Who makes the lowest distortion pro-sound subwoofers?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2005
Paid Member
Forget about the driver. Think about the cabinet design. Folded horns are the only way to go.

See:
http://www.hometheatershack.com/forums/subwoofer-tests/980-diy-pa-horn-xls-12-a.html

115dB at 40Hz from a single Peerless XLS12 driver. Who'd thought it manage that?!


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


For music, dynamic integrity and low distortant in the bass and midbass is more important than hitting ultra low 10-20Hz.
 
tktran303 said:
Forget about the driver. Think about the cabinet design. Folded horns are the only way to go.

I agree that thinking about the cabinet design is in order, but I don't think folded horns are the best option in a home environment.

And while I like the idea of a XLS12 in a properly documented front loaded horn, the approach we are discussing here is far superior to the link you gave.

The DIY horn in that project just touches 115dB@40Hz at full power with the harmonic distortion surpassing 20% ( :hot: ), my simulation above shows 134dB@40Hz at full power from the same box volume! On the same note, 115dB SPL(max) would be achieved at 25Hz!

Derate this with 3dB for single speaker use (think: smaller box, comparable price) and another 3dB for power compression, yelding 128dB@40Hz when driven from the same amp (around 2kW).

This leaves us with a mere 13dB headroom (or -13dB distortion level, for the sake of simplicity) compared to the horn. And don't forget that the XLS12 isn't free to begin with, current price is just below 250 Euro with the horn unequally more difficult (and expensive) to build.

If we really want to improve upon what we got, I think a tapped horn (with an appropriate driver choice) would be the way to go. :bigeyes:
 
sek said:
If we really want to improve upon what we got, I think a tapped horn (with an appropriate driver choice) would be the way to go. :bigeyes:

Here is a 20hz TH for the 18N860. 700 liters and 125db max (in half space, 135db in eighth space) from 20-150hz. Grey graph is half space, black graph is eighth space. You would need to build two resonators in it to tame those peaks. You could make it somewhat smaller at the expense of efficiency.
 

Attachments

  • neu.jpg
    neu.jpg
    69.2 KB · Views: 763
Hi MaVo,

I take it the 125dB at 20Hz is excursion limited.

In my above simulations I incorporated a high pass filter (LR4), as this is best practice at the powers we're dealing with. ;)

Can you model a low cut of any sort with Hornresp?

Additionally, how much smaller could somewhat smaller probably be? ;)
We're talking roughly four times the box volume requirement here (compared to a single BR or BP)!
 
The limiting excursion peak in my simulation is around 30hz, so a highcut wont help much to raise max spl - besides, a highcut is mandatory for a tapped horn (20hz in this case). if you make a horn for a 30hz cutoff, the spl limit increases to maybe 130db while size will go down.

I tried the 12 inch bms, but didnt like it. it gets a peaky response easily. but with the right design, it could perform very well.
 
But please note that the closed alignment, delivering 125dB@40Hz and 120dB@30Hz, is at it's excursion limit.

The vented alignment has some considerable headroom here, safely delivering about 132dB@40Hz and 128dB@30Hz (with 5kW peak power!), see attached graph.

So putting a single 18n860 into your 200l box and giving it a vent would be the more efficient approach (read: half the price, twice the output!!!). :D
 

Attachments

  • bild 5.png
    bild 5.png
    20.5 KB · Views: 664
Ex-Moderator R.I.P.
Joined 2005
Looks nice, but you will loose sensitivity and its a bit low with 95db as is ... I dont play very loud so I guess a bit Eq will work ok

But you are spot on if the goal was around 93db sensitivity, I suppose

I will think about making holes in my matrix boxes :bawling:

I suppose running them double mono with single plateamp will work OK crossed around 60hz ... theres a new Hypex800 :D
 
tinitus said:
Looks nice, but you will loose sensitivity and its a bit low with 95db as is ...

Nope, sensitivity doesn't change with the number of speakers, as it is a property depending on input power. Twice the number of drivers, twice the power, hence the same sensitivity... ;)


I dont play very loud so I guess a bit Eq will work ok

LOL, you're funny. If you don't play loud, why bother with such expensive speakers anyway?


I will think about making holes in my matrix boxes

No, don't! I've supplied you with another graph, showing both closed and vented in comparison. Both single driver, high pass filtered, driven with the same power (2kW) at the respective excursion limit. The closed box fares pretty good (for a closed box, that is ;)), so leave it alone until you really miss the lowest registers. I predict that you won't! :D


I suppose running them double mono with single plateamp will work OK crossed around 60hz ... theres a new Hypex800 :D

Talking about the DS8.0? No, it won't! The DS8.0 delivers the single 18n860 with about 500W, costing a little above 500 Euro according to Hypex's website (which turns out to be a buck per watt).

As much as I hate to talk you away from a Hypex product, it will limit your max. SPL to 114dB@40Hz and 108dB@30Hz. Still very loud, but a speaker with half the excursion (and power) capability would suffice in that application. Why I'm telling you that? Because such speakers can be had at half the price, too!

Amplifiers like the Powersoft Digam Q or QSC Powerlight series (just to provide you with some examples) do that for the same relative cost (per watt), but are able to make actual use of the speaker's quality.

You could of course always try on a budget with two of these, but good luck with that one... :clown:
 

Attachments

  • bild 7.png
    bild 7.png
    20.3 KB · Views: 617
Well, 3dB difference in max. SPL means twice the power requirement. Thus, a speaker 3dB more powerful can mean one speaker less required. Remember that amplification is cheap compared to high quality speakers.

And besides that, nobody's talking about 500 bucks, seriously! ;)
 
You could ask a moderator to split off your part of this thread. It's actually two parts, that about the 18n860 and that about the AE speakers. Or you just open up another thread. :)

This way, thadman could get his topic back and you would generate more interest in yours. ;)

You could then start out by stating your design objectives, as you didn't make yourself clear regarding your requirements yet.

Regarding shipping the AE units, are they available anyway? I see no spec sheets, no price quotes and a lot of dead or void links on the website.

Cheers,
Sebastian.
 
sek said:


Nope, sensitivity doesn't change with the number of speakers, as it is a property depending on input power. Twice the number of drivers, twice the power, hence the same sensitivity... ;)

Wrong -- twice the number of drivers, input power is doubled (+3dB), output SPL is doubled (+6dB), so efficiency in dB/W is increased by 3dB.

This is a well known fact for all direct radiator loudspeakers.

Ian
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.