2
\$\begingroup\$

I am new to working with electronics and I am building a simple speaker.

The Problem

The speaker I built so far (poor sound quality aside) stops working if the volume is above ~80%. By stop working I mean, emitting a stuttering buzz as if only a fraction of the sound is making it through in between noise.

As per the suggestion in the comment, I have measured the voltage out of the booster and it is 5(.02)V both while playing and while "stuttering".

This is an example of the problem (audio).

The Speaker

  • battery: 18650 (3.7V) with TP4056 for charging
  • power: MT3608 boost converter to from 3.7V to 5V
  • speakers:2x 3W 4ohm
  • amplifier: PAM8403
  • audio-input: 3.5mm jack to amplifier
\$\endgroup\$
18
  • 3
    \$\begingroup\$ Does it come alive again when you cycle power, or do you have to bulid another one each time? \$\endgroup\$
    – TimWescott
    Apr 19 at 1:01
  • 1
    \$\begingroup\$ Lack of decoupling capacitors (poor power supply regulation) or incorrect grounding can cause instability at high enough gain (= freaky loud noises) \$\endgroup\$ Apr 19 at 11:52
  • 1
    \$\begingroup\$ Please edit your question to describe this "stuttering buzz". And, please measure the voltage from your boost converter when this stuttering buzz is happening, and include your results in your question. Again, preferably this will be a oscilloscope output, but if you don't have that, a description of what a voltmeter does will be helpful. We can't help you if you don't help us! \$\endgroup\$
    – TimWescott
    Apr 19 at 14:43
  • 1
    \$\begingroup\$ Yeah, it doesn't look like there's nearly enough decoupling or bulk capacitance on those \$\endgroup\$
    – pjc50
    Apr 19 at 15:32
  • 1
    \$\begingroup\$ A quick suggestion, try powering it directly from a reliable 5V source. Then you can at least rule out the boost converter if it works this way. \$\endgroup\$
    – Wesley Lee
    Apr 19 at 19:32
1
\$\begingroup\$

What you are describing is distortion. Your speaker isn't "breaking." If it broke, it wouldn't work anymore afterwards.

There are a couple of possible causes for distortion in your case:

  1. The battery can't supply the needed current.
  2. Your audio signal is too high.

First the battery:

  • You are using an 18650 cell to power your amplifier.
  • You are using a voltage booster to get 5V for the amplifier.
  • You have two 3 watt speakers.
  • The amplifier loses a little power as heat, maybe 10 or 15 percent.

You need about 7 watts at 5 volts. That's about 1.2 amperes from the voltage booster. That's closer to 1.6 amperes from the battery. Without specifications, I can't tell you if that is too much for the 18650 you have.

Your voltage booster is rated for 2 amperes. It ought to be OK.

  • Turn the volume down.
  • Measure the battery voltage and the 5V output. Write them down.
  • Turn the volume up to just before the distortion starts.
  • Measure the battery voltage and the 5V output again. Write the values down.

They should both be stable - neither value should drop much.

  • If the battery voltage drops but the 5V stays up, then the battery isn't up to the task.
  • If the battery voltage is stable but the 5V drops, then the booster isn't up to the task.
  • If both drop, then the battery is too weak and the regulator might be too weak. Use a better battery, test again.

If the battery voltage and the 5V are both good then the problem is that your signal is too high.

The PAM8403 IC used in your modules amplify by a little over 100 times.

A signal input of 1V peak to peak would give an output voltage of 100 volts peak to peak.

That's impossible, of course. The module only has a 5V power supply - the output can only be a maximum of 5V peak to peak.

You have to keep your input signal low enough that the output signal after the amplification will be less than 5V peak to peak. That would be about 50 millivolts peak to peak.

Typical earphone outputs such as your smartphone or MP3 player will have an output level of 200 millivolts peak to peak - or more.

You can reduce the amplification by changing two resistors.

Here's your PAM8403 module:

enter image description here

There are two resistors on there that are marked "103" (black with white lettering.) Those are 10 kiloohm resistors.

Replace them with 47 kiloohm resistors. That's an "eyeballed" value that ought to lower the amplification to a reasonable level. If you still get distortion at high volume, just use a couple of bigger resistors.

Rather than replace the resistors, you could just put a resistor in series with each input. That will also lower the amplification. It will also mess with the bass sound, though.


The 10k resistors are smaller than the resistors recommended in the PAM8403 datasheet. That gives it more amplification than it is supposed to have, and leaves it more susceptible to distortion and maybe other problems.

\$\endgroup\$
1
  • \$\begingroup\$ Thank you very much @JRE, for this comprehensive answer. This helps A LOT also to understand these concepts more broadly. I will try your suggestions and provide an update. Thanks again. \$\endgroup\$
    – Paladinic
    Apr 19 at 20:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.