Denoise a switching power supply
Oct 19, 2009 at 1:49 AM Thread Starter Post #1 of 12
nightanole

nightanole

100+ Head-Fier
Joined
Dec 24, 2008
Posts
325
Likes
10
Im looking for a 12-14v 3amp min power supply for a power amp, but it seems they are all switching types. is there a cheap way to denoise these?

When amb usb stuff gets power, they use beads and a cap, could i do something similar?

Or maybe get a 14v supply and make a 12v regulator out of a few parts? I dont have room for a tread.
 
Oct 19, 2009 at 10:06 PM Post #2 of 12
It depends on what you want to do, are you trying to reduce the power supply voltage ripple, or is there actual common-mode switching noise?

If you want to reduce the ripple, just throw extra capacitance at it.
If you're concerned about actual noise, i'd suggest a common-mode choke and some form of RF shielding -- or finding a better switching supply.
 
Oct 20, 2009 at 4:01 AM Post #3 of 12
IMO, SMPS noise is a grossly over-hyped concern for most power supplies made in the last five years or so. Yeah, you might have problems with early 1980s supplies you get in a surplus shop, or some really cheap Chinese supplies - but if you can hear any noise from an audio project connected to a reasonably decent modern SMPS, I'd be more inclined to blame the project than the SMPS.

All Electronics have a "desktop" (laptop-style) 12VDC, 7.5A Elpac SMPS for $27 on their website, and I'm willing to bet it'll work fine as-is, without any "denoising"...
 
Oct 20, 2009 at 5:52 AM Post #4 of 12
It may indeed be that a noisy switcher will "work fine", but I have yet to see one that's anywhere near as quiet as a decent off-the-shelf linear, much less a high-quality audiophile linear power supply.

I'm purposely not answering the original question because I don't have a good answer. If you don't have enough room even for a TREAD -- not that I think that would be a good solution to switching power supply noise problem -- I don't see that you have room for the big magnetics needed in a proper CLC type filter that would squish the noise.
 
Oct 20, 2009 at 5:50 PM Post #6 of 12
There were good points made here & would like to add that its not trivial to design an "audiophile-acceptable" SMPS supplying the current you are looking for.
For example, (post-)ripple filtering is not as simple as adding capacitance. It requires an inductor/capacitor filter tuned to operating characteristics of the target parts under target load.

My first attempt at this (in CTH) needed to go through revisions & a final tweak to be on par w/battery power and that was just for driving the tube heater. For other applications you might be looking @ the PSRR of what you are trying to feed.

I'd not expect off-the-shelf SMPSs, not designed for this use, to have decent filtering. But amp designs w/high PSRR may cope with them well enough for your needs.

I'd heed Tangent's comments here. Don't think it'd be worth a one-off design effort for sure. But if you are up to it get a hold of a good Switch Mode Power Supply design book & be ready to experiment.
 
Oct 20, 2009 at 5:58 PM Post #7 of 12
Tangent, I'm sure they measure quite poorly, and look horrible on a 'scope, but does that really matter for audio purposes if you can't hear any difference?

Everyone automatically hates on SMPSes the same way everyone automatically hates on using USB power for anything, and - I strongly suspect - for much the same reasons: a combination of "conventional wisdom", blind prejudice, and the fact that you connect that to a 'scope and crank the resolution up to 1 millivolt and OMG look at that, that's horrible, get it away from my precious audiophile system, yech.

My whole point was that nightanole doesn't appear to even have a PSU yet, but is looking for solutions to a problem that he(?) doesn't even know exists yet (i.e. PSU noise), and which I maintain isn't going to be an issue as long as he(?) doesn't buy the oldest, cheapest SMPS that can be had - assuming the end point is, indeed, to power an amplifier and not draw smooth straight lines on an oscilloscope...

"12-14V 3amp min" plus "power amp" makes me think "chipamp", and I can't think of any power chipamps that have particularly horrible PSRR, etc, off the top of my head. Even the various crazy Gainclone cults have embraced the SMPS as a perfectly workable power source...
 
Oct 20, 2009 at 6:53 PM Post #8 of 12
Well to be honest it will be a 41hz amp32 powering some 4ohm higher end car audio (older pioneer rev's) on a infinite baffle with a foot print the size of a clock radio. I was going to go with a ebay special 14v lcd power supply for around $10-15. I didnt have a problem with the power supply from cisco for the starving student but i wanted to touch base for a larger amplifier. Linear power supplies cost more then the amp32 from my pricing of just ac transformers that can supply 4-5amp at 12v.
 
Oct 20, 2009 at 8:54 PM Post #9 of 12
Quote:

Originally Posted by cfcubed /img/forum/go_quote.gif
For example, (post-)ripple filtering is not as simple as adding capacitance. It requires an inductor/capacitor filter tuned to operating characteristics of the target parts under target load.


Filtering the output voltage ripple really is as simple as adding capacitance to the output of an existing SMPS. If you're designing a SMPS from the ground up, then yes, the LC filter is going to need some attention.

Also, voltage ripple and SMPS noise are two different beasts. Generally, the voltage ripple is a design criteria, and the SMPS noise is a result of how good the designer is at doing a PCB layout and picking devices.

By switching noise I mean:
-- Noise caused by reverse-recovery current of power diodes
-- high-frequency (above several megahertz) ringing between some device capacitance and some parasitic board inductance.
If there is switching noise from either of these reasons, then it's more of an EMI issue and it's exponentially harder to deal with.


nightanole, if you're powering a car stereo, definitely go with an SMPS. The 41Hz Amp32 is switchmode as well, isn't it? In my opinion it would be overkill to throw a linear supply in front of a switchmode amp.
 
Oct 5, 2018 at 9:35 PM Post #12 of 12
Hello,

I needed to make a power supply using a switching regulator to drop voltage without wasting a lot of power, but I needed a extremely quiet supply.

The lt3042 voltage regulator is good, but the switchers chop was getting through.

I went on a development of useful filters to put between them.

I had forgotten that almost all ferrite beads are made for signals with no current and
saturate their magnetic material with just a few mA's and turn into just a short bit of wire.

I looked at surface mount inductors, but their inter-winding capacitance let the switch edges through.

The laird 35F0121-1SR-10 was a lone surprise after a lot searching.
Says it's got a 10amp limit, but that is an overheating limit. Looks good to 100mA's.
Even has a good spice model.

I was able to make an astoundingly quiet supply.

To see the microvolts of noise I had to use multiple lt1028's to get a gain of 100,000 to see what was left.
The spectrum analyzer clearly saw fundamental and harmonics coming out of the switching regulator.

One bead ( and some caps) cut it down to just a little of the fundamental and no harmonics.
Over a thousand fold lower. After a second there was no trace of anything.

The output of the ultralow noise voltage regulator was noisier. And it's rated at a couple of micro volt.

BTW, the new acrylic film SMT caps go up to 10uF and seem extremely well behaved.

Hope this is useful.
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top