This is my first time building.
I got together some parts and ideas from advice in a post I made earlier and some builds I've seen. I've seen quite a few videos and builds using the ND65 and ND91 speakers and just today I saw this as well: https://www.parts-express.com/Passive-Aggressive-Mini-Speaker-Kit-Pair-300-7148?quantity=1 I remember seeing they are based on the Aurasound NS or something like that.
My goal is to try and build a decent small Bluetooth speaker to plug in my Qudelix 5k Bluetooth DAC/DSP equalizer. I'm still trying to decide if I want to go with the ND65 or the ND91, but im fairly dead set on doing one of the two, now. I have become aware that if I go with the ND91 I will have to build a much bigger enclosure, and my goal is to make this as small as reasonably possible. But I also want hz to go below 100, and as low as I can get it. I can tune it with the passive radiators and the Qudelix 5K's DSP, but I also need to make sure my enclosure is big enough. I've tried using https://speakerboxlite.com/subwoofer-box-calculator/tab-graphs and I think I have an idea of what I want to do. But I'm worried I did something wrong. And I think I might need to add an LPAD to bring the sensitivity of the tweeter down closer to ND65/91... but I also have the Equalizer, so will I need it? Can't I just mass-lower those frequencies? It's more expensive, but should I do a 3rd order crossover? I've also been thinking of getting the TPA3116 50w amp instead,.. but that seemed like a bit much.
Last edited:
The ND91 is much more capable and has higher sensitivity. It will do OK in around 1 liter with the proper tuning, so the "optimal" volume on this one can be taken with a grain of salt. I didn't look at speakerboxlite, because I already had the ND91-4 in Unibox, so I can't comment about what you saw there.
Regarding your EQ comments, keep in mind that a passive radiator design can't handle a lot of boost below the tuning frequency of the enclosure/driver/passive radiator (Fb). Excursion rises rapidly with decreasing frequency in that region, so it's very easy to run up against the speaker's limits.
As long as you don't run out of range, you can probably handle the sensitivity mismatch between the drivers with EQ.
Since it's your first project, I'd suggest not adding the complexity of the 18 dB/octave crossover, unless you have a specific reason to. Using "stock" crossover values that assume flat impedance and frequency response is sub-optimal also, but with enough EQ you can probably deal with its shortcomings that way.
Also, the Passive Aggressive crossover is in the build document. That might be a good place to start.
https://www.parts-express.com/pedocs/manuals/Passive Aggressive manual.pdf
Regarding your EQ comments, keep in mind that a passive radiator design can't handle a lot of boost below the tuning frequency of the enclosure/driver/passive radiator (Fb). Excursion rises rapidly with decreasing frequency in that region, so it's very easy to run up against the speaker's limits.
As long as you don't run out of range, you can probably handle the sensitivity mismatch between the drivers with EQ.
Since it's your first project, I'd suggest not adding the complexity of the 18 dB/octave crossover, unless you have a specific reason to. Using "stock" crossover values that assume flat impedance and frequency response is sub-optimal also, but with enough EQ you can probably deal with its shortcomings that way.
Also, the Passive Aggressive crossover is in the build document. That might be a good place to start.
https://www.parts-express.com/pedocs/manuals/Passive Aggressive manual.pdf
Last edited:
IT DOES!... and that's why despite the fact I want to make this as small as possible, it is still tempting me.
I will keep that in mind. Also, I can put up to a -24db cliff after a frequency if I need to clip it to protect something.
I'm fairly confident. I also want the highs to be a bit louder anyway. I'm a bit of a treble and sparkle fan.
I was afraid to, but willing. But yes, I think I agree. I won't do the 3rd or 4th order crossover.
I remember the uF and mH being QUITE different from the ones I got out of https://www.diyaudioandvideo.com/Calculator/SpeakerCrossover/
.... I'll look again and do some more research. THANKS! I really appreciate the help!
OH, WAIT! I just noticed... one last thing...
Do you mean 1 ND91 in a 1ltr enclosure, or can it be two?
Also, does that mean I can't make a chamber for each ND91 and give each it's own passive radiator?
If you want to use two ND91's, you'll need to roughly double the volume of the enclosure.
If you are using a mono signal for the woofers, you can put them both in one chamber. You can also put them in separate chambers if you want, but if you're trying to keep things small one chamber probably makes more sense.
If you are going to drive the two woofers with separate left and right signals, having them in separate chambers is optimal, but also not a big deal for the vast majority of music.
One other point about passive radiators: typical advice is to use enough of them or a large enough one that their volume displacement is twice that of the woofer they're used with. When you're trying to make things really small, that can be an issue as well.
If I misunderstood any of your points or didn't give you enough info, please ask more questions.
If you are using a mono signal for the woofers, you can put them both in one chamber. You can also put them in separate chambers if you want, but if you're trying to keep things small one chamber probably makes more sense.
If you are going to drive the two woofers with separate left and right signals, having them in separate chambers is optimal, but also not a big deal for the vast majority of music.
One other point about passive radiators: typical advice is to use enough of them or a large enough one that their volume displacement is twice that of the woofer they're used with. When you're trying to make things really small, that can be an issue as well.
If I misunderstood any of your points or didn't give you enough info, please ask more questions.
Apparently I lied about that being the last question... but i just noticed that too.
But then I realized that the ND91-4 had an Xmax of 4.6, and the ND90-PR had an "Xmech" of 9... not Xmax. I do not know what the difference is... theoretically if the SD is the same, but the Xmax is twice as much, then VD should be twice as much, yes?
I could do a mixing circuit, but that seems kinda rediculous. I could set my phone to mono, but it doesn't sound right...
I thought the PRs designed for this series, ND90-PR, were designed to be 1 for 1.
But then I realized that the ND91-4 had an Xmax of 4.6, and the ND90-PR had an "Xmech" of 9... not Xmax. I do not know what the difference is... theoretically if the SD is the same, but the Xmax is twice as much, then VD should be twice as much, yes?
Perhaps I just go mono then and have one chamber with one ND91, one Tweeter, and one (or two) PR. Yeah I dont have stero, but I'm willing to have a compromise somewhere. My JBL GO 3 I've had for years is mono. But how do I get a mono signal to/out of the amplifier? My Qudelix 5K doesn't seem to have a mono setting anymore...
I could do a mixing circuit, but that seems kinda rediculous. I could set my phone to mono, but it doesn't sound right...
I wasn't sure which passive radiators you were looking at, so just trying to cover all the bases.
Xmax on a driver is usually defined by voice coil length/magnetic gap geometry or by distortion level. It's normal for it to be less than Xmech (aka Xdamage), which is the physical limitation of the suspension or some other mechanical limit in the system. In some drivers there's not a lot of difference between the two, but in others Xmech can be multiples of Xmax. Similarly, some drivers play pretty cleanly well beyond Xmax, while others get muddy fast at higher excursion.
In a passive radiator they also use the term Xmax, but on some of the smaller Parts Express ones it tends to feel more like Xmech - you're really not going beyond that number.
Some of this comes down to your own expectations and the anticipated drive level. If you model things and a single passive radiator gives the performance you need, then that's good enough. If you want to push the driver beyond Xmax, you may need two ND90-Pr's.
I think some of the small amplifier boards include a summing feature. I don't have experience with ones that do, but that might be an option.
If you are going to be using a wired line level connection (RCA, headphone, etc.), you can put some isolating resistors on the L and R signal lines, then tie the two resistor outputs together to sum that way. 4.7 kΩ is a typical value.
Xmax on a driver is usually defined by voice coil length/magnetic gap geometry or by distortion level. It's normal for it to be less than Xmech (aka Xdamage), which is the physical limitation of the suspension or some other mechanical limit in the system. In some drivers there's not a lot of difference between the two, but in others Xmech can be multiples of Xmax. Similarly, some drivers play pretty cleanly well beyond Xmax, while others get muddy fast at higher excursion.
In a passive radiator they also use the term Xmax, but on some of the smaller Parts Express ones it tends to feel more like Xmech - you're really not going beyond that number.
Some of this comes down to your own expectations and the anticipated drive level. If you model things and a single passive radiator gives the performance you need, then that's good enough. If you want to push the driver beyond Xmax, you may need two ND90-Pr's.
I think some of the small amplifier boards include a summing feature. I don't have experience with ones that do, but that might be an option.
If you are going to be using a wired line level connection (RCA, headphone, etc.), you can put some isolating resistors on the L and R signal lines, then tie the two resistor outputs together to sum that way. 4.7 kΩ is a typical value.
Got a 3d printer?
Why not?
https://www.instructables.com/Mr-Speaker-3D-Printed-DSP-Portable-Speaker/
Why not?
https://www.instructables.com/Mr-Speaker-3D-Printed-DSP-Portable-Speaker/
Again, thank you. If i need to add a second, I can and will.
I am, but the recommendations I saw were two 1k ohm resistors. Is there a reason you'd use 4.7k instead?
You'll see a range of values suggested. There's little consistency in input and output impedances at line level, so almost any general recommendation is theoretically suboptimal for many scenarios. But the impact is typically small also unless you're running abnormal/tweaky hardware or you pick a crazy resistor value.
If you add another passive radiator later, watch your tuning. You have to add more mass to each of them when you increase the number of passive radiators. Adding a maximum of 2x the bass mass of the passive is the rule of thumb. Don't add more than you need to achieve your desired tuning frequency though. It can sap efficiency if added mass gets out of hand (frequency response can start to droop on the lower end).
If you add another passive radiator later, watch your tuning. You have to add more mass to each of them when you increase the number of passive radiators. Adding a maximum of 2x the bass mass of the passive is the rule of thumb. Don't add more than you need to achieve your desired tuning frequency though. It can sap efficiency if added mass gets out of hand (frequency response can start to droop on the lower end).