Author Topic: Trying to understand a couple of LM386 circuits  (Read 3436 times)

0 Members and 1 Guest are viewing this topic.

Offline SzewczykmTopic starter

  • Regular Contributor
  • *
  • Posts: 55
Trying to understand a couple of LM386 circuits
« on: November 12, 2013, 03:48:04 pm »
I posted another message looking for good component by component resources on amplifiers.  I do have some specific questions and I thought it would be better to start a new thread with those.

Here is one circuit: http://web.mit.edu/6.s28/www/schematics/lm386.htm

I'm wondering about all of the caps and how the values were determined.

C1: I've learned that this cap is there to remove the DC offset voltage. I get that you want it to exceed the input voltage, but I'm wondering how the value of 10uf was determined.

C2: This one is easy. It's outlined in the data sheet. Pins 1 and 8 are used to set the gain and they give you values to figure out if you should leave it open, use a resistor, or to use a cap.

C3: It's labeled "Bypass". I looked this up and found that I normally call the little .1uf cap near the power input on an IC a Decoupling Cap. I get why this is used but I'm confused as to why they'd dedicate a pin to it on the IC and not just assume you'd use it on pin 6.

C4: At first I thought this might be a filter, but it's not wired that way. I've read about this combination, using a resistor and a cap to create a delay. But why would you need that in this circuit? I thought that maybe it was being used as protection in case there was no speaker but I'm really not sure what that's there for.

C5: This looks like another DC blocker. But... why at this point?

Even if you put a name to something that I can go look up, that would be helpful.

Here is a second circuit:

http://www.instructables.com/id/LM386-Audio-Amplifier/step3/Observe-Diagram/

There are a few repeated themes, but this circuit has a low pass filter on the power input.  Why would you do that?

It looks like there is a high pass filter on the audio output.  C3 is 220uf and R4 is 10 ohm which according to my calculator gives a frequency of 72.343 hz.  So, are we blocking low frequencies on the speaker output?  Why would we do that?

Thanks for any hints or insight.

Mike
 

Offline edavid

  • Super Contributor
  • ***
  • Posts: 3383
  • Country: us
Re: Trying to understand a couple of LM386 circuits
« Reply #1 on: November 12, 2013, 04:48:44 pm »
C1: The impedance of the capacitor is in series with VR1, so its value determines the low frequency cutoff of the amp.  No need to make it enormous since the speaker can't reproduce low bass.

C3: Bypassing the low current input stage separately allows use of a much smaller bypass capacitor, and reduces interaction between output and input stages.

C4: The RC is called a Zobel network.  Without it, the LM386 would probably oscillate at about 1MHz.  This is a more advanced topic, just accept it for now.

C5: Most speakers aren't designed to tolerate DC flowing through them.
 

Offline codeboy2k

  • Super Contributor
  • ***
  • Posts: 1836
  • Country: ca
Re: Trying to understand a couple of LM386 circuits
« Reply #2 on: November 13, 2013, 02:59:58 pm »
To add a bit more..

As edavid said, the R1-C4 in series is a Zobel network, its placement there at the output of the amplifier is designed to compensate for the inductance of the speaker, and helps the amplifier to see less inductance and more resistance, i.e. to try to make it so that the output stage of the amplifier sees a purely resistive load.  Removing the inductance of the speaker is good for stability (as edavid said, without the RC, it may oscillate at about 1Mhz.. oscillation equates to unstable)

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf