Should phantom power be supplied by a linear supply?
- Thread starter Lerok
- Start date
Help Support GroupDIY Audio Forum:
Hard to say, but one of the possibilities is that it's a multi-output smps with a single switcher. In this configuration, the PS is stable as long as the total load of the rails exceeds a minimum.
rjb5191
Well-known member
Here's some anecdotal evidence. For my Daking FET II comps, one came with an APS 48V 0.25A linear supply and the other with a Meanwell 48V 0.5A Switcher. Switching between the two supplies on the same unit, there is a distinct difference. The linear supply is punchier and bigger and the switcher thinner and more anemic and I strongly preferred the linear supply. I'd say the improvement is about 20% and worth it to me to find another linear supply.
All this is 100% subjective. I can hardly accept it as "evidence".
Then there is something wrong with the smps or the way it is implemented in the unit. have you checked the voltage, and how clean it is?
It's always a risky concept to put figures on something that cannot be measured. There are no units for punch, bigness, thinness, anemy.
I have used enough smps to affirm they are at least as good as linear ones for powering most audio processing equipment, as long as they do not radiate unwanted interference.
Now, have you read the title of this thread? There are circumstances where smps are not ideal. Another is tube guitar amps, where the PSU "sag" is an integral part of the sound.
rjb5191
Well-known member
The supplies I mentioned are both external "brick" supplies with one being a Advanced Power Systems (good quality) class 2 transformer / linear type and the other a Meanwell smps. The meanwell smps is what Daking supplied as a replacement for what I imagine was the original class 2 linear supply that went bad or was lost, so the meanwell is newer and the APS older. They both put out the correct voltage but I have not measured for noise. There is no undue noise in the audio listening-wise so there's no obvious indications that the SMPS is bad. The linear supply simply sounds a bit better than the SMPS when powering the same unit. I have no skin in this game, it's just what my ears tell me consistently. It led me to contact Daking aas I wanted to find another advanced power systems supply to better match the unit, but they are not made anymore unfortunately.
The nature of phantom power that is supplied common mode to the microphones provides a degree of rejection.
Can you qualify what you mean by "a bit better"?
The most obvious vector for audible phantom power problems could be low voltage, then perhaps large amounts of ripple/AC noise on the Phantom DC.
JR
Matt Syson
Well-known member
Switchers have a significant amount of HF noise on their outputs. Somewhere betweeen 60 and 100KHz. This can easily be 20millivolts or so, depending on actual loading and then what level of 'post (supply) module filtering is used. Advanced power (XP or other 'industrial equivalent) specify around 1 or 2 millivolts and will depend on current drawn but the 10 millivolts 'ripple and noise' would only be reached when you are nearing full capacity of the module, which at less than 14 milliamps per mic channel represents a lot of channels.
I find it very surprizing that Daking put on the market the smps version without properly evaluating it.
What is their current implement of phantom PS?
First thing I did when I got my MOTU M4 was take it apart and look at how it was designed. It uses three switching power supplies to step USB 5V to + and - audio rails and a 48V phantom rail. It's a studio grade interface from a prestigious manufacturer. I don't perceive any audible noise when phantom is turned on, even at high gain.
It's important to note that even if switching noise makes it into the 48V rail, as long as the 6.8k feed resistors (as well as other circuit impedances) are well-matched, that noise will be common-mode and should be cancelled out in the preamp.
Mismatches in other circuit impedances will manifest in more ways than just phantom power noise.
Phantom-powered devices generally have fairly low output impedances in the audible range (<100 Ohm), and should be pretty capable of shunting away any audible noise from the phantom supply. The 6.8k feed resistor and the device's output impedance make a voltage divider that should attenuate any noise voltage quite a bit.
If you're really worried you can follow up a switching regulator with a linear post-regulator with good PSRR, and/or a capacitance multiplier.
It's important to note that even if switching noise makes it into the 48V rail, as long as the 6.8k feed resistors (as well as other circuit impedances) are well-matched, that noise will be common-mode and should be cancelled out in the preamp.
Mismatches in other circuit impedances will manifest in more ways than just phantom power noise.
Phantom-powered devices generally have fairly low output impedances in the audible range (<100 Ohm), and should be pretty capable of shunting away any audible noise from the phantom supply. The 6.8k feed resistor and the device's output impedance make a voltage divider that should attenuate any noise voltage quite a bit.
If you're really worried you can follow up a switching regulator with a linear post-regulator with good PSRR, and/or a capacitance multiplier.
rjb5191
Well-known member
Just to be clear, this is the power supply for their FET II compressor. There's no phantom power involved here. I thought it was relevant because the sonic merits of linear vs switchers in regard to phantom power was being discussed so that was extrapolated to non phantom audio circuits. For the sake of relevance, my modified rolls phantom hex sounds better than the phantom given by my 500 series chassis. Go figure. I'm not trying to convince anyone here that I'm right. It's more for those that are interested in optimizing the sonic performance of their gear so they can try things for themselves and decide what's best for their ears. In my case, I wish the meanwell would sound as good or better because it is smaller and much lighter, but unfortunately it doesn't. When I have the time, I'll post some clips and you can decide for yourselves.
Why would Coil Audio go through the trouble of building this?: CA-70S 2U Microphone & Line Preamp | Coil Audio
Snake oil salesman? Or maybe their ears tell them something.
Last edited:
Then, if it's corroborated by DBLT,there is something wrong with the phantom PSU in your 500 chassis.
I have two 500 chassis, an 11-slot and a 6-slot. The 11-slot uses the silent arts linear PSU and the 6-slot uses one that I built out of 5 X-Power smps. There is no audible difference I can detect between the two sets.
Or you just don't have proper test equipment. There are plenty of ways that SMPS could cause problems for audio circuits, but there are also plenty of ways to work around those, or design so that they don't occur in the first place. But you can't design high frequency switching power supplies with your ears, you need the appropriate test equipment for verification.
You just have to translate it to something more objective, like frequency response differences (possibly differing with level), or level differences, or non-linear effects (modulation, new frequencies popping up from interactions between signal and power supply noise), etc. One weird thing to watch out for is that sometimes "thinner and more anemic" actually means "cleaner" in a technical sense. A little bit of harmonic distortion or a little bit of intermodulation can spread out the spectrum just enough to give a little "heavier" or "thicker" sound. Not saying that is what happened with the Daking preamp example, just that you have to go in without set preconceptions, and just see what the data tells you.
rjb5191
Well-known member
Ok. Here's the samples. Everything the exact same except I switched power supplies. Gave each run 5 minutes warm up time before recording. They sound different to me! I'm not excluding that one of the supplies may not be performing optimally but there's no obvious indication that one isn't.
https://www.dropbox.com/sh/9grvnutws35vn5a/AAA6G7pqRl1yh1SmmmaGL4Raa?dl=0
Dropbox link since the attachments don't seem to be working for me.
View attachment PS Test A.wav
View attachment PS Test B.wav
https://www.dropbox.com/sh/9grvnutws35vn5a/AAA6G7pqRl1yh1SmmmaGL4Raa?dl=0
Dropbox link since the attachments don't seem to be working for me.
View attachment PS Test A.wav
View attachment PS Test B.wav
Last edited:
Well, it seems you're using a prerecorded signal, so why don't you do a mult and then a null test?
rjb5191
Well-known member
I just did a null test and there's an audible difference above the noise floor.
Last edited:
You can't do a proper null test if you recorded your samples in different time. You need to record simultaneously and use the same converter.
It will be difficult to time synchronize samples processed at different times.
The sound character of the null result can advise about the difference. If it sounds phasey like a flanger that suggests a tiny time offset. If it is heavy in HF or LF content that can suggest a frequency response difference.
JR
The sound character of the null result can advise about the difference. If it sounds phasey like a flanger that suggests a tiny time offset. If it is heavy in HF or LF content that can suggest a frequency response difference.
JR
