# LM3875 for 4 ohm speakers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

#### sutton1

Hey all,

I'm building Peter Daniel's LM3875 gainclone kit, and having concerns about the gain of the amp. Clearly this design has worked for a ton of people, so I know I am misunderstanding something. I am a huge noob here (degree's in mechanical engineering, not this magical electrical stuff), so correct me when I'm wrong. The way it's set up gives a closed-loop gain of 100. If I'm playing from my iPod, which I intend to, that's an input voltage of 1.5V.

1.5V*100 is 150V coming out of the chip? Then 150V into a 4 ohm load is going to give 5625W and draw 37.5A. Things are going to go up in smoke! So, if I've got this right, I'm going to need some more input resistance?

Even if I use 8ohm speakers, that's still 2812W and 18.7A.

#### Frank Berry

Turn the volume of your iPod down.

#### sutton1

But those numbers seem ridiculous. Since posting, I read in the datasheet that the chip has overvoltage protection which limits the output current. Is that what I've been missing?

Also, Peter says that R1 (the input resistor) is optional and that he just puts a wire usually. That would make for a very high gain using the A = Rf/R1 +1 equation.

#### frank1

You can't get output voltage from nowhere. It has to come from the power supply and cannot exceed this. In fact, the maximum output voltage swing will be somewhat less than the power supply voltage.
If you attempt to exceed this the amp will clip.
I suspect that this is the fact you are missing.

Frank

Last edited:

#### sutton1

Okay. Cool, thanks

#### JMFahey

you may see it from the other end.
Your chipamp can provide, say, 16V RMS output.
If gain is set to 100, you'll need 16/100=0.16V RMS.
But your IPod *can* provide 1.5V RMS .
OK, either lower the IPod volume setting until it outputs 0.16V *or* use a volume pot set to get the same level reaching the amplifier.

#### gootee

Also, it would "only" have been 2812 Watts, not 5625 Watts. You have to square the RMS voltage, not the peak voltage, then divide by R.

For a sinusoid, the RMS value is the peak value divided by the square root of two (about 1.414).

#### Jay

Are you sure Peter Daniel's 3875 has such a high gain? That's too high. Too many distortion.

#### sutton1

That's in voltage ratio. If I've got this right, 100 times increase in voltage is 40dB.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.