LM3886 optimisation and power supply review - Page 3

ddd · 13th August 2012, 01:57 PM

Will you try the "true" circuit now?
The pot position mustn't make any difference.

AndrewT · 13th August 2012, 02:16 PM

12mVdc of output offset for a complete sweep of the vol pot is considerably better than the 100mV to 200mV of offset changes I have seen in other Direct Coupled implementations.

Cvp,
do not become complacent.
You must ensure that faults elsewhere do not damage your speakers.
Detect DC and correct excessive DC and isolate if outside the correction range is a minimum requirement for a DC coupled implementation.

cvpho1 · 13th August 2012, 03:31 PM

First I'll do the same measurements but with a 8ohm load and input signal.

On the second board I'll try the fixed polarisation resistor (aprox 20k as the feedback).

Regards

Quote:

Originally Posted by ddd

Will you try the "true" circuit now?
The pot position mustn't make any difference.

jackinnj · 13th August 2012, 04:00 PM

Bob Cordell's book "Designing Audio Power Amplifiers" has a very nice LM3886 design -- he takes advantage of the lower distortion when the inverting input is used. Use a dual FET opamp first as buffer, then as inverter (OPA2604). He also uses another OPA2604 as servo. If you take a look at the documentation for the LM3886 -- specifically AN-1192 you'll get the idea.

Bob demo'd his design at one of the NJAS meetings a couple years back and it sounded very good indeed. He also built in a bit of loop compensation --

jackinnj · 13th August 2012, 05:29 PM

Here's one I drew up for the LM47890 -- which is a pair of LM3886 -- mostly follows Cordell:

cvpho1 · 13th August 2012, 06:37 PM

Thank you, the design you suggest is fine but the requirements from the start were: simple and small amp.
Best regards

Quote:

Originally Posted by jackinnj

Here's one I drew up for the LM47890 -- which is a pair of LM3886 -- mostly follows Cordell:

cvpho1 · 13th August 2012, 07:03 PM

Quote:

Originally Posted by AndrewT

12mVdc of output offset for a complete sweep of the vol pot is considerably better than the 100mV to 200mV of offset changes I have seen in other Direct Coupled implementations.

Hi Andrew,

To be honest we don't have here a real "Direct Coupled".... there is still the Cin before the pot.

PS: In front of the amp will be a Technics SH-GE70

jackinnj · 14th August 2012, 12:25 AM

Quote:

Originally Posted by cvpho1

Thank you, the design you suggest is fine but the requirements from the start were: simple and small amp.
Best regards

I've probably sold 800+ PCB's for various iterations of the LM3886. This example I think to be the best.

A servo will prevent heartbreak.

cvpho1 · 14th August 2012, 07:42 AM

Measurements in normal mode, after 180min listening jazz at normal level for a 20m2 room.

WARM after 180 min
L R
3mV 2mV min 0%
3mV 2mV med 25%
1mV 0mV med 50%
-2mV -3mV med 75%
-8mV -10mV max 100%

Note 1: temp went from 29C to 38C

Note 2: as before, above is mentioned (in %) the position of the pot, not the value of its resistance

cvpho1 · 15th August 2012, 06:32 PM

I've redone the measurements with signal at the input after a few hours of listening. Basically the values are the same:

L R
3mV 2mV min 0%
3mV 2mV med 25%
1mV 0mV med 50%
-2mV -3mV med 75%
-8mV -10mV max 100%

I noticed that at max volume the DC offset is fluctuating.

Any idea what are acceptable values?

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
LM3886 Power supply	jfnewell1965	Power Supplies	25	2nd June 2011 03:18 AM
+/- 12v regulated power supply review needed	renfrow	Power Supplies	6	26th October 2008 10:35 AM
Power supply for LM3886	brconner	Chip Amps	7	23rd June 2008 01:48 PM
Power supply for LM3886	gcds	Power Supplies	10	24th March 2007 06:32 AM
LM3886 Power Supply Schematic?	newbie1	Chip Amps	4	16th January 2004 01:39 PM

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

13th August 2012, 01:57 PM	#21
ddd diyAudio Member Join Date: Oct 2005 Location: Sofia	Will you try the "true" circuit now? The pot position mustn't make any difference.

13th August 2012, 02:16 PM	#22
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	12mVdc of output offset for a complete sweep of the vol pot is considerably better than the 100mV to 200mV of offset changes I have seen in other Direct Coupled implementations. Cvp, do not become complacent. You must ensure that faults elsewhere do not damage your speakers. Detect DC and correct excessive DC and isolate if outside the correction range is a minimum requirement for a DC coupled implementation.

13th August 2012, 04:00 PM	#24
jackinnj diyAudio Member Join Date: Apr 2002 Location: Llanddewi Brefi, NJ	Bob Cordell's book "Designing Audio Power Amplifiers" has a very nice LM3886 design -- he takes advantage of the lower distortion when the inverting input is used. Use a dual FET opamp first as buffer, then as inverter (OPA2604). He also uses another OPA2604 as servo. If you take a look at the documentation for the LM3886 -- specifically AN-1192 you'll get the idea. Bob demo'd his design at one of the NJAS meetings a couple years back and it sounded very good indeed. He also built in a bit of loop compensation --

13th August 2012, 05:29 PM	#25
jackinnj diyAudio Member Join Date: Apr 2002 Location: Llanddewi Brefi, NJ	Here's one I drew up for the LM47890 -- which is a pair of LM3886 -- mostly follows Cordell:

14th August 2012, 07:42 AM	#29
cvpho1 diyAudio Member Join Date: Oct 2011	Measurements in normal mode, after 180min listening jazz at normal level for a 20m2 room. WARM after 180 min L R 3mV 2mV min 0% 3mV 2mV med 25% 1mV 0mV med 50% -2mV -3mV med 75% -8mV -10mV max 100% Note 1: temp went from 29C to 38C Note 2: as before, above is mentioned (in %) the position of the pot, not the value of its resistance

15th August 2012, 06:32 PM	#30
cvpho1 diyAudio Member Join Date: Oct 2011	I've redone the measurements with signal at the input after a few hours of listening. Basically the values are the same: L R 3mV 2mV min 0% 3mV 2mV med 25% 1mV 0mV med 50% -2mV -3mV med 75% -8mV -10mV max 100% I noticed that at max volume the DC offset is fluctuating. Any idea what are acceptable values?