Lm3886 with single ended supply

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The LM3886 data sheet has a figure showing a "Single Supply Application Circuit"

1) has anyone tried building this?

2) what does it sound like

3) are there inherent disadvantages in a single ended supply besides needing capacitors on the output to cope with the DC?

Thanks,

Simon.
 
Hello,

well i havn't build a LM3886 on a single supply before, but yes there are some disadvantages using a ouput cap.
one is that it introduces some distortion, especialy in the higher frequency's.
although this is very low but measureble.
another is that you must use a very large value (in the order of 10.000 uf) to get sufficient low end responce in a 4 Ohm load
A third disadvantage is that those capacitors are costly and they take up alot of space in the amp.
and from this extra money you can decide to use a dual power supply instead.
if you decide to use a single power supply be sure to use high quality caps (Nippon Chemicon or Panasonics) bypassed with some Wima 2.2uf or so, this enhances the sound.

for the rest the sound should be the same.

An advantage of using a single supply is that you have perfect DC protection allready build in.

succes in building your amp:)

Greets captain out.............................!!
 
Hello,

well i havn't build a LM3886 on a single supply before, but yes there are some disadvantages using a ouput cap.
one is that it introduces some distortion, especialy in the higher frequency's.
although this is very low but measureble.
another is that you must use a very large value (in the order of 10.000 uf) to get sufficient low end responce in a 4 Ohm load
A third disadvantage is that those capacitors are costly and they take up alot of space in the amp.
and from this extra money you can decide to use a dual power supply instead.
if you decide to use a single power supply be sure to use high quality caps (Nippon Chemicon or Panasonics) bypassed with some Wima 2.2uf or so, this enhances the sound.

for the rest the sound should be the same.

An advantage of using a single supply is that you have perfect DC protection allready build in.
!

Try BTL in single supply mode...but u still need to create a virtual ground at the input side...advantages no capacitor at the output
 
one advantage of a single supply design is to bring a protection for the speaker in case
the amp fails and shows dc on its output. No relay needed.
I am not shure it sounds worse than symmetrical supply.

Yes...having a low drop p-type high current mosfet will switch & latch (in off stae) in case of short circuit in output semi's
 
Last edited:
Bridge Tied Load aka Bridged. It means to drive each side of the speaker with an amplifier, with the signals of opposite polarity, instead of connecting one side to ground. It gives double the voltage swing into the speaker, and approximately double the power, assuming the amp can handle the current needed.
 
Even with a dual rail supply you still have capacitors in the circuit - the power that is going to the speakers has to come from the power supply and most likely you have a bunch of big cheap electrolytics in there. People have reported improved sound quality when using higher quality caps in the power supply. And a dual rail supply has twice as many caps as a single rail supply. So I'd agree that in theory a single supply rail with cap at the output isn't necessarily a big handicap. Try putting a cap on the output of an existing amplifier - can you hear a difference ? Try a better quality cap - can you still hear a difference ?

If you run out of space in the amp, put the capacitor outside the amp. That way it'll stay cooler. Heck, maybe you can put it inside the speaker box.
 
Has anyone got schematics for hooking up two LM3886 in bridge mode (or preferably, four chips in bridge-parallell mode) and running them from a single sided power supply? The National application note does not show this, only bridge or bridge-parallell from dual-sided supply or a single chip from a single-sided supply...
 
Has anyone got schematics for hooking up two LM3886 in bridge mode (or preferably, four chips in bridge-parallell mode) and running them from a single sided power supply? The National application note does not show this, only bridge or bridge-parallell from dual-sided supply or a single chip from a single-sided supply...

Post #9!

http://sound.westhost.com/project14.htm

Regards zeoN_Rider
 
Has anyone got schematics for hooking up two LM3886 in bridge mode (or preferably, four chips in bridge-parallell mode) and running them from a single sided power supply? The National application note does not show this, only bridge or bridge-parallell from dual-sided supply or a single chip from a single-sided supply...
if you are not able to use these two sources then I think you are not ready to attempt a complex build like this.

Far easier to build a discrete output stage that can deliver your power needs.
 
Parts questions

Going back to the original question: building a gainclone with a single ended supply, I think I am ready to order parts and start building. I am going to use a regulated, snubberized single ended supply. I am going to use the application schematic from National Semi. Referring to that schematic (attached below), I have the following questions:

  1. Do any of the resistors need to be more than 1/4 watt?
    1. The 10 ohm out put R?
    2. The 2.7 ohm zobel?
  2. If I eliminate the pot (Rin), do I just replace it with a 10k, 1/4W R or should I use an 'L' network with ~1k series and ~20k to ground as I have seen on other schematics?
  3. For capacitors
    1. Cs is clearly an electrolytic, probably should be 100V (0-64V supply). If I am using the snubberized type supply, then Cs should just be ~100nF, not 1000uF, correct?
    2. Both Ca 0.1uF are shown as electrolytic (+ shown), but at that value, I would think a film type? Which is correct?
    3. The 10uF input cap is also shown with a (+), but I am sure it should be a high quality film type, yes??
  4. Regarding grounds - I have read the application sheet and many of the threads here and I think I have at least some idea of what to do, but I am not sure which "grounds" in the schematic to consider which type of ground.
    1. Rn is clearly signal ground
    2. I would think the Ca that ties to pin 4 would be power gnd., but what about the Ca on pin 7? I thought pin 7 was signal gnd? But then, pin 7 also connects to the emitter of Q1 which would make it power gnd???
    3. I assume Ci & Csn are "by-pass" gnd and would have their own star that then goes to the signal star? What about Cm? same for that?
    4. Cs is surely power gnd and probably a good point for the star, yes?
  5. Anything I missed re:schematic that I should think of before picking up the soldering iron?
Thanks for all the help. This forum had really been great.
 

Attachments

  • SELM3886.pdf
    109.5 KB · Views: 252
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.