Power line noise and USB audio

jmar · 30th May 2006, 02:14 PM

I have a PCM2902 that is using it's own power supply...nicely built on a PCB...works great except that if a glitch is imposed on the power line through something healthy like a refrigerator or a flourescent fixture on/off, it will disable.

If I go to device manager, I won't see "USB audio device" and the USB composite will show an exclamation mark of not installed.

If I reboot.....everything is fine until another AC glitch.... and it happens again.

I have the power supply filtered well on the LM317 input and output but still having problems.

Both jet audio and media player have been used.

Has this been a problem for anyone else?

jahonen · 30th May 2006, 03:40 PM

Does it the same when the audio is not connected, i.e. the USB cable is the only one connected? Try disconnecting the cables to find out which one is the interference source. For power, try using USB-power or just ordinary battery.

Problem is most likely common mode interference, it still remains regardless how much conventional (differential mode) filtering you apply.

Regards,
Janne

pburke · 30th May 2006, 04:22 PM

just a guess: the problem may be the computer's power supply dropping out at that point. Power on the USB bus is required to sense the devices on the built-in ports

try the computer on a big UPS - not a bad thing to get anyway. There are tons of cheap units on ebay that'll do the job for these brownout moments

Peter

jmar · 31st May 2006, 12:54 PM

I have the unit self powered, so I have the USB power line completely disconnected....(pin 1 on USB connector does not see pin 3 on the PCM2902).

Pin 1 isn't connected at all.

I was concerned about this at one point but thought it wasn't a problem?

jmar · 31st May 2006, 02:12 PM

Also....

It only has this problem with the PCM2902 interface.

If I switch back to my sound card, it won't react to a glitch.

Sometimes I have to reboot, but lately, if I turn on/off the flourescent fixture (which is on the same circuit)....it will just stop the song.

I can hit play again and it will function but it always takes the windows volume control down about 50%.

It will do this wih both Jet audio and media player.

I'm going to try some more noise suppresion but sure would like some clues on this.

Thanks.

jahonen · 31st May 2006, 02:26 PM

Like I said, try to determine the coupling path of the interference (which disrupts USB-line communications). Does unplugging of the audio output cable (i.e. disconnecting the output cable and then playing music and experimenting with the light switch) entirely have any effect?

If you experiment with the filter, I suggest that you'll add some common mode filtering, unlikely the differential mode interference is the problem.

Regards,
Janne

jackinnj · 31st May 2006, 02:29 PM

just a guess -- the Codec demands 50 to 60 ma -- if you are using a 1,000uF storage cap on the LM317 it will discharge to less than 3.3V (from 5.0V) in 30 milliseconds or so --

i can knock out the device manager for my data-acquisition cards if i overload the current -- won't recognize the device until a hard reboot.

jmar · 31st May 2006, 02:41 PM

I'll be damn....if I disconnect the RCA's from the board it won't do it.

I have an AD823 opamp on the output of the PCM2902.

Now that I look at this board closer, I forgot to add some filtering components I had made room for on the board....a 470pf across the output and a 100k for impedance setting.

OOPS.....maybe that will clear it up....I'll let you know.

jmar · 31st May 2006, 03:04 PM

Well....no change, but definetly it's getting back to the computer through the receiver RCA jacks.

I have 470pf caps across each output but no change....also added the 10 ohm resistor between digital and analog grounds (which I had left out).

No hum at all in the thing....dead quiet....did have a ground loop that I cured initially.

Still acting up though...in fact now, not only will it stop the song when I hit the light switch but it will start it again if I turn it back on.

???

jmar · 31st May 2006, 09:28 PM

Jacjinnj....

I don't have a problem unless these RCA cables are connected to the amplifier.

So that eliminates any power supply problem.

I think somehow the data (D- and D+) is being corrupted by the induced noise....what is wierd is that it that the audio will actually turn on again if I play with the light switch enough.

How can any of this be a power supply problem?

Would a flourescent switch actually cause line voltage sag?

Perhaps there is a grounding design flaw on the PCB but I sure don't "hear it".

30th May 2006, 02:14 PM	#1
jmar diyAudio Member Join Date: Jan 2004 Location: Indiana	Power line noise and USB audio I have a PCM2902 that is using it's own power supply...nicely built on a PCB...works great except that if a glitch is imposed on the power line through something healthy like a refrigerator or a flourescent fixture on/off, it will disable. If I go to device manager, I won't see "USB audio device" and the USB composite will show an exclamation mark of not installed. If I reboot.....everything is fine until another AC glitch.... and it happens again. I have the power supply filtered well on the LM317 input and output but still having problems. Both jet audio and media player have been used. Has this been a problem for anyone else?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to remove noise from DC line?	tripathdude	Digital Source	1	20th March 2008 04:26 PM
Noise in my Line Splitter - HELP	weston530	Parts	2	4th July 2005 07:13 PM
EDN on Line Noise, Hash, Nasties	jackinnj	Everything Else	0	19th March 2004 10:14 PM
Transmitting audio over power line	paulb	Solid State	3	5th April 2002 08:08 AM
Power Line Noise	Thoth	Everything Else	18	5th September 2001 01:55 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

30th May 2006, 03:40 PM	#2
jahonen diyAudio Member Join Date: Jan 2002 Location: Kuopio, Finland	Does it the same when the audio is not connected, i.e. the USB cable is the only one connected? Try disconnecting the cables to find out which one is the interference source. For power, try using USB-power or just ordinary battery. Problem is most likely common mode interference, it still remains regardless how much conventional (differential mode) filtering you apply. Regards, Janne

30th May 2006, 04:22 PM	#3
pburke diyAudio Member Join Date: Nov 2003 Location: Madison Wisconsin	just a guess: the problem may be the computer's power supply dropping out at that point. Power on the USB bus is required to sense the devices on the built-in ports try the computer on a big UPS - not a bad thing to get anyway. There are tons of cheap units on ebay that'll do the job for these brownout moments Peter

31st May 2006, 12:54 PM	#4
jmar diyAudio Member Join Date: Jan 2004 Location: Indiana	I have the unit self powered, so I have the USB power line completely disconnected....(pin 1 on USB connector does not see pin 3 on the PCM2902). Pin 1 isn't connected at all. I was concerned about this at one point but thought it wasn't a problem?

31st May 2006, 02:12 PM	#5
jmar diyAudio Member Join Date: Jan 2004 Location: Indiana	Also.... It only has this problem with the PCM2902 interface. If I switch back to my sound card, it won't react to a glitch. Sometimes I have to reboot, but lately, if I turn on/off the flourescent fixture (which is on the same circuit)....it will just stop the song. I can hit play again and it will function but it always takes the windows volume control down about 50%. It will do this wih both Jet audio and media player. I'm going to try some more noise suppresion but sure would like some clues on this. Thanks.

31st May 2006, 02:26 PM	#6
jahonen diyAudio Member Join Date: Jan 2002 Location: Kuopio, Finland	Like I said, try to determine the coupling path of the interference (which disrupts USB-line communications). Does unplugging of the audio output cable (i.e. disconnecting the output cable and then playing music and experimenting with the light switch) entirely have any effect? If you experiment with the filter, I suggest that you'll add some common mode filtering, unlikely the differential mode interference is the problem. Regards, Janne

31st May 2006, 02:29 PM	#7
jackinnj diyAudio Member Join Date: Apr 2002 Location: Llanddewi Brefi, NJ	just a guess -- the Codec demands 50 to 60 ma -- if you are using a 1,000uF storage cap on the LM317 it will discharge to less than 3.3V (from 5.0V) in 30 milliseconds or so -- i can knock out the device manager for my data-acquisition cards if i overload the current -- won't recognize the device until a hard reboot.

31st May 2006, 02:41 PM	#8
jmar diyAudio Member Join Date: Jan 2004 Location: Indiana	I'll be damn....if I disconnect the RCA's from the board it won't do it. I have an AD823 opamp on the output of the PCM2902. Now that I look at this board closer, I forgot to add some filtering components I had made room for on the board....a 470pf across the output and a 100k for impedance setting. OOPS.....maybe that will clear it up....I'll let you know.

31st May 2006, 03:04 PM	#9
jmar diyAudio Member Join Date: Jan 2004 Location: Indiana	Well....no change, but definetly it's getting back to the computer through the receiver RCA jacks. I have 470pf caps across each output but no change....also added the 10 ohm resistor between digital and analog grounds (which I had left out). No hum at all in the thing....dead quiet....did have a ground loop that I cured initially. Still acting up though...in fact now, not only will it stop the song when I hit the light switch but it will start it again if I turn it back on. ???

31st May 2006, 09:28 PM	#10
jmar diyAudio Member Join Date: Jan 2004 Location: Indiana	Jacjinnj.... I don't have a problem unless these RCA cables are connected to the amplifier. So that eliminates any power supply problem. I think somehow the data (D- and D+) is being corrupted by the induced noise....what is wierd is that it that the audio will actually turn on again if I play with the light switch enough. How can any of this be a power supply problem? Would a flourescent switch actually cause line voltage sag? Perhaps there is a grounding design flaw on the PCB but I sure don't "hear it".

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)