The "designing the perfect subwoofer driver" thread... (1 Viewer)

[Shawn Solar]

I've read through all the posta so far and tend to agree a managable enclosure(3ft^3) is way better than a 6ft^3 enclosure. A little bigger is ok but too big and its harder to fit two in the room
A problem I find with kilowatt amps is fan noise and the fear of flip'n the breaker. But if it is nessacary then ok.
I like the idea of using a passive radaitor as it will improve effiency and lessen the load on the amp. which brings me to the question. Is it possible to cut the amp power in half while at the same time keeping the same box and performance with one or two PRs? This would solve the problem of running two subs and lessen the added cost of amplifiers and having enough free circuts in the hometo run them.

 

[Brian Bunge]

I wish you guys were in my neck of the woods...

Vince,
ROAD TRIP!!! I can be there in less than 4 hours!!!
Brian

 

[Michael R Price]

I think we can all have what we want if we have 2 separate drivers. I still think that's a bit much to ask, but anyway...

One would be like the one Jack modeled (best for 3-5 ft^3 sealed with 1.5kW), the other would be like a double Tempest (for 8-12 ft^3 sealed/vented with more normal amplifier power like 250-800 watts[?]).

What's the formula for calculating anechoic sealed box output using only Vd?

I personally would go for the second driver option, btw. Some of us like the new 'holy grail' of big output from a small box. Some of us would like to push the typical DIY 'water heater' sub performance through the roof.

Also, what's the possibility of an Adire driver that would be good for IB use? High efficiency, low Fs, maybe an 18-incher? Well, shouldn't have asked, it's already amazing that Dan is designing a driver practically 'just for us'. Thanks, Adire.

 

[Vince Bray]

ROAD TRIP!!! I can be there in less than 4 hours!!!

bring it!!! let me know and we'll send the spl sensitive to a quieter movie

 

[Greg Monfort]

>At 20hz? I get 105db at 20hz with 750 watts in, with 1500 I get 108db, and with a whopping 2600 watts you can finally reach 110db at 20hz. You can hit 110db at 30hz with 750 in...

====

I get even less after factoring in VC heating. I'd be surprised if it did a real world 100dB/20Hz/750W/m measured outdoors, and as far as applying more power, you usually don't see much (if any) more output beyond its half power rating regardless of how much power it feels the need for.

There are some designs out there that have very little compression but you pay dearly for it and I thought this was about keeping driver cost reasonable.

====

>Anyway, I think what is needed is just plain more xmax!

====

If the driver can't handle/convert the power to drive it then what's the point?

GM

 

[DanWiggins]

Boy oh boy... Now you all know what it's like to design a driver - talk about tradeoffs, eh?
A few things to think about:
1. PRs/vents don't "improve" efficiency. They gain you a bit more output over a narrow band (typically an octave, at most) as compared to a sealed box, but other than that, you don't really get more efficiency.
2. Power compression is a real concern; we've gone a long ways towards reducing it on the cheap (and there's a few other things we have up our sleeve, too), but it still is something to concern yourself with. Small boxes playing deep means LOTS of power, which means power compression.
3. Ask yourself - do I really NEED 115 dB SPL anechoic from a single 3 cubic foot box? How about 105-108 dB SPL @ 20 Hz anechoic - that will get you to the 115+ range in-room. And really reduce the power requirements.
Anyway, there is some good info here! But we should start looking at narrowing down the "design". I'd recommend the following two decisions:
A: sealed or vented/PR?
B: Larger or smaller than 4 cubic feet?
Once those two are answered, it really starts anchoring a driver design. Yes, these are two of the toughest questions to ask, but until these are answered, it's nearly impossible to set anything else.
Lastly, as far as why have so much linear excursion? It's because it means VERY low THD, even if you aren't pushing the limits. Better to be moving 16mm one way on a driver rated for 30mm+, than 16mm one way on a driver rated for 18mm one way.
Dan Wiggins
Adire Audio

 

[Dustin B]

Hmm, I don't want to have to spend over $1000 on amps. So my vote is for larger than 4ft^3.

As for vented/sealed. I want to do a dual driver sub for my next one that can reach reference in room. If that can be done sealed, without ridiculous power requirements, I'll vote that way. Mainly because to get adequate porting for a pair of these drivers in a vented allignment will just require too large an enclosure.

 

[Anthony_Gomez]

110db at 20hz @87L:

sorry should have posted that this was with a pr/vented alightment tuned to 20-22hz (I forget what i used)

 

[Jeffrey Noel]

3. Ask yourself - do I really NEED 115 dB SPL anechoic from a single 3 cubic foot box? How about 105-108 dB SPL @ 20 Hz anechoic - that will get you to the 115+ range in-room. And really reduce the power requirements.

Dan, I would be extremely happy with this. Cutting down on the power is also a big plus for me since college is expensive enough.

Would this be sealed? That's what I'm hoping for; a killer sealed sub!!!

 

[Anthony_Gomez]

Well, the HS500 and the BPD1503 can do that in 3ft^3..however, you will need to EQ it.

...so DAN, The mission (IMO) is a 2-3ft^3 sealed box. subwoofer should be fully linear (in the flat BL region) in the HS500 power range. XBL2 can take care of the inductance peaks...and have the Fs low enough that the built in EQ of the HS500 can bring the room flatto 20hz! this simplifies things for many people. For the extra enthused, they can get a BFD for room modes.

 

[Greg P]

I am for a small enclosure (less than 3ft ^3). I dont mind using kilowatts. They are pretty cheap know a days. The QSC RMX 2400? Some where around 1100 watts a channel at 4ohms. Ive seen it all over the net for $550. Thats pretty damn cheap. Getting a lesser amp and 2-3 good PRs will about equal the price. The Crest LT-2000 can be had for a little over $650. Its a a more efficient class D amp.

Sealed. However I think the Brahma would fit my needs, but I bought the 1503 because it is probobly half the price of the proposed 15 brahma. Cant wait to model that Brahma 15

 

[Shawn Solar]

Having read dans reply on PRs, I vote for a sealed design. The huge ports needed for such high excursion drivers is really quite costly . I paid $50 just for 6" flanges and luckily found a nearby construction site that had a few left over 6" pipe, or else I would of been forced to to buy 10 ft of the stuff. But enough of my ramblings.

I think 108-110db at 20hz is really decent. Even with my 1503 in a ported 5ft^3 I get around 115db at 20hz. So 3ft^3 at a trade off of 5db is a fair trade off.

 

[Brian Bunge]

Dan,

A. Sealed

B. Under 4ft^3

As cheap as high power amps are getting this makes the most sense to me. Construction is easier also. I'd rather put a few extra hundred dollars more into amp power and have a lower Qtc/better transient response than put that money into PR's and have lesser transient response. I know for HT this shouldn't be an issue. But seeing as my system has to perform double duty I'd prefer sacrificing a bit of output in favor of sound quality.

Brian

 

[Greg Monfort]

>Lastly, as far as why have so much linear excursion? It's because it means VERY low THD, even if you aren't pushing the limits. Better to be moving 16mm one way on a driver rated for 30mm+, than 16mm one way on a driver rated for 18mm one way.
====
True, but the extra Vd has historically been expensive to make, and since the extra THD is down where it decays out before it becomes audible, I still don't see the point. I doubt there's anyone on this list more demanding WRT keeping distortion vanishingly low than me, but there's a point of diminishing returns below ~35Hz.
Now if you can make 60mm p-p with 10% distortion on the cheap, then count me in, I'll mass load them puppies up to get the most out of them in a dipole sub.
GM

 

[Jack Gilvey]

Now if you can make 60mm p-p with 10% distortion on the cheap, then count me in, I'll mass load them puppies up to get the most out of them in a dipole sub.

Greg, I need you to teach me about dipole subs now, Dan knows why.

 

[Randy G]

108-110 anechoic at 20hz is WAY more than satisfactory for me. My vote:
1) definitely sealed
2) small as possible
Watch ouuuut Velodyne...

 

[DanWiggins]

Greg Monfort posted:

Dipole, eh? How about this:
Sd: 1140 cm^2
Xmax: 35mm one way
Fs: 14.5 Hz
Qms: 24
Qes: 0.45
Vas: 360L
Re: 5.4 Ohms (two in parallel for a nominal 4 Ohm load)
Le: 4.3 mH
Motor-based distortion under 5% @ 30mm one way
Now THAT should work well for a dipole! A pair in a room should get you over 120 dB SPL @ 25 Hz. So reference levels from a dipole sub - that's not bad...
And in 10 cubic feet tuned to 14 Hz it's a monster. AND, in 5 cubic feet sealed it's a serious unit as well.
So Jack, Greg, others, there's one set of possible params. Whaddayathink?
Dan Wiggins
Adire Audio

 

[Jeffrey Noel]

What exactly are the benefits of a dipole sub? And what exactly a dipole sub? Would it be set up the same as regular dipole speakers?

TIA

 

[Jeremy_Wolf]

I'll throw in another vote for sealed as well. I am a huge fan of the linkwitz transform and would love to have a monster pe and xmax to create the uber linkwitz transformed sub. A sub that in a 5 ft^3 box that has a f3 of 10hz after the transform would be insane.

 

[Jeff Rosz]

hello jack,

in answer to your post about the LT...

a plain jane unity gain black box....even easier.

a modular design with swappable parts?....yes, that can be done with with sockets. also there are small metal female pins that can be soldered into holes on the pcb.

ok now i am curious. Dan, to what practical limits, high and low, can we take the 6 or so parameters you gave? i think i have come up with a nice lil sub but i just dont know if it can be done. i need to change a few things.

also, maybe this has been answered already, but are we talking dual 4 or 8 ohm coils? a Qes of .45 indicates 8ohm in parallel, yes?

thanks

jeff