LM386 Question: What is the purpose of the 1nf capacitor between input and ground (pin 3)?

[StellarRat]

I was working on my little radio last night and trying to improve the output of the LM386. Two changes I made seem to have had a huge effect on the output volume. The first was to use a small audio transformer between the output (pin 5) and my speaker (not shown in schematic.) I understand why that helped. But, I also found a schematic online that shows a 1 nf capacitor going from pin 3 (input) to ground. Using that capacitor gave a large boost in output as well. Nearly all the other schematics I've seen online do not show this small change, but it sure seems important (at least in my circuit.) So, what does it do and why did it help so much? I understand bypass capacitors in a regular amp prevent degenerative feedback, but this doesn't seem to be the same thing. I've included the schematic.

 

[KeepItSimpleStupid]

I suspect it adds a volume based high frequency roll-off. At full volume about 16 kHz. As the volume is decreased the corner frequency gets higher in frequency. So, it also acts as an RF filter. This can prevent overload.

Do you happen to live near a radio station transmitter?

 

[StellarRat]

Well, the LM386 is attached to the detector of my radio through a 10uf isolation capacitor. As far as having an actual radio station nearby, not really. The volume increase at output is very substantial though. I didn't measure it, but by ear I'd say it's nearly 50% louder than it is without that capacitor. It seems like more than you would expect from just a filter. I was really shocked at how much difference it made specially since most diagrams do not include this.

 

[zahwi]

The 1nF supposed to decrease the volume. The reason that it increases the volume could be bad earth routing or insufficient supply decoupling.

 

[Gasboss775]

Well, the LM386 is attached to the detector of my radio through a 10uf isolation capacitor. As far as having an actual radio station nearby, not really. The volume increase at output is very substantial though. I didn't measure it, but by ear I'd say it's nearly 50% louder than it is without that capacitor. It seems like more than you would expect from just a filter. I was really shocked at how much difference it made specially since most diagrams do not include this.

My theory is that in the absence of the capacitor RF is getting through to the lm386 and reducing the efficiency of the amplifier by wasting power amplifying an inaudible signal.

 

[KeepItSimpleStupid]

What Gordon said.

 

[Gasboss775]

If you have a 'scope you might be able to see the amplified RF at the output of the LM386.

 

[cowboybob]

I also think Gordon nailed it. Cut off for that RC low-pass is 15,915kHz.

Although:

If you have a 'scope you might be able to see the amplified RF at the output of the LM386.

might be tough because the RF is likely filtered out.

 

[KeepItSimpleStupid]

I said the same thing in post #2.

I suspect it adds a volume based high frequency roll-off. At full volume about 16 kHz...So, it also acts as an RF filter. This can prevent overload..

 

[Gasboss775]

Although:
might be tough because the RF is likely filtered out.

I recently measured the frequency response of an lm386N-4 ( A TI specimen ) and was surprised to find that it had some gain up to as high as 1Mhz, in fact it was more or less flat to well over 100Khz, perhaps I just had a lucky specimen. It worked well as a square wave oscillator up to around 40Khz after that the transitions get steadily longer and you get a waveform like a triangle wave with hard clipping.

Like this:

 

[cowboybob]

I said the same thing in post #2.

Indeed you did. My apologies, KISS.

 

[StellarRat]

So your guys' theory is that RF was also being amplified by the LM386 until I put the capacitor on there....interesting. Now, I'm going to show my lack of knowledge (I'm really just a beginner at this I can read schematics, but theory and math aren't there yet) does this mean that an amplifier "gives" up gain on any particular frequency as the range of frequencies it has to amplify increases?? Instinctively that makes some sense to me. I suppose there is a formula for this? I guess another question would be why do most of LM386 schematics not have this capacitor i.e. I wonder why this particular designer put it in there? Whatever the case it makes it far superior for my purposes.

 

[KeepItSimpleStupid]

A single pole filter (RC network) has it's -3 db point at f=1/(2πRC). That point is where the gain is down by √2 or about 70.7%.
R is 10K or less and C is 1 nF.

In this circuit R varies.

A simple detector is a diode, if you go back to a crystal set and their are inherent diodes in the input of an opamp. The amplifier has such high input Z, it picks up stuff. So the RF gets rectified and amplified which means the amplifier overloads. If the supply rail is 12 and it somehow pushes the DC output to 10, it's hard to amplify more than a 4 V p-p output swing. e.g. an output of 10 +-2V.

ou might see it as an output offset and you could disturb it too. i.e. measuring it makes it go away/

 

[StellarRat]

OK, I'm going to get my brother's scope this weekend and look at every part of this radio and see if I can find these RF "leaks" or maybe some other interesting cause. This is fascinating.

 

[audioguru]

The datasheet of the LM386 shows a curve of its frequency response. At the 455kHz frequency of the IF of an AM radio and at the low end of the AM broadcast band its voltage gain is 20 times. If there is any IF or RF then it can cause the LM386 to clip the signal on top of the audio. If the signal source is low and the 10k volume control is set to halfway then it forms a 5k resistor feeding the 1nF capacitor to ground making a lowpass filter with -3dB at 32kHz and absolutely no effect on audio frequencies. Then only audio signals will cause full maximum output.

The LM386 is designed to have a maximum low distortion output of 0.45W into an 8 ohm speaker when it is powered from 9V or 0.32W into a 4 ohm or 16 ohm speaker. Then an output transformer added and using an 8 ohm speaker reduces the maximum output.

 

[StellarRat]

OK, you guys are pretty smart! I found a transistor in my RF amp (the very first stage in my radio) with an emitter going to ground without an emitter bypass capacitor. Apparently the schematic was wrong OR the designer just didn't realize it was a problem. I haven't used the scope yet, but once I put a .047uf capacitor from the emitter to ground the 1nf became irrelevant AND the RF amp is giving me more gain now. I think this might be my first real "victory" in electronics! From my reading in knew that bypass capacitors prevent degenerative feedback in amps. I assuming that the RF amp was injecting amplified RF back into the radio without the cap and as you guessed it was giving the LM386 too much to do. You guys gave me the clue I needed. Thanks!

 

[zahwi]

Your solution indicates that those guys were wrong and misleading.
Adding the 0.047uF increases the RF gain and the RF level at the audio amp. Going by their theory your audio output should have reduced farther but you observed the opposite.

 

[Gasboss775]

StellarRat & zahwi so the problem got sorted, but we don't quite no why. If we can figure that out the victory will be all the more so. Further speculation is difficult without a full schematic.

 

[StellarRat]

I'll post the full schematic when I have some time to draw it.

 

[zahwi]

StellarRat & zahwi so the problem got sorted, but we don't quite no why. If we can figure that out the victory will be all the more so. Further speculation is difficult without a full schematic.

The OP solution was adding a cap to the emitter, that increases the RF gain, more RF more audio more sound. This goes against the theory that the RF blocked the LM386.