My LM3886 PCB

[Razor_x]

I am designing my own PCB schematic for two LM3886 amplifiers on the same board (stereo). This is what I have made:



I have transformer with two center trapped secondaries that I 'm going to use with this amp.
Can you please check my design for any mistakes? Every advice is welcome before it is to late.

Thank you in advance!

RazoR

[picowallspeaker]

I don't understand the 1 ohm resistor after the rectification . Is it the correct value ?
Some may use it after the first smoothing capacitor but followed by another with the same value . I would keep the power rails symmetric and shorter , in this case you should cut the input traces and do some wiring over the PCB . Then ,each channel has its own star ground ,but you should think it globally , so instead of a star you could have drawn a circle or a square , which is the same for the purpose .
After : you could have it done in half the space you used , since all the "job" is done by the chip...

[Razor_x]

Yes, I thought of putting 1 ohm resistors after first smoothing capacitor, but now I am going to put jumper at its place.
I forgot to draw two input signal input capacitors (3,3u).
So is the grounding correct?

[pacificblue]

Originally posted by picowallspeaker
I don't understand the 1 ohm resistor after the rectification . Is it the correct value ?

The rectifier on the drawing looks like a KBU8 or similar. The datasheet specifies a protective resistor of 0,8 Ohm with 5000 µF and 1,6 Ohm with 2500 µF. 1 Ohm with 3300 µF seems to be a good enough approximation.

Nobody in the forum seems to use those resistors and nobody seems to have problems without them. That does not mean they should not be used. Manufacturers usually don't specify those things for fun, because customers hate any unnecessary additional cost and losses.

The alternative would be a high current bridge rectifier, like the KBPC in 35 A version or so. They can deal with big capacitors without protective resistors.

[picowallspeaker]

So the resistor is there only for the in rush current at power up , even to prevent a low wattage transformer to melt ?!?
The grounding for me is wrong . It does symbolize what a star ground is (for a channel) , but it looks like a spider web in this case : Don't be afraid , it probably works , sure it will without any problems .Just the concept ,right ?

[pacificblue]

Originally posted by picowallspeaker
So the resistor is there only for the in rush current at power up , even to prevent a low wattage transformer to melt ?!?

No, it is there to protect the rectifier by limiting the charging current to the capacitors.

[akis]

Amazingly I am also making a guitar amp for my son (who already has two guitar amps), using the same chip as the post amp.

Here is my schematic and board layout (I will have to post multiple times). I only do it in the evenings after work so please feel free to shout if you spot errors.

[akis]

By the way, I have not decided on the values of the feedback circuit just yet, so it is currently random on the schematic (x20 I think).

Anyway, this is the PCB (sorry it is too small to see clearly perhaps) ?

[picowallspeaker]

I thought the supply should be able to deliver full power immediately , as cables for speakers should be near zero resistance . maybe it is not the case for LM38xx series ,as they don't exhibit superb slew rates ;if people can detect difference in cables (transmission) , surely 1 ohm at the generator could cause fatigue at power peaks.
It is ok to over dimension the power of the rectifiers and of capacitors in the supply
BTW I don't understand the huge transformers often used for chip amps.Are they pretended to be used lately in case of chip failure for a discrete design?!?!?!

[akis]

And this is the separate PSU board. It will sit very close to the board, inches away, but separate nevertheless.