LM380 circuit with cap between inputs

[disorder]

I need a desktop amp for work. I have a few LM380 chips and was thinking about the following circuit...

https://www.eleccircuit.com/wp-content/uploads/2010/01/lm380-mini-size-amplifier.jpg

It looks pretty standard to me aside from the cap C3 between the inputs. Does this act as a low pass filter? Higher frequencies fall into common mode?

Also, is the DC blocking input cap C1 needed here?

Thanks!

[PRR]

I think you actually know what you are doing, but want an amen.

Go with what is on the LM380 datasheet, not something you find on some unspecified blog.

The input cap is optional.

[thermionix]

Is C1 backwards?

[disorder]

Go with what is on the LM380 datasheet, not something you find on some unspecified blog.

Sound advice, I think I will follow it. I do spend too much time at night looking at circuits on questionable blogs.

[anotherjim]

Most chip amp circuits published, including datasheet ones, assume a really good DC power supply is nearby. If that isn't the case (average wall wart/brick), then V+, close to the chip, needs a biggish electro cap in addition to the 0.1uF ceramic that should already be shown. 1000uF is a good start.

[PRR]

> Is C1 backwards?

You'd think.... but no. See LM380 datasheet. The inputs are ground (V-) referenced. As disorder says, the cap is not essential.

Oh, I see, technically the inputs self-bias with 50K resistors, times PNP base current, the inputs are at a small positive voltage. But less than a tenth volt, and an e-cap is fine to several-tenths reverse (though you could turn it around and feel better).

The cap could reduce "scratch" in the volume pot. But LM380 was aimed at VERY cheap systems. Where saving a cap made a real difference on loss/profit. In DIY the cost-cutting does not have to be so severe.

[thermionix]

(though you could turn it around and feel better).

I already do.