LM3886 noise issue

zeus_threat · 24th July 2008, 06:18 PM

Hi i have a strange issue with my gainclone. Schematic attached here. There is a small noise(need to put my ear on speaker to hear) which is not 50Hz hum or high freq buzz but someting in between bout 1Khz/500Hz. This noise occurs when i connect it to my high pass active crossover.

Connected to another source the gainclone has no issue
Another discrete amp connected to the same active Xover no issue.

Can anyone help. Thanks

zeus_threat · 24th July 2008, 06:27 PM

In case it can help here is the PCB layout.

pacificblue · 24th July 2008, 07:38 PM

Your chance to find a solution will increase, if you post the cross-over schematic as well.

You could compare the values of Ri, Rin, Ci and Cin of the gainclone to the corresponding resistors and capacitors of the other discrete amplifier to start with.

AndrewT · 25th July 2008, 08:27 AM

Hi,
the decoupling is not located in the best position.
PIN5 desperately needs decoupling.
The schematic shows the R+C after the R//L, but the PCB is before.
Try changing the two 10R to 4r7 or maybe even 2r7.
The 22uF in the NFB is too small. Try 47uF.
I cannot see the speaker return.

zeus_threat · 25th July 2008, 02:40 PM

Hi andrew thanks for the reply will do the mods you advised. pin 5 is actually connected to the 220uF for the +ve rail. Dou you mean add additional decoupling for pin 5?

There is an error in the schematic for the zobel network its correct on the PCB. (Not in original circuit i've added it.

Speaker return is the green circle next to the caps the other circle is for power ground.

Do you have any other suggestions.

Thanks

zeus_threat · 25th July 2008, 02:43 PM

Thanks pacific for your reply i'll add it soon. The input network and feedback of my discreet amp are:-

Cin- 10uF Rin 15K for input
and for feedback Rfeedback=15K Rground=680 and Cground=22uF

AndrewT · 25th July 2008, 03:17 PM

Look at the length of the route from PIN5 to the decoupling cap.
You must have the HF decoupling either on or right next to the supply pins.

Look also at the Zobel return between the decoupling caps. Could they be picking up output pulses and feeding them back into the supply pins? Remember what Walt J. said "all pins are inputs"

Cin=10uF is even worse for allowing voltage to develop across the NFB cap.

Cin must limit the passband so that the PSU and the NFB do not have to handle frequencies at or near their poles.

The other components are also different from the posted schematic.
What are you doing to us?

zeus_threat · 25th July 2008, 07:37 PM

Hi please find attached the high pass x-over shcematic. I'll be posting PCB layout soon.

jaycee · 26th July 2008, 12:19 AM

Put the Zobel on the speaker terminals. I find a 10R 3W + 100nF mylar capacitor work well.

Here is my layout, very similar to yours, except mine is not a true Gainclone as there is only 330uF per rail of capacitance on the board near the chip, with the rest of the capacitance (2x4700uF per rail) is on a seperate board.

For reference, here is the schematic.

zeus_threat · 26th July 2008, 02:49 PM

Thanks jaycee, but cannot see your layout can you please attach it thanks.

Since its weekend i've been able to trace the source of the issue. Since it was more quiet around i've been able to listen to my discrete and my gainclone both are dead quiet with any external source but the issue i am having is the cross over itself its the one oscillating. Am not an expert with low level circuit layout but i will post schematic and PCB layout here in case anyone can help. Additional improvements to the gainclone are welcome but like am saying its working well and its not the culprit

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
Deluxe Reverb Re-Issue High Frequency Noise.	jordankersten	Tubes / Valves	1	22nd December 2008 09:20 AM
Help needed to sort noise issue with amp build	ShinOBIWAN	Solid State	28	7th December 2008 05:29 PM
DCX2496 ch6 noise issue	akunec	Digital Line Level	5	30th March 2006 10:10 PM
LM3886 p2p - ground issue	Stabist	Chip Amps	0	20th January 2005 02:03 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

24th July 2008, 07:38 PM	#3
pacificblue diyAudio Member Join Date: May 2008	Your chance to find a solution will increase, if you post the cross-over schematic as well. You could compare the values of Ri, Rin, Ci and Cin of the gainclone to the corresponding resistors and capacitors of the other discrete amplifier to start with.

25th July 2008, 08:27 AM	#4
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, the decoupling is not located in the best position. PIN5 desperately needs decoupling. The schematic shows the R+C after the R//L, but the PCB is before. Try changing the two 10R to 4r7 or maybe even 2r7. The 22uF in the NFB is too small. Try 47uF. I cannot see the speaker return.

25th July 2008, 02:40 PM	#5
zeus_threat diyAudio Member Join Date: Nov 2005 Location: Mauritius	Hi andrew thanks for the reply will do the mods you advised. pin 5 is actually connected to the 220uF for the +ve rail. Dou you mean add additional decoupling for pin 5? There is an error in the schematic for the zobel network its correct on the PCB. (Not in original circuit i've added it. Speaker return is the green circle next to the caps the other circle is for power ground. Do you have any other suggestions. Thanks

25th July 2008, 02:43 PM	#6
zeus_threat diyAudio Member Join Date: Nov 2005 Location: Mauritius	Thanks pacific for your reply i'll add it soon. The input network and feedback of my discreet amp are:- Cin- 10uF Rin 15K for input and for feedback Rfeedback=15K Rground=680 and Cground=22uF

25th July 2008, 03:17 PM	#7
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Look at the length of the route from PIN5 to the decoupling cap. You must have the HF decoupling either on or right next to the supply pins. Look also at the Zobel return between the decoupling caps. Could they be picking up output pulses and feeding them back into the supply pins? Remember what Walt J. said "all pins are inputs" Cin=10uF is even worse for allowing voltage to develop across the NFB cap. Cin must limit the passband so that the PSU and the NFB do not have to handle frequencies at or near their poles. The other components are also different from the posted schematic. What are you doing to us?

26th July 2008, 12:19 AM	#9
jaycee diyAudio Member Join Date: Apr 2004 Location: Norwich, UK	Put the Zobel on the speaker terminals. I find a 10R 3W + 100nF mylar capacitor work well. Here is my layout, very similar to yours, except mine is not a true Gainclone as there is only 330uF per rail of capacitance on the board near the chip, with the rest of the capacitance (2x4700uF per rail) is on a seperate board. For reference, here is the schematic.

26th July 2008, 02:49 PM	#10
zeus_threat diyAudio Member Join Date: Nov 2005 Location: Mauritius	Thanks jaycee, but cannot see your layout can you please attach it thanks. Since its weekend i've been able to trace the source of the issue. Since it was more quiet around i've been able to listen to my discrete and my gainclone both are dead quiet with any external source but the issue i am having is the cross over itself its the one oscillating. Am not an expert with low level circuit layout but i will post schematic and PCB layout here in case anyone can help. Additional improvements to the gainclone are welcome but like am saying its working well and its not the culprit