Isobaric design for mini travel speaker?

[clengman]

I've been roaming around the forums here for a little while trying to glean advice for a small, battery-powered speaker I'd like to try to build.

Size is important. I'd like to be able to pack it in a suitcase for air travel or in a backpack for camping trips. We've been using an HMDX "Jam Classic" bluetooth speaker until now. It works, but it also sounds like a squawky, noisy, distorted, tiny little speaker. I'd like to make something slightly bigger with a properly designed enclosure that will hopefully sound a little] more like a real speaker, will give me useable frequency response down to 70Hz or so, and not sound so noisy and forced.

I found a Dayton Audio ND65 8 ohm driver that seemed like it would be a good choice. I was originally going to use the single driver in a ~45 c.i. vented enclosure and response would look like first screenshot.

Then I read around a little more about small speaker design and played around with the isobaric option in WinISD. With two drivers cone to cone I can make a speaker roughly the same size with freq response as in second screenshot. Downsides are higher power consumption at similar volume and the added weight of the additional driver.

So my first question is, do you think there will be enough of an advantage in bass extension to make this worthwhile? I see what it looks like in the software, but I have little experience translating this to the real world.

Second, are there any drawbacks to this type of design that I'm missing?

 

[turk 182]

i may be off base here but is an isobaric design ok for fullrange applications? i know they work for low frequencies but i don't know what the upper end response would be like.

 

[clengman]

You may have a point. I have no idea.

I think isobaric is usually used for small subwoofers, and often with a bandpass type of enclosure. In both cases high end response would not be important. I'm not sure what happens if you put an isobaric driver in a conventional BR enclosure.

Thanks for the input!

 

[turk 182]

WinISD is ok for looking at the low end but leaves one in the dark for the higher end.

 

[picowallspeaker]

Buy the two speakers then make your experiments

...consider that if you use them for bi-channel listening as in the norm
you'd get +6dB from dual sources

 

[clengman]

Buy the two speakers then make your experiments

Yes, if I buy two drivers I'll try probably try both ways. However, if there was some theoretical reason why an isobaric driver pair would never work in this application, I would buy just the single driver.

...consider that if you use them for bi-channel listening as in the norm
you'd get +6dB from dual sources

Yes, but the only reason to buy the second driver would be to trade efficiency for better bass extension in a small enclosure. If I used two drivers in stereo I would indeed get more volume for the power consumed, but I'd also have to double the size of the boombox to get the same bass response as I would from one driver. Or compared to isobaric driver pair I'd have to double the size and the bass response still wouldn't be as good. That's the opposite of what I intended for this speaker.

 

[picowallspeaker]

If I used two drivers in stereo I would indeed get

..what the artist/producer wants to be broaded !

But your assumptions are right; still nobody has tried it with small drivers.
The fact is that you add the mass of air's inertia to the movement of a cone so it is in some way retarded...
From an energy saving POV it should be avoided since you gain nothing at the expense of one more coil to drive. Subwoofer is far more clever as it is specifically designed for it, to produce an extra octave to little speakers

 

[chris661]

Isobaric face-to-face won't sound good for full-range use.
However, you can mount the speakers so that the outer one faces outwards, and the inner one also faces outwards. A small chamber would be needed to keep the cones acoustically coupled.
It's probably be worth low-passing the inner driver to avoid comb filtering through the kHz range.

Chris

 

[Scottmoose]

Last I checked, there was no such thing as an 'isobaric driver' per se, just drive units which are more or less well suited to it depending on application.

Otherwise, what Chris said. Face to face is out; you won't have any midband, let alone HF. The twin coupled type, both facing forward, one mounted internally behind the other a la the Linn Isobarik is viable. As a general ROT give it a minimum of about 3x Xmech spacing between motor and the forward most edge of the rear driver for safety. Re combing & low-passing the rear unit, there should be enough of a blocking impedance (as well as physical blocking) with the motor & the stiff air-mass in the small chamber between them to avoid the worst effects, although it certainly wouldn't hurt. Acoustically the downsides of isobaric loading of differential heating, driver variability etc. tends to cancel out many of the nominal advantages, so the main gains are reduced cabinet size for a given alignment, although the internal driver & chamber between the units does eat into that somewhat.

 

[planet10]

If you want to build a small enclosure, isobarik is not the way to do it. Better to choose a driver that fits into a small enclosure.

The overhead to put together a small FR isobarik can take up as much volume as the box. And that does not factor in the FR ripple caused by the coupling chamber depth.

For my small suitcase system i chose the 2.5 litre uMar-Ken for the Alpair 6.2p. I have drawn up 1.5 litre milliOnkens for Alpair 5.2 and FF85wk.

And if you are willing to give up some bass & top and have a biggish budget, the ScanSpeak 10F fits in an optimal 0.8 litre miniOnken.

dave

 

[clengman]

Last I checked, there was no such thing as an 'isobaric driver' per se, just drive units which are more or less well suited to it depending on application.

When I referred to an "isobaric driver" I was thinking of the two drivers, electrically and pneumatically coupled, as a single unit.

Otherwise, what Chris said. Face to face is out; you won't have any midband, let alone HF. The twin coupled type, both facing forward, one mounted internally behind the other a la the Linn Isobarik is viable. As a general ROT give it a minimum of about 3x Xmech spacing between motor and the forward most edge of the rear driver for safety. Re combing & low-passing the rear unit, there should be enough of a blocking impedance (as well as physical blocking) with the motor & the stiff air-mass in the small chamber between them to avoid the worst effects, although it certainly wouldn't hurt. Acoustically the downsides of isobaric loading of differential heating, driver variability etc. tends to cancel out many of the nominal advantages, so the main gains are reduced cabinet size for a given alignment, although the internal driver & chamber between the units does eat into that somewhat.

Thanks for this. I think I'll get two of the drivers to play around with just to see what I can come up with.

My original thought was to mount the two drivers magnet to magnet inside a piece of PVC tubing, then glue that unit into a hole in the front baffle. As I was doing some sketching, though I thought of a way to put everything (electronics and battery included) into a very small package. It really simplified the construction as well, but it requires mounting the drivers cone to cone. It's a shame that won't work, but I might be able to make a magnet-to-magnet work almost as well. Depends what type/size pipe I can find and how well the drivers will fit into it.

 

[clengman]

If you want to build a small enclosure, isobarik is not the way to do it. Better to choose a driver that fits into a small enclosure.

The overhead to put together a small FR isobarik can take up as much volume as the box. And that does not factor in the FR ripple caused by the coupling chamber depth.

For my small suitcase system i chose the 2.5 litre uMar-Ken for the Alpair 6.2p. I have drawn up 1.5 litre milliOnkens for Alpair 5.2 and FF85wk.

And if you are willing to give up some bass & top and have a biggish budget, the ScanSpeak 10F fits in an optimal 0.8 litre miniOnken.

dave

I'm trying to go quite a bit smaller (and less expensive) than these. I'm using a Dayton Audio ND65-8. According to WinISD it will work well in a 45 c.i. ported enclosure tuned to ~75Hz. It'll make some sound down to around 70Hz.

If I was to put that driver in a 90 ci box tuned to 70Hz I'd get a little peak in the 70-75Hz range and a little more extension below that.

If I do the isobaric trick, I can get that same response with my original 45 c.i. enclosure. I realize I have to add the volume of the extra driver and the airspace between the two drivers, but if I can find a good piece of ~2.5 plastic pipe, it should only add around 7 c.i. I still save a considerable amount of volume.

I don't know what FR ripple is. I'll have to look that up.

Thanks for the thoughts. I'm enjoying thinking all this through and learning as I go.

 

[planet10]

I'm using a Dayton Audio ND65-8. According to WinISD it will work well in a 45 c.i. ported enclosure tuned to ~75Hz. It'll make some sound down to around 70Hz.

0.74 litre. Pretty close. To go that low in such a small box, it is not going to be very efficient.

And the required vent might be a significant factor in the size.

I don't know what FR ripple is.

Frequency response ripple, caused by early reflections from the coupling chamber back thru the cone.

dave

 

[clengman]

0.74 litre. Pretty close.

Yes. I should have said much smaller OR much less expensive. I don't think I can drop the bucks for that scanspeak driver.

To go that low in such a small box, it is not going to be very efficient.

It's true. It's a very low sensitivity driver. and isobaric pair will be even lower, but I don't need it to be loud. I'm planning to get the drivers and amp (TPA3118 PBTL amp) and test with some AA batteries to see where the voltage needs to be to get sufficient volume. I may readjust my priorities at that point.

And the required vent might be a significant factor in the size.

I wasn't sure if the vent volume would have to be subtracted from the enclosure volume or not. Sounds like you are saying it does.

but I was planning to use a small diameter port (otherwise it will have to be far too long) so vent volume should be less than 1 ci.

but I'm also considering a passive radiator. Just need to do a considerable amount more research to understand how to select and tune them.

Frequency response ripple, caused by early reflections from the coupling chamber back thru the cone.

dave

How does this manifest in the sound of the speaker? interference and notchy frequency response?

 

[planet10]

Vent volume does add to box volume. You don't want to make it too small or it may chuff.

I did a miniOnken in the same volume as the SS 10F, i had to double the vent back an extra time to get enuff length while maintaining a reasonable cross-section.

dave

PS: the ND65 is an Aura OEM driver. As a generalization, the small Aura tend to go quite low for their size.

 

[astrojet]

I built these with Peerless 830984 2.5" Full Range 8 ohm for $25 from Madisound for my travel speakers. Sealed and very small volume. I use them with an elekit mini tube amp or a fostex 5w job and am completely happy for hotel sound. When properly travelling, I power the amp with a small LiPo battery.

 

[Greg B]

If I do the isobaric trick, I can get that same response with my original 45 c.i. enclosure. I realize I have to add the volume of the extra driver and the airspace between the two drivers, but if I can find a good piece of ~2.5 plastic pipe, it should only add around 7 c.i.

Maybe if both facing the same direction, I think a bit more. A quick back of envelope calculation tells me it's more like 20 cubic inches if you run them back to back, which is optimum. (cancels out some distortion)

If you've got a hankering to try this, go right ahead. Might be a fun experiment.

Still, I tend to agree with the others that it doesn't really make sense for a mini FR speaker. You are better off simply using a driver with a lot of excursion, putting it in a small sealed box, and EQ'ing your way to happiness.

Isobaric is a useful trick, but IMO more suited for things like domestic subwoofers, where you could make what would be an 8 cu ft box 4 cu ft instead, and get a somewhat cleaner sound.

 

[chris661]

I'd use a port/PR for something small and portable - the extra help in the LF is nothing to be sniffed at. You can EQ ported boxes down to the port tuning frequency, so even one with a response that slopes down can be made flat, using less cone excursion than a sealed design of the same LF cutoff.

Chris

 

[clengman]

I'd use a port/PR for something small and portable - the extra help in the LF is nothing to be sniffed at. You can EQ ported boxes down to the port tuning frequency, so even one with a response that slopes down can be made flat, using less cone excursion than a sealed design of the same LF cutoff.

Chris

I'm not finding a lot of PR that are an appropriate size for a small speaker. But these things: 2pcs 60 90mm Low Frequency Radiator Vibration Plate Bass Passive Speaker DIY | eBay

are available on ebay. Can anyone comment? There's very little information in the listing. From the picture it looks like the "effective area" is in the neighborhood of 30-35 cm^2. Listing says weight is 38g.

I was looking at an online calculator and if I assume that those numbers are correct, if I put this radiator in the box that I had originally planned (0.74l) it will have Fs of 36Hz. To get to the box tuning that I had originally planned (70Hz) the plate would have to be much lighter, just about 10g.

I'm not sure what my question is here. I'm just trying to organize my thoughts and try to figure out how I might be able use a PR effectively in my design. I appreciate any thoughts.

 

[chris661]

Those would be perfect. I don't think there's enough data given to calculate the PR resonance, but they almost always need to have mass added.

Once assembled, I'd run a loud-ish sine sweep, and see where the main cones stop moving. That indicates your PR tuning. Then you can adjust the PR mass to get it where you want it to be. If it really needs to be lighter, sanding away some of the PR surface might be safe, depending on the material. Alternatively, a silicone-based glue could be applied to the surround of the PR to reduce compliance, also pushing the resonance up.

Chris