Stereo LM1875 / TDA2050 PCB with Power Supply

[kuldeepsingh]

Hi friends,

I am planning to make a Stereo PCB for LM1875 / TDA2050 with Power Supply. Just finished with initial PCB routing. The power supply routing will be converted to planes and there will be a central ground plane.

[Bone]

Why does the right hand amplifier not have wide power traces on pins 3 & 5 like the left hand one? And why the different zig zags on the signal earths

[AndrewT]

I count 4 terminals on the ground trace !
Which one are you planning to use? Note which one.

I see a gap in the trace tapping points for the PSU decoupling caps. Inside that gap are two terminals and two traces going elsewhere.

Where did you pick up these PCB design tips?

[kuldeepsingh]

I made planes now as shown. Trying to fine tune the ground plane.
Since its a single side PCB, the +- supply lines needed to be connect by jumper wires, VP_out <-> VP_in and VN_out <-> VN_in.

And the zig zag trace connects signal ground to main ground, its a kind of experiment only, i expect it to simulate 1Ohm ground resister. But i hope it doesn't cause any inductive effect. (Mr Andrew, kindly give your opinion on this). And the "one" you mentioned, i plan to make that at GND terminal connector, which also connects to transformer's center tap.

[AndrewT]

I have already offered comment, but you seem to have overlooked it.

[pacificblue]

The zig zag trace does not simulate a resistor but an inductor, which you preferably avoid in ground traces.

Where do you connect the transformers center tap?

You are not really using electrolytics for the RC circuits at the amp outputs?

[kuldeepsingh]

@ pacificblue
Then will widening the zigzag trace will inductance to minimum level at the same time providing a longer trace for signal ground. The only idea is to raise the signal ground from main ground without using any resister. Is it acceptable.

And the transformer canter tap will connect to GND header (just above the bridge rectifier). I am using 0.1uF at output zobal, ya, that can be converted to non-polarized cap.

And Mr. AndrewT, I got your point, but it will be more easy to apply that in double sided design. In my place the PCB fab cost doubles for Double sided PCB, and that's why i have difficulty to terminate all ground traces to a single point in single side PCB.

[AndrewT]

Originally Posted by kuldeepsingh

............but it will be more easy to apply that in double sided design. In my place the PCB fab cost doubles for Double sided PCB, and that's why i have difficulty to terminate all ground traces to a single point in single side PCB.

It's your single sided that is wrong.
Don't complicate it with double sided. Learn how to do single sided first.

[kuldeepsingh]

Then please guide me a little. I havn't taken any professional course in PCB routing or in electronics . can you edit my uploaded PCB schematic picture in paint & just give me a visual clue :]

[pacificblue]

Originally Posted by kuldeepsingh

Then will widening the zigzag trace will inductance to minimum level at the same time providing a longer trace for signal ground.

Widening it won't make a big difference for the inductance. Ground traces should not be meandering.

Originally Posted by kuldeepsingh

The only idea is to raise the signal ground from main ground without using any resister. Is it acceptable.

A 0,254 mm wide 35 µm thick trace will approximately have 1 Ohm for every 50 cm of length. There should only be one signal ground for both channels.