I had a soundbar for my computer audio and I finally got fed up with it. Half a need for better sound and half to acquire more speaker design knowledge, I made these. I used basic math only to calculate it's size and shape, its roughly .18 cubic feet with a 3 1/2 inch full range and a 90w piezo tweeter. I went with the Piezo because I wanted to make the design without a crossover. The piezo's work nicely, but they definitely introduce listener fatigue fairly quickly
I'm still breaking in the BMR's so I can fully comment on the sound, but so far I'm impressed. Please ignore the block of wood adding the tilt, it's temporary. This was made out of convince, that is the parts were all part of past projects. I would love to hear design critiques, no matter how harsh they may be.... I'm here to learn.
These measurements are for a general idea of the speakers capabilities, I used REW but with an uncalibrated mic
I'm still breaking in the BMR's so I can fully comment on the sound, but so far I'm impressed. Please ignore the block of wood adding the tilt, it's temporary. This was made out of convince, that is the parts were all part of past projects. I would love to hear design critiques, no matter how harsh they may be.... I'm here to learn.
These measurements are for a general idea of the speakers capabilities, I used REW but with an uncalibrated mic
Nice! You might want to reduce the tweeter level with a (rather high value, because of piezo tweeter, maybe 10-20 ohms) resistor.
And the woofer is not suited for a vented enclosure because of high qts. Is that a port in your picture?
You could stuff the port and add a series cap to the woofer for better low extension and linearity. Something between 270 and 560 uF (non polarized) maybe. If you have a simulation tool try to find a good value.
It may also be possible to use the vented box with a high pass cap, but that would definitely need simulation.
And the woofer is not suited for a vented enclosure because of high qts. Is that a port in your picture?
You could stuff the port and add a series cap to the woofer for better low extension and linearity. Something between 270 and 560 uF (non polarized) maybe. If you have a simulation tool try to find a good value.
It may also be possible to use the vented box with a high pass cap, but that would definitely need simulation.
Could you post a view of the frequency response where the vertical (dB scale) is approximately 50dB from bottom to top?
The "fatigue" could be from the 10dB boost at 6kHz, or the >12dB boost >10kHz. The piezo definitely needs attenuating.
Chris
The "fatigue" could be from the 10dB boost at 6kHz, or the >12dB boost >10kHz. The piezo definitely needs attenuating.
Chris
The frequency response looks fairly good for a piezo.
That doesn't do much. A Piezo acts like a capactor, if you put a resistor in series, the top end goes down more than the lower frequencies. To make it quieter, put a capacitor in series. If you want to actually cross them over, put a 20 Ohm resistor in parallel to the piezo, a xo will 'see' mainly the resistor. A 2µF capacitor is a good point to start.
You could also incorporate an adjustable L-pad on the piezo if you want to be able to dial the tweeter level up and down. This is often done by ear, so it's nice to have fine adjustments. That could be replaced with a fixed resistor L-pad ultimately if you want.
One approach here:
https://www.diyaudio.com/community/threads/piezo-vs-voice-coil-tweeters.3005/page-4
There are other things you can do to adjust the piezo's response, but I'd be inclined to just try the L-pad first and see if that's enough.
Your BMR looks like it has a rising response tendency on-axis in the upper frequency range also. Ultimately a simple crossover (single inductor on your BMR, and a circuit to attenuate the piezo) might help your speaker perform more consistently.
https://www.parts-express.com/pedocs/specs/297-2156--tectonic-tebm65c20f-8-spec-sheet.pdf
One approach here:
https://www.diyaudio.com/community/threads/piezo-vs-voice-coil-tweeters.3005/page-4
There are other things you can do to adjust the piezo's response, but I'd be inclined to just try the L-pad first and see if that's enough.
Your BMR looks like it has a rising response tendency on-axis in the upper frequency range also. Ultimately a simple crossover (single inductor on your BMR, and a circuit to attenuate the piezo) might help your speaker perform more consistently.
https://www.parts-express.com/pedocs/specs/297-2156--tectonic-tebm65c20f-8-spec-sheet.pdf
Last edited:
I forgot to mention that I have this connected to a yamaha avr and I have to turn the volume up almost to -10 for it to be somewhat loud. I'm guessing the 90w piezo is drawing all the power away. I do have some other "normal" tweeters, one pair is an 8 ohm and the other pair are 6ohms. I'm thinking about using those in the next design and that will require a crossover.
The sensitivity of your BMR driver is quite low (81 dB at 1 watt). That's more likely to be the source of the volume issue.
The piezo's wattage rating is just an indication of how much 8-ohm rated power is safe for it; it's not an indication of what it will actually draw. The impedance of it will be very high at lower frequencies, so the actual power draw will be minimal.
Example piezo impedance plot from here:
https://www.diyaudio.com/community/...super-tweeter-piezo-tweeter-revisited.399363/
The piezo's wattage rating is just an indication of how much 8-ohm rated power is safe for it; it's not an indication of what it will actually draw. The impedance of it will be very high at lower frequencies, so the actual power draw will be minimal.
Example piezo impedance plot from here:
https://www.diyaudio.com/community/...super-tweeter-piezo-tweeter-revisited.399363/
Last edited: