Buffered LM3886 + overthetop regulated power supply

This is partly an exercise in learning to use PCB CAD software so it's more complex than strictly necessary but I'd still like to get a good quality, good sounding end product. Any comments welcome.

[IMGDEAD]http://www.deviantart.com/download/178399209/gainclone2_by_random007-d2y7ppl.jpg[/IMGDEAD]

Notes:

VAC A & B are the two secondary windings from a 160VA 30v toroidal transformer.

Power supply and main regulator circuit are from Roman Black's gainclone amp modified to be a symmetrical design.

Buffer/preamp pre-regulator design from Pre-Regulator

Buffer/preamp regulator from LM7815/LM7915 data sheet.

Buffer/preamp from ESP - A Better Volume Control (part 2) with a linear potentiometer used at input.

LM3886 circuits are pretty much generic.

Power switch and transformer circuitry not in the schematic.

Some of the odd component choices are because I have them in my parts collection already (eg lots of 2200uf 105degree capacitors), various power transistors and so on.

The links/jumpers are there so that it can be powered up and tested in stages.

I'll have to redraw the star ground, it's confusing the way it is with a shaded box.
 
Last edited:
Hi,
the 3886 does not need a buffer at it's input.
Your inadequate source may need a buffer at it's output.

This is a very complicated PCB project. Why can't you start with something much simpler?

See explanation below. It's probably misleading for me to call the op-amp a buffer.

Inadequate source? Can you explain?

All together complicated, granted, but each section is simple and easily understood. It's an exercise in learning that suits my level of experience.

Just out of curiosity, why the use of two op-amps (non-inverting buffer -> inverting op-amp with gain), as opposed to a single non-inverting op-amp with gain?

The op amps are there to change the response of a linear potentiometer at the input to reproduce a log response. See the ESP link. Seems like an interesting idea.
 
I really never did like that power supply design. Hey lets make a poor lm317 regulate till it dies and then have an external part shunt the current. Why Not just get a 5amp rated lm338

http://diyaudioprojects.com/Technic...338-5-Amp-Adjustable-Regulator-Data-Sheet.pdf

The author of the design seems very pleased with it but certainly I'll look at the 338, thanks for the suggestion.

As everyone is saying. Why the op-amps at the input at all. The Gain Clone has enough gain on its own to amplify a 100mV line input. Why add extra noise, distortion and complication with the op-amps.

See the ESP link. Granted, I'm having second thoughts.

Oh, I didn't see a pot connected to the feedback loop in your schematic.

Sorry, there is a mistake in my diagram. Will update it so that the way the feedback loop interacts with the pot is shown. *headslap*

NOT a good idea to use a potentiometer in the feedback loop of an op-amp. If the pot fails, the op-amp will go into maximum gain which would be horrible if not disastorous.

See the orignial ESP article as my schematic is critically flawed. Will update it so that the way the feedback loop interacts with the pot is shown.
 
No offence, but this just looks bad all around. Let me walk you through my thoughts.
first the design was meant to be a 7-9 part per channel dc coupled project with a 15-30k input with nothing on the output but a zobel network.

You have an input that varies (puke if you dont have a dam good buffered source that ant a tube) betweek 50k and 5k, and has a gain of 10 (will this saturate the lm3886? what will the gain of the lm3886 be?). Its going into a massive 10uf (guess we are not using high quality parts unless you have $50 in D sized cans) cap into a 100k drain(noise).

After it gets out of this "volume control" it goes into the amp and then out into a 10ohm resistor bypassed with a choke. Even the feedback loop has a cap bypass. And on that note why is it being fed back into pin 10 (positive input). SHouldnt it be going to pin 9?

Now lets talk about the power supply. I cant find one person who likes regulated, but hey. and its regulated using a 160va (on the small side considering the unit can suck that much continuous).


Personally i would scrap the whole thing. start here:
DIY LM3886 Chip Amplifier (Gainclone) Project

Real simple and bullet proof. Then go nuts. Learn how to make a soft start circuit for the power supply. Dc offset sensing protection relay for the outputs. You could do source selection, motorize volume control, etc.

Keep the spirit of the gain clone. Short signal path, low parts count in the amplifier section,unbuffered (why do you need to buffer a 25k load with a 50k buffer?), with an unregulated power supply. Then you can add all the the goodies that the commerical amps have to keep from popping breakers at startup from the 300va torrid transformer and keep from popping tweeters by using a speaker protection relay.


If you really wanted to be fancy you could copy the ac/dc b1 buffer and make a fancy power supply and cap couple that (then your source would see mega ohm load instead of a 50k opamp) and use it to drive your lm3386.
 
I missed that power transistor but if the regulator fast enough?

The transistor has a worst case (maximum power) switch time measured in nanoseconds: entirely over engineered for an audio frequency task.

No offence, but this just looks bad all around. Let me walk you through my thoughts.
first the design was meant to be a 7-9 part per channel dc coupled project with a 15-30k input with nothing on the output but a zobel network.

You have an input that varies (puke if you dont have a dam good buffered source that ant a tube) betweek 50k and 5k, and has a gain of 10 (will this saturate the lm3886? what will the gain of the lm3886 be?). Its going into a massive 10uf (guess we are not using high quality parts unless you have $50 in D sized cans) cap into a 100k drain(noise).

After it gets out of this "volume control" it goes into the amp and then out into a 10ohm resistor bypassed with a choke. Even the feedback loop has a cap bypass. And on that note why is it being fed back into pin 10 (positive input). SHouldnt it be going to pin 9?

Now lets talk about the power supply. I cant find one person who likes regulated, but hey. and its regulated using a 160va (on the small side considering the unit can suck that much continuous).


Personally i would scrap the whole thing. start here:
DIY LM3886 Chip Amplifier (Gainclone) Project

Real simple and bullet proof. Then go nuts. Learn how to make a soft start circuit for the power supply. Dc offset sensing protection relay for the outputs. You could do source selection, motorize volume control, etc.

Keep the spirit of the gain clone. Short signal path, low parts count in the amplifier section,unbuffered (why do you need to buffer a 25k load with a 50k buffer?), with an unregulated power supply. Then you can add all the the goodies that the commerical amps have to keep from popping breakers at startup from the 300va torrid transformer and keep from popping tweeters by using a speaker protection relay.


If you really wanted to be fancy you could copy the ac/dc b1 buffer and make a fancy power supply and cap couple that (then your source would see mega ohm load instead of a 50k opamp) and use it to drive your lm3386.

I'm certainly going to drop the op-amp stage entirely, especially as I've just seen a nice looking stepped discrete resistor dual gang log control on ebay for under $30US.

Thanks for suggestions!
 

sasmit

Member
2006-05-31 10:57 am
You're going to end up with a larger heat sink with a regulated power supply, the power dissipation that the manufacturer of the IC has accounted for was with a unregulated psu. The sag in the rails which you would have had with a normal psu would have reduced the overall power dissipation. If you're using a regulated psu you're rails are going to stay pretty much the same even during heavy loading ( assuming your transformer is adequately sized ).

This will be in addition to the heat sinks which you would use for your
regulators. you would have to use insulating pads for either the LM3886 or the pass transistor if you plan to mount them on a single heat sink.

Designing the pcb with both the regualtor and the power amp together taking care of the heatsink mountings , is fairly challenging even for experienced pcb designers.

IMHO you should design separate pcb's for your regulator and power amp. If you're motivation for designing a single pcb was to reduce pcb cost, then I suggest put the art work in a single pcb in such a way that you can cut across the sections in the pcb's with a hacksaw or design perforations into the pcb with a panelling script ( for eagle ).
 
Last edited: