Candidates for ELF performance?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I want to build a small sub i can use in my bedroom, but want to build something thats powered with an ELF style response, preferably a small sealed enclosure.

I dont have anything really specified yet, i'd like to keep it small, but make it powerful enough to have output and performance like the velodyne DD series.

I have yet to build an amplifier and circuit for it, because i want to make sure i get the first part right, being the driver and enclosure combination!

Does anyone have ideas on what drivers would work well for something like this?
 
I want to build a small sub i can use in my bedroom, but want to build something thats powered with an ELF style response, preferably a small sealed enclosure.

I dont have anything really specified yet, i'd like to keep it small, but make it powerful enough to have output and performance like the velodyne DD series.

I have yet to build an amplifier and circuit for it, because i want to make sure i get the first part right, being the driver and enclosure combination!

Does anyone have ideas on what drivers would work well for something like this?
A high power (>1000 watt), large Xmax (>18mm), low Fs (<25) driver with a 30-40 pound ceramic magnet structure will get you in to a similar response ballpark to the Velodyne DD series.
The Velodyne DD specs don't mention SPL levels, just equalized frequency response of +/- 3dB 18.7 -120 Hz for the 10" to 14.4 -120 for the 18".

Those specs can be reached by many drivers with lesser specifications at lower levels.

A DBX DriveRackPa can do most of the electronic functions, as could Mini Dsp.

The SpeakerPower Torpedo SP-4000 would be a great amp choice if you opt for very high power, it does an honest 4000 watts at 2 ohms, has a good sounding limiter, and a built in crossover.

The VLF SPL level you want will determine the cost of your project.
 
Disabled Account
Joined 2008
If you want to build a true ELF, you must put the system resonance above your XO frequency. That means drivers with high Fs and very small boxes, which again means you need lots of power and drivers than can handle it.

I made a pair that was very good with two Beyma 12P1000nd woofers in a 35 liter cube. I put one driver on each side. DCX2496 for XO and EQ, and 2x250W class-D power. Works very well.
 

Attachments

  • DSC_8614.jpg
    DSC_8614.jpg
    103.7 KB · Views: 256
If you want to build a true ELF, you must put the system resonance above your XO frequency. That means drivers with high Fs and very small boxes, which again means you need lots of power and drivers than can handle it.
There are many ways to achieve extended low frequency, the approach you (and Bag End) employ is one of them.
High Fs drivers typically don’t have much Xmax, so have much less LF output potential than similar size boxes using low FS, high Xmax drivers.
 
So what would you use for low to moderate listening levels? And for the bedroom, I feel like 25 Hz is a reasonable target?
“Moderate level” is a relative term, someone used to 140 dB at 25 Hz in their car might think 120 dB of LF is moderate.

I am using an 8” woofer in my kitchen, it can put out around 110 dB (but not down to 25 Hz) at 1 meter before my tuner sub output clips.

A 10” Tang Band WT-1427G would probably be adequate for your bedroom use, you can compare specifications of multiple woofers easily here:

http://www.parts-express.com/resources/woofer-selection-guide.cfm
 
Its possible to find, but they might not be cheap. The Beyma drivers I used has got 16 mm p-p Xmax, which is fairly ok I think.
Displacement capability (Vd) determines low frequency SPL potential.

The Beyma 12P1000nd currently cost around $379 for 8mm Xmax, a Vd of 440 cm3, 86 cents per cubic centimeter displacement.

An Eminence Lab 12 costs around $165 with 13 mm Xmax, a Vd of 658.7 cm3, only 25 cents per cubic centimeter displacement.

The Lab 12 is by no means the lowest cost per cubic centimeter displacement, I just happen to have owned many of them and know they sound good and have very low distortion.

You picked a rather expensive way to achieve low frequency response.
 
(Not really relevant but the Eminence Lab12 is not able to reach its Xmax, since 10mm seems to be its physical limit)
I have frequently measured excursions of 13 mm or more (26mm peak to peak) in tests with my Lab 12s.
In fact, the grill cloth is set slightly more than13mm from the surround in my ported cabinets so that when the grill starts to flap I know it is time to back down a bit.

Distortion goes up rapidly above 26mm P to P, but stays under 10% below.
 
I have frequently measured excursions of 13 mm or more (26mm peak to peak) in tests with my Lab 12s.
In fact, the grill cloth is set slightly more than13mm from the surround in my ported cabinets so that when the grill starts to flap I know it is time to back down a bit.

Distortion goes up rapidly above 26mm P to P, but stays under 10% below.
Hi Art,

Did you make measurements (preferably plots) when the cone was reaching 13mm?
 
So a low Fs, high power, high excursion driver, built with a high resonance, is the combination I'm looking for to have an excellent subwoofer? As long as my amplification is applied correctly?

A high system (woofer+box) resonance would be needed. So a small box must be used to put the resonant frequency of the driver up past the top of the operating range of the subwoofer.

That way, you're using the ELF principle rather than a Linkwitz Transform.

Also, use more than a pair of 8" drivers. This was my first subwoofer, and normal listening levels were fine, but going any louder meant the drivers went clank fairly quickly. Simmed f3 was 28Hz.

I'd say a 12" driver with a healthy Xmax (such as the Lab12) would be fine.
Just be sure to add some sort of excursion limiter, limiting amplifier output according to frequency - that way, you can turn it up as loud as you like, without being worried about the driver being damaged.

Chris
 
Hi Art,

Did you make measurements (preferably plots) when the cone was reaching 13mm?

Yes, examples are in this thread:

http://www.diyaudio.com/forums/subwoofers/191833-push-pull-vs-normal-distortion-compared.html

The 20 Hz full power tests got the cones dancing quite a bit.
The suspension does seem to keep damage from happening, though the cones were obviously “flapping”, there was no voice coil “clacking”.

Amplifier power is able to push the Lab 12 cone further than I can by hand, have not seen the cone move 44mm P to P (Xlim is 22mm) though I have seen at least 35mm Pto P when doing some testing.

Scared me at first, then I got the BC 18SW115-4 and they made the Lab 12 excursion look small.
 
Hi Art,

Measuring below the Fb and/or Fs doesn't show true Xmax or linear excursion + acceptable THD figures. Drivers can pass these points easy especially when driven below their Fs in free air or below the Fb of the system. In other words they can 'dance' above the point where the cone, suspension or motor looses their control over the excursion even if it doesn't look that way. Often it doesn't look that way when powered with sine waves. This 'over excursion' will not result in extra dB's but in dynamic power compression (losses) but doesn't necessary have to damage the driver.

For measuring true Xmax you will need advanced equipment. More practical for us; use a virtual model to compare the difference between max SPL measured and max SPL calculated. If you use the measured max SPL as input for your model, it can calculate the true Xmax figure.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.