Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

LM3886 PCB Any Problem?

Status
Not open for further replies.

dr.power

Member
Hello Guys,

sorry but can you guys Please let me know if there is any problem/ loop area/ Antenna and s o on in the below PCB which is the PCB for LM3886 power amplifier?

Thanks a lot
 
The PCB looks good. PCB is art. What looks good to one person does not look good to another.

I am not a big fan of star grounding. In part because the center of the star needs to be at the right spot. Why the 1000uF cap and not the LM3886 pin7 or the input or output connector ground? I am not saying to move the ground point, I am just asking why.

If I made the board I would, at a minimum, make the bottom 1/3 the board all ground copper. Go for the maximum solid connections to a strong ground. Connect J1,2,3 C8888, C555 and U300pin7 all in a solid block with out a star. This is only a preference of mine not a requirement. Many of my boards are about 90% copper.

Your board will work. I am just ranting.
Please post your schematic.
 
The PCB looks good. PCB is art. What looks good to one person does not look good to another.

I am not a big fan of star grounding. In part because the center of the star needs to be at the right spot. Why the 1000uF cap and not the LM3886 pin7 or the input or output connector ground? I am not saying to move the ground point, I am just asking why.

If I made the board I would, at a minimum, make the bottom 1/3 the board all ground copper. Go for the maximum solid connections to a strong ground. Connect J1,2,3 C8888, C555 and U300pin7 all in a solid block with out a star. This is only a preference of mine not a requirement. Many of my boards are about 90% copper.

Your board will work. I am just ranting.
Please post your schematic.

Hi And thanks for your comment,

I just went ahead and ended up with this layout, Thats why I put it here so that you guys help me out to improve it. I am suspect to the loop areas though.

heres the schematic:
 
what's the rail voltage? the reason i ask is that one of the "rules of thumb" for power amps is that the power amp clips at between 1 and 1.5Vrms input (standard line levels used in home audio). so for instance, a 100W amplifier clips at 30Vrms out, so it's set for a voltage gain of 30. so the 100W amp (assuming an 8 ohm load)would have an input waveform with a peak value of 1.414V, and the output waveform with a peak value of 42.42V, so the rail voltage would be about +/-45V or more (many manufacturers use +/-50V). an LM3886 is a 75W amp, so for an 8 ohm load you would use +/- 40V rails with a clip voltage of about 35V (it will actually be a little higher). this is an RMS voltage of 24V, and again an input voltage of 1V, so you might actually want R701 to be 24K.
 
No fuses on the 2 supply rails? Or fusible resistors?

No pulldown resistor on the input? A 470k or 1meg would be advised.

No Zobel network (as shown in the datasheet) to reduce chance of HF oscillation?
 
what's the rail voltage? the reason i ask is that one of the "rules of thumb" for power amps is that the power amp clips at between 1 and 1.5Vrms input (standard line levels used in home audio). so for instance, a 100W amplifier clips at 30Vrms out, so it's set for a voltage gain of 30. so the 100W amp (assuming an 8 ohm load)would have an input waveform with a peak value of 1.414V, and the output waveform with a peak value of 42.42V, so the rail voltage would be about +/-45V or more (many manufacturers use +/-50V). an LM3886 is a 75W amp, so for an 8 ohm load you would use +/- 40V rails with a clip voltage of about 35V (it will actually be a little higher). this is an RMS voltage of 24V, and again an input voltage of 1V, so you might actually want R701 to be 24K.

The power sullpy is +-20Vrms.
The datasheet of the chip has a good example how consider the output/input voltyages by having the supply voltage , the speaker impedance and the requared power.
 
Some body told me that the lines shown in the below pic (the withe ones) are casing the AREAL LOOP.
Can you guys confirm it? What is an areal loop actually? Is it considered just for inputs or...?
What It does and how it affects the circuit?

By the way I am wondering how part of the ground trace acts as an areal loop?!
 
Last edited:
so, to use the "rule of thumb", take 20V, and multiply by 0.707, and you get 14.14, so R701 would be about 14k... or are you saying the output will be 20Vrms max? using the term "+/-20Vrms" is kind of self-contradicting, since the +/-20V is about DC rail voltages, but the "rms" part is about AC voltage..... maybe i'm just over-analyzing...

as for a ground loop, it doesn't look like you have one, but your "star" ground does look kind of out of whack, but it still might work. if i might make a suggestion, though, you may want to tap R701's feedback from the output connector pad, rather than along the trace going to it. so just like you did for the star ground, you would route the speaker side of R701 to the connector pad by having a diagonal trace from the pad, then a straight trace parallel to the output trace to the high side of R701.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top