diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Class D (http://www.diyaudio.com/forums/class-d/)
-   -   Noise troubleshooting Class D (http://www.diyaudio.com/forums/class-d/112889-noise-troubleshooting-class-d.html)

Tom7227 1st December 2007 04:07 AM

Noise troubleshooting Class D
 
Maybe someone can assit me with a troubleshooting idea.

From time to time I get class D amps repaired, ...the typical blown output FET or power supply FETS blown to pieces.

But I notice that on some amps I hear a high pitch whine in the speaker.

I question weather this is normal or is there a problem. I use a full range speaker to troubleshoot and I am thinking maybe thats why i hear it...

or is there really a problem...

if you scope the output of the class D amp, what would one expect to see ? power supply noise or high frequecy noise from the switch on and off of the H bridge?

Tom

Pano 1st December 2007 04:45 AM

If you scope the output you should just see the remains of the carrier. Should be a fuzzy sine wave.

Setting the scope to normal audio sweep rates should show you the same sine wave, but of course as an indistinct block or fuzzy line. You can measure DC offset that way, too.

That's all you really should see.

On the power rail you will see the switching fundamental plus a lot of other noise.

Be careful with grounds and the scope probe on BTL amps!

fredos 3rd December 2007 03:27 AM

Maybe is talking about beating probleme...Watch syncronisation between channel and power supply...

Fredos

Tom7227 3rd December 2007 02:04 PM

I have never heard or that before, could you explain that?

Do you mean a 60 HZ issue with an input that has a hot ground?

fredos 3rd December 2007 09:26 PM

Beating is made of switching residual on un-syncronized signal. If a channel run at 330Khz and the second one at 340Khz, you have about 10Khz of difference between them. This will make an audible 10Khz modulation over signal...And worst, if frequency of them are not stable, you will modulate the 10Khz difference...Imagine! Add to this a switching supply running at 108Khz (for example!), you will add another 16Khz of shifting to the modulated interference (beating), ie: third harmonic of 108Khz will be at 324Khz...And to worst that a bit more, use more than 2 stereo amplifier that dont have stable crystal reference......

More clear now?

Fredos

Tom7227 3rd December 2007 11:55 PM

Yes, I see what your talking about, but I guess since I cant see the schematic of the power supply, I can't actully test this.

Moreover is some acceptable, if bass is the primary frequency, would you even hear a 10 - 15KHZ signal thru the speaker...

fredos 4th December 2007 04:18 AM

With some woofer, yes!

Fredos

Eva 4th December 2007 10:13 AM

Beating frequencies are only charactetistic of clocked systems. Crosstalk between self-oscillating systems manifests itself as noise.

fredos 4th December 2007 04:49 PM

I already experience beating with self oscilating design...Specialy when they are supply by a SMPS and feed to a high effiency speaker (100db and more..)...

Fredos

Tom7227 4th December 2007 07:38 PM

Well, perhaps back to my original thread...how do you troubleshoot this?


All times are GMT. The time now is 01:18 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