7
\$\begingroup\$

I recently dug out of my box o' stuff a pair of seemingly decent 2.5ohm bookshelf speakers. I wired them to an old mp3 player I have and while they sound good at low volumes, I get lots of distortion at high volume. Given that they are only 2.5ohm, I'm assuming that this is because they are trying to draw more current than the poor mp3 player can produce.

Rather than buying one (or new speakers for that matter), I'd like to try and build an amplifier for these speakers. I was looking at using an IC such as the TDA7924 but like all of these chips, the advised load impedance is 4 ohms. It doesn't specifically state anywhere that the absolute minimum is 4-8 ohms, but has anyone had any experience with this sort of thing? Will it still work at a lower load impedance if I just make sure that the volume is limited? If not, will I be able to use this IC if I chuck a big 2 ohm power resistor in series (as much of a waste of electricity as that is)?

I'll probably get some anyway to play with but I am a poor student and I would prefer not to blow up several $12 chips. thanks!

\$\endgroup\$
2
  • \$\begingroup\$ You only have 2 speakers, right? You can't wire multiple speakers in series per side? \$\endgroup\$
    – endolith
    Nov 5, 2010 at 12:52
  • \$\begingroup\$ you're correct, only 2. I understand this is not optimal but it is more of a learning experience then anything else \$\endgroup\$
    – jeremy
    Nov 5, 2010 at 13:29

4 Answers 4

4
\$\begingroup\$

If you don´t mind to lose stereo sound, you can wire up the speakers in series resulting in one load of 5 ohms.

I think that a 2 ohm resistor is not a good solution because a common output voltage will be arround 10 volts and with current of 5 amp you´ll need a resistor of 50 Watts, which is pretty expensive. Maybe buying an cheap tweeter will be better and improve high frequency sound quality.

Looking in the datasheet I found that this IC specifically (TDA7924) have an short-circuit and overload protection. Using a load of 2.5 ohms may not be a problem if you don´t exagerated in the volume.


Some other thoughts, how did you measured the speaker impedance? If it was only using an DC multimeter the number you got (2.5 ohms) is only the resistance of the speaker. The impedance also needs to consider the inductor. If the DC resistance is 2.5 ohms probably the total impedance is alredy 4 ohms, so you don´t need any modifications on the circuit. The speakers whose impedance are not 4 or 8 ohms are rarelly manufactured.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Thanks for your help! The resistance was measured across the two speaker terminals (including the crossover circuit) with my multimeter and is actually written on the box as 2.5 ohm. unfortunately rare speakers make for rare problems! \$\endgroup\$
    – jeremy
    Nov 5, 2010 at 10:57
4
\$\begingroup\$

You could try using an audio transformer to get a better match between the output impedance of the amp and your low-impedance speakers. Transformers match impedance by the square of their turns ratio, so if your device has an 8 ohm output, to match a 2.5 ohm load you'd want a ratio of sqrt(8/2.5), or about 1.8:1. Using a 2:1 transformer would make a 2.5 ohm load look like 4*2.5=10 ohms, which would still be a closer match to 8 than 2.5 is.

\$\endgroup\$
1
  • \$\begingroup\$ if i want to match my audio amp speakers outs (400 watt into 4 ohm, 200 watt into 8 ohm) to my microwave oven transformer (nominal 0.3 ohm DC and run at 240 volts 50hz) using 8 to 0.3, perhaps in order not to blow it up or catch fire I'll try a 5.2:1 ratio transformer...? that would be about perfect actually for my experiment: fractal wood burning using audio amp speaker outputs connect first to a 5.2:1 audio matching giving about 312 volts into my box of doom (from a speaker out of say 60 volts at full volume knob). see funk.nz/laser \$\endgroup\$
    – Tomachi
    Mar 7, 2020 at 5:15
3
\$\begingroup\$

Ahh!!! I remember doing one in my school days. I have few tips for you where you can still have these speakers work perfectly fine with the least risk ;) coz I tested them on mine. For reference I have used the datasheet from this site http://www.st.com/stonline/books/pdf/docs/1057.pdf

The logic here is the lower the resistance of speaker the higher current it will end up drawing from the amplifier, so what you have to take care is to make sure the current your speaker are drawing to keep it within safe operating range of the amplifier. Some tips you can use are

  1. Keep the input voltage much less that the MAX SOURCE VOLTAGE of the amplifier, in this case its 40V. You can use a 12V supply with 1AMP current.

  2. Make sure you have good HEAT SINK, a big one. preferably use the "heat sink paste" and screw it up tight and arrange it properly in the board such that it gets a good AIR CIRCULATION.

  3. And since the speaker are just 2.5 ohms from my experience they are going to be real loud and noisy at higher volumes (I re-winded the coils for higher ohms in my case and used custom cloth spider and cone to get the best out of my speakers ) So keep the volume low :)

\$\endgroup\$
2
\$\begingroup\$

I would suggest you do some research into Op Amps, as they will allow you to manipulate the output of the mp3 player into a more suitable input for your 2.5Ω speakers. You can in essence create a pre-amplifier for your speakers.

More information on Op Amps can be found at Wikipedia.

While your amp does not emphatically state itself to be an op amp, have you metered the impedance of the speakers? I found a link at http://www.prestonelectronics.com/audio/Impedance.htm, which states "a 4 ohm speaker will typically measure about 2.5 - 3 ohms, and an 8 ohm speaker will typically read about 5-6 ohms, while a 16 ohm speaker will measure around 12 ohms."

If you speakers actually meter to 2.5Ω and the typical speaker meters 2.5Ω also, then your speakers should work fine on the "op amp" you have selected then you may find you are able to utilize your amp.

\$\endgroup\$
1
  • \$\begingroup\$ michael, the part listed is an op amp (a very powerful one, but an op amp nonetheless). My question is more to do with the rating of high power amplifier chips and their empirical use rather than amplifier design. I am an EE student so I have a reasonable amount of knowledge of amplifiers already. \$\endgroup\$
    – jeremy
    Nov 5, 2010 at 10:55

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.