diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Power Supplies (http://www.diyaudio.com/forums/power-supplies/)
-   -   12v DC wall wart creating noise (http://www.diyaudio.com/forums/power-supplies/209567-12v-dc-wall-wart-creating-noise.html)

razorrick1293 25th March 2012 11:34 PM

12v DC wall wart creating noise
 
Hi there,

I've recently tried to use a 12v 0.8A wall wart for a pre-amp project, never used them before, usually just build a mains supply right into the box. The problem is, when im using it I get quite a loud pop every 5 seconds or so. Ive tried a couple of different adapters, same problem. On the input socket I have a 10uF cap accross the terminals, with a 1N4001 diode going into a 470nF cap (just to protect the circuit in case of -ve voltage where there shouldnt be), then supplying the op amps. the output is a simple op-amp buffer, ive checked the coupling and I measure no DC output. When i swap the adapter for a battery the pop is no longer there. Am I missing anything? is there a cap/resistor/diode that should/shouldnt be somewhere?

Ricky

benb 26th March 2012 03:39 AM

This sounds unusual. The "tried a couple different adapters" suggests it's your circuit, but your description of it doesn't suggest a problem. You're talking a 'pop' in the signal output of the opamp?

Is this an old-fashioned 50/60Hz transformer wallwart, or a newfangled switcher (which would feel lighter than a 10 watt wallwart "ought to" feel)? Switchers are the new thing for everything (charging cellphones, flipcams, powered USB hubs), and I haven't even tried one for an audio project. The RF hash and switching harmonics can't be good for audio, though I can't imagine one making the pop you're experiencing. The old fashioned ones have been fine for powering simple audio projects.

I presume this is a DC output wallwart? I'd put it on an oscilloscope to make sure. And if it's rated at 12VDC at full load, it might be as high as 18V at light or no load.

hpp3 26th March 2012 04:04 AM

Since the problem doesn't exist with battery power, and you've tested other adapters, I would suspect something in the electrical system of your house. Do you have the wall wart plugged directly into your outlet, or a power strip, perhaps a fancy one with power-conditioning that may be on it's way out? Does the 'pop' correspond with another device in your house turning on or off? Do you live in close proximity to a welding or machine shop?

I've seen adapters do strange stuff, and the (usually) inexpensive regulating electronics in those things seem to be more susceptible to influence from outside interference than a solidly-built diy supply.

razorrick1293 26th March 2012 07:58 AM

I've tested 12.1 v at 100mA it's quite light so I would guess it was a switching one, I'll try hooking it up to a scope again, ive tried it with an AC one too (12VAC 1A, rectified to 20V at 100mA) ripple was quite low, (used a 1000uF reservoir cap) but the pop was still there. It was just plugged into a mains socket 240v 50Hz, we're not close to any kind of factory or shop, and the noise is unaffected by other things plugged in around it, i will try plugging it in somewhere else though.

nazaroo 17th April 2012 02:12 AM

Quote:

Originally Posted by razorrick1293 (Post 2960491)
I've tested 12.1 v at 100mA it's quite light so I would guess it was a switching one, I'll try hooking it up to a scope again, ive tried it with an AC one too (12VAC 1A, rectified to 20V at 100mA) ripple was quite low, (used a 1000uF reservoir cap) but the pop was still there. It was just plugged into a mains socket 240v 50Hz, we're not close to any kind of factory or shop, and the noise is unaffected by other things plugged in around it, i will try plugging it in somewhere else though.

I've tried several of these non-transformer walwarts for various audio projects, especially guitar-pedals. I had a lot of problems:

(1) The SoundSource (white ones) delivered plenty of current, but would place all kinds of weird 'hashing' and whistling noises on the power-supply rails of the guitar pedals. Sometimes this would be only a mildly annoying background, two quiet to cause me to take any evasive action.

(2) However, in some cases, there was a complicating bad interaction between pedals due to reliance on common source for PS. Some pedals (Digitech vs. BOSS vs. Behringer) could not stand being on the same supply, even though there was plenty of current avail.

(3) Especially bad was a Digitech Compressor pedal which although it had great behavior on its own, could only be used alone!. Frustrating.

To make a long story short, I never got to the bottom of why these various transformerless switching supplies couldn't be made to work.

In one case, I added about 5,000 uF of filtercap to a homemade pedal to remove 99% of PS noises, but I was never happy with this extreme solution, and wondered if the turn-on surge was safe.

They are no longer selling the older style larger transformer walwarts in most stores up here. I think it is because everyone is cannabilizing the transformers for the copper as scrap, so the 'dollar' stores are getting more for them from scrappers than they can get selling them.

This is a disaster for audio however, because now small transformer costs will go through the roof.

I gave up using cheap switching supplies for audio, except for the cheesiest and most non-critical applications, like grunge-metal overdrive distortion pedals. This isn't a recommendation.

If you can find older style transformer wallwarts at surplus stores, buy them up before they disappear.

benb 17th April 2012 02:45 AM

In the USA there are still lots of older, 50/60Hz transformer wal-warts in thrift stores (as well as an increasing number of the newer switchers), though I expect over time they will become rarer, as the switching supplies saturate the market and older ones slowly dribble into the thrifts from people's closets.

nazaroo 17th April 2012 02:53 AM

oops, meant to edit but quoted.

oh well.

Yes I think you can still get the older style with transformer, but this may get harder in the future.

Anyway, in theory a switching supply should work with audio, because the switching should be going on above the audio range.

However, as everybody discovers, theory and fact don't always coincide.
These 'supersonic' noises can modulate with audio and add distortion and deterioration of sound

I think a switching supply can be designed with audio in mind, but it needs extra components.

benb 17th April 2012 03:06 AM

Quote:

Originally Posted by nazaroo (Post 2988309)
oops, meant to edit but quoted.

oh well.

Yes I think you can still get the older style with transformer, but this may get harder in the future.

Anyway, in theory a switching supply should work with audio, because the switching should be going on above the audio range.

However, as everybody discovers, theory and fact don't always coincide.
These 'supersonic' noises can modulate with audio and add distortion and deterioration of sound

I think a switching supply can be designed with audio in mind, but it needs extra components.

Furthermore, when a switching supply is lightly loaded (or sometimes even when not), the switcher may "intelligently" skip cycles, or do other things that cause it to generate SUBharmonics of the switching frequency that land in the audio range, and that's even without the switching frequency or harmonics modulating some other nearby ultrasonic frequency source (like, maybe, a CFL lamp on the same power circuit), with the difference frequencies being in the audio range.

As the late Bob Pease said (and like many good sayings, he stole it from someone else): In theory, theory and practice are the same. In practice, they're not.

nazaroo 17th April 2012 03:11 AM

Quote:

Originally Posted by benb (Post 2988322)
Furthermore, when a switching supply is lightly loaded (or sometimes even when not), the switcher may "intelligently" skip cycles, or do other things that cause it to generate SUBharmonics of the switching frequency that land in the audio range, and that's even without the switching frequency or harmonics modulating some other nearby ultrasonic frequency source (like, maybe, a CFL lamp on the same power circuit), with the difference frequencies being in the audio range.

As the late Bob Pease said (and like many good sayings, he stole it from someone else): In theory, theory and practice are the same. In practice, they're not.

OMG Bob Pease my favorite designer!

I have his book the one where he discusses CAD circuit design, and there is a photo of him throwing his computer off the roof.
From the moment I gazed at that photo, I was a Bob Pease fan.

abraxalito 17th April 2012 03:15 AM

Very interesting and enlightening observations, thanks :)

Quote:

Originally Posted by nazaroo (Post 2988309)
I think a switching supply can be designed with audio in mind, but it needs extra components.

Methinks the onus is going to be on the audio designers to engineer their products to work with cheap switchers. Rather than the PSU designers to clean up their act - quality audio will always tend to be a minority sport compared to mobile phones and tablets.


All times are GMT. The time now is 04:48 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2