).

This caveman will continue to double 24VAC to a Zener/BJT regulator, why be fancy?

> the microphone’s output levels are quite low in order to transmit the captured signal in high quality.

Well, no: because Phantom is a poor match to 200 ohm lines, so only a volt or so is possible without trickery too elaborate for the older Phantom mikes.
 
80hinhiding said:
Do you have preferences over which BJT (s) you use, and zener?  How much filtering capacitance do you like to use to keep ripple down?  In a 48V phantom situation for at least 4 mics.

Adam
The zener and pass transistor can be general purpose, as long as they check the voltage boxes. There are many examples of this around the internets.

Prudent design involves a HF bypass cap to ground at the output (already there). 

Indeed Paul is correct that this is mostly an exercise to sell ICs.

JR

PS: I have already responded to a thread about this on another forum, to not repeat myself so much, so close together.

https://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=12&t=1213
 
There are cases where using a DC-DC converter is unavoidable, like retrofitting phantom to a unit that does not have AC inside. Another example is a PSU I made for a 500-format lunchbox that has two smps. I didn't want to add another smps (which would have been significantly underloaded). In that case I would favour one with isolated output, which allows connecting the phantom negative right where it belongs. I never had any issue with noise, at least nothing that couldn't be managed with a simple CLC filter.
 
abbey road d enfer said:
There are cases where using a DC-DC converter is unavoidable, like retrofitting phantom to a unit that does not have AC inside. Another example is a PSU I made for a 500-format lunchbox that has two smps. I didn't want to add another smps (which would have been significantly underloaded). In that case I would favour one with isolated output, which allows connecting the phantom negative right where it belongs. I never had any issue with noise, at least nothing that couldn't be managed with a simple CLC filter.
I am repeating myself, but if you already have one SMPS you can use that switching waveform to drive a charge pump or cap doubler, tripler... Switching at high frequency means you can get away with using smaller caps, and there are no sync concerns.

JR
 
JohnRoberts said:
I am repeating myself, but if you already have one SMPS you can use that switching waveform to drive a charge pump or cap doubler, tripler... Switching at high frequency means you can get away with using smaller caps, and there are no sync concerns.

JR
If the smps had been my own design, that's what I would probably have done, but I don't relish interfering with an undocumented Chinese design.
 
Back
Top