Maximize the Sound from a Buzzer

Question

I am working on a gizmo that will alert the user to an event using sound and light. It will be in a small enclosure. My objective is to make the buzzer as loud and annoying as possible, with an option to soften it. This is what I have to work with power wise: 5V from USB 3V logic from microcontroller gizmo will include a radio so want to avoid a switching boost converter I put together the following circuit on a breadboard which works more or less as expected: I don't need to vary pitch and the buzzer I just happened to have on hand is an active buzzer (that is to say it will buzz with only DC) which frees up a timer on the microcontroller. I have looked through the E14 offerings and 85 dB seems to be about the best I can expect at 5V. Before I blindly pick one and move on here are my observations and questions: 1) There is a sharp transition between loud and soft. It seems subjectively like it goes from a soft buzz to full throttle with a small change in R5 on the schematic. 2) Are there high efficiency buzzers I am missing or is the only way to get higher volume with higher power and thus, higher voltage. 3) I might consider an amplifier with speaker but it would have to be simple, sturdy, and compact. Thoughts and ideas are greatly appreciated! Frank 
 Attachments:

dougw · Answer

Piezo buzzers will provide higher sound pressure level with lower input power than magnetic buzzers. The input power is minimized and the output sound pressure level is maximized when the buzzer is operating at its resonant frequency.

You can get 125 dB with 200 mA if you really want it that loud.

A PUI (AI-5025-TWT-R) generates 115 dB with 75 mA at 12 V.

At 5V you can get 95 dB with 5 mA. (Mallory PT-3529PQ).

You can put several in parallel if you want louder sound.

dr-jhc · Answer

Another option could be to do something mechanical to provide better impedance-matching of the buzzer to the air: essentially some way to make the buzzer move air more efficiently. Some kind of waveguide (like you get for bass on ported music speakers) can work well. Your frequencies won't be low, so the waveguide won't be large.... but you need some way to design it. As dougw said, because your frequency is fixed, resonance will help you. Attaching the buzzer to another, larger, object that has the same resonant frequency can help you get more air moving (so higher SPL).

shabaz · Answer

Hi Frank, Interesting topic I've looked in the past for good buzzer circuits too, and it seems quite hard to find an off-the-shelf single chip to do this these days. One trick I recall was to use a schmitt hex inverter or quad NAND integrated circuit, and parallel up some of them for drive, and put the piezo buzzer across it, so that it is in parallel too, for doubled-up Vp-p. It would be possible to use one of the gates for an RC oscillator.

jw0752 · Answer

Hi Frank, I doubt if it will make much difference but you can squeeze an extra volt into the Buzzer if you use an N Channel MOSFET instead of the NPN Transistor. Depending on how you are driving the transistor it may work without modification or you may need a 10K resistor to pull the gate to ground. John

dr-jhc · Answer

Thanks for the link - that will keep people busy for a while! I'd start with a horn that looks like it might work reasonably well (based on that article) then 3D print some variations of these and test them. I'm sure you'd get something good enough quite quickly doing this.

jw0752 · Answer

Hi Frank, I have traditionally used an NTD 4906 for these applications as they are logic level and will saturate by 3 volts. I just set up a test with 5 volts on my load to the drain of an NTD4906 and I was able to get full saturation at 2.9 volts under a 750 mA load. There should be some equivalent N ch MOSFETS out there as these are listed as obsolete now. The trick will be to make sure they will handle the current of the buzzer and that they are specified as "logic level" or have a Vth low enough to work. John

jw0752 · Answer

I took a look at the BS170 data sheet and its Vgsth is slightly higher than the NTD4906 but the big short coming is that at full saturation it has a much higher DS resistance. John

Maximize the Sound from a Buzzer

Top Replies