5
\$\begingroup\$

I have a PAM8403-based amplifier module and am going to build a small portable speaker, mostly to listen to lectures and audio books. The module looks like this:

small amplifier circuit board

First I attached two 4 ohm 3 watt speakers and it worked fine. But I've read that it would make a clearer sound with 8 ohm speakers. And in the module properties it says it can work with 4-8 ohm speakers.

  1. A clear sound would be a benefit for me, but will I lose power (loudness) with higher resistance speakers?

  2. How is the sound quality related to the speaker resistance? What is the theory behind it?

All the speakers I am going to use would be the same diameter.

Here is the efficiency chart from the PAM8403 datasheet:

graph showing efficiency higher for 8-ohm speaker

As far as I know, efficiency is about delivered power, not quality of the sound. Please correct me if I am wrong.

\$\endgroup\$
5
  • 4
    \$\begingroup\$ It would be helpful for us to know where did you read that and who claimes so. Basically speaker sensitivity tells you how loud a speaker is per input power. Another thing is how much power the amp can push into different impedances. So you can't know whether a 4 ohm or 8 ohm speaker is louder just based on their impedance. The supply voltage is also important determining which impedance is better. \$\endgroup\$
    – Justme
    Feb 5, 2020 at 13:22
  • 6
    \$\begingroup\$ It wouldn't surprise me if the person who came up with that claim was overloading their amp with speakers with too-low impedance, and then discovered that when they used higher impedance speakers the sound 'improved' because the amp was no longer overloaded - so their conclusion was that higher-impedance speakers sound better when in fact not overloading your amp is what really sounds better ... \$\endgroup\$
    – brhans
    Feb 5, 2020 at 13:31
  • \$\begingroup\$ @Justme I've read it in more than one place :) Here is a link, says: "These are 8Ω 3W speakers which will sound more clear then 4Ω but not that loud": instructables.com/id/Portable-Bluetooth-Speaker-2x3W-Finished- \$\endgroup\$
    – Sohail
    Feb 5, 2020 at 14:25
  • 1
    \$\begingroup\$ @Sohail that does not say all 8 ohm speakers sound better than all 4 ohm speakers. It just means he tried one set of speakes first, and then tried another set of speakers, and from those two sets of speakers, he found them better. Maybe they are better speakers, regardless of impedance. \$\endgroup\$
    – Justme
    Feb 5, 2020 at 15:52
  • \$\begingroup\$ @Justme You are right about that. I am going to find where else I read about it. One other page that I know is a non-english forum, where somebody suggested using the 8Ω speakers over 4Ω for more voice clarity. I will post here if I could find its source. \$\endgroup\$
    – Sohail
    Feb 5, 2020 at 20:59

2 Answers 2

7
\$\begingroup\$
  1. With the higher impedance (8 ohm) speaker, you will deliver 3dB less power to the speaker than you would with a 4 ohm speaker. However, the efficiency of the speaker will likely have a much larger effect on the volume. In general, speakers with lower efficiency will have less distortion. Note that typical efficiency for speakers ranges from about 4% for loud, cheap ones to 0.2% for reasonably good audio equipment. (Of course, it's easy to make a speaker that's both distorted and inefficient, so it's not a hard rule.)

  2. Amplifiers will show higher distortion for lower impedance speakers because the higher currents required to drive them stress the drivers more, introducing higher levels of nonlinearity.

If you're looking to reproduce legible voice, you can easily handle the levels of distortion that you're looking at in this example; you probably wouldn't even notice it if you weren't listening for it. If you were looking to listen to music, my advice would be different.

\$\endgroup\$
3
  • \$\begingroup\$ I am confused with the "efficiency" you are talking about. I guess an efficiency of 4% would mean the device is wasting 96% of power! Please correct me. \$\endgroup\$
    – Sohail
    Feb 5, 2020 at 14:01
  • 1
    \$\begingroup\$ This is typical for a moving coil speaker. I don't blame you for being surprised to learn this, I was too. Yes, 96% or more of the power goes into heat. If you're ever designing a high-powered speaker, especially for continuous operation, you'll find that there's a power rating and a (generally more hidden) continuous power rating. You probably won't burn it with voice or music, due to the high crest factor, but if you're running tests with sine waves (for instance) you can easily burn it up. \$\endgroup\$ Feb 5, 2020 at 14:08
  • 1
    \$\begingroup\$ That is true, speakers are very inefficient converting electrical power to sound power. Which is why speaker efficiencies are in daily life measured in SPL output per watt typically, not how many watts of sound power is output. \$\endgroup\$
    – Justme
    Feb 5, 2020 at 16:01
3
\$\begingroup\$

The PAM8403 data sheet (link to datasheet) doesn't seem to contain anything to back up what you've read.

Look at the characteristic labelled "Total Harmonic Distortion Plus Noise". For the specified noise level, it can deliver twice as much power into a 4 ohm load as into an 8 ohm load. That implies you would be better off using a 4 ohm speaker and turning down the volume a bit, to reduce distortion further.

\$\endgroup\$
3
  • \$\begingroup\$ Thank you for pointing me to those charts in the datasheet. In the THD+N charts, the minimum THD+N for 4Ω speaker is 0.1% and for 8Ω is 0.08%, albeit on a much lower power. Also if compared on the same power (lower than max, like 1 watts), the difference is a little more visible. This amount of difference sounds too tiny to consider as a difference in sound quality, right? \$\endgroup\$
    – Sohail
    Feb 5, 2020 at 20:52
  • \$\begingroup\$ @Sohail unless you are operating the amplifier at maximum power, the distortion is likely to be tiny anyway. \$\endgroup\$
    – Simon B
    Feb 5, 2020 at 20:54
  • 1
    \$\begingroup\$ @Sohail Assuming you are using cheap speakers, the THD of the speaker is likely to be at least 100 times higher than the amp. The charts show the THD for the amp driving resistors, not speakers. \$\endgroup\$
    – alephzero
    Feb 6, 2020 at 0:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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