Implementing the SW-12-16FR servo open baffle subwoofer driver

TRADERXFAN · « **Reply #20 on:** 23 May 2008, 10:22 pm »

Quote from: Danny on 23 May 2008, 06:32 pm

I was able to get a pretty flat response from 20Hz to 300Hz with no resonance at all.

Traderxfan,

Try applying that information to a box with a wider opening (22" tall by 13.5" wide) and only 13.5" deep and see what you get.

<I hesitate to post this out of fear of looking dumb. Mathematics is not my strength >

If I understand the equations, I get 250 hz -but I am prob doing this equation wrong. And as I said, the value for the calculation is supposed to be measured as the depth of the box from the face of the driver. The driver being tilted front to back, there is no representative value. If it is linear, then maybe you can take the midpoint, then it would be about double, 500 hz. I don't know how to incorporate the height. There is some calculations for a flat panel mount for that dimension, but this has depth so i assume its not usable to make a calculation, but it obviously has some other effects.

calculation is: F=0.25*v/L

v=343 m/s speed of sound
L= 13.5” or 0.3429 m depth of frame

So 0.25*343=85.75
Divided by .3429
=250.07 hz

Bottom line though... if you don't hear or measure a resonance, must not be there, or is compensated by something. Then this is all pointless.

Could someone help me understand the significance of the impedence plot. I take it that the impedence plot shows the resonance of the driver, and if there is no significant effect on the driver resonance frequency on the impedence plot when mounted in the baffle, then it means the baffle isn't having a detrimental resonance. Is that even close???

And I thought it was puzzling why there is no output dropoff as frequency gets lower as predicted too... Why would that be?

Thanks for your help and patience.

-Tony

Danny Richie · « **Reply #21 on:** 23 May 2008, 10:36 pm »

Well, I think you missed it, but don't worry about it. You don't have to be able to do the math to get it.

In his example: For a baffle of D = 20" (0.5 m) length and with L = 10" (0.25 m) estimated, the resonance peak in the dipole output is at F = 0.25*v/L = 343 Hz.

His baffle is 20" deep and the resonance was 343Hz.

If the depth gets shorter (like 13.5" deep) then the resonance gets higher. Also, if the height and width increase the amount of loading decreases. It is like the port stayed the same length but got bigger around. That also pushes the resonance higher up.

Quote

Could someone help me understand the significance of the impedence plot. I take it that the impedence plot shows the resonance of the driver, and if there is no significant effect on the driver resonance frequency on the impedence plot when mounted in the baffle, then it means the baffle isn't having a detrimental resonance. Is that even close???

Yep, you pretty much got it on that one, and that is why Bruno asked to see the impedance plot. If there was a resonance there then it should show up as a bump in the impedance.

TRADERXFAN · « **Reply #22 on:** 23 May 2008, 10:53 pm »

Thanks...

So if you took a single driver, mounted it vertically in the middle of a frame that was 13.5 x 13.5 x 13.5 " HWD, THEN that would give you the 500 hz resonance as predicted by the calculation, correct?

TRADERXFAN · « **Reply #23 on:** 23 May 2008, 10:57 pm »

And if I may go a little further on the impedence question, does the amplitude of the impedence increase at ~22hz mean anything? Was that amplitude equal with or without the baffle?

Danny Richie · « **Reply #24 on:** 23 May 2008, 11:02 pm »

I am leaving for a trip right now (5 minutes away) and am not going to take the time to do the math, but that looks about right.

The rise at 22Hz is a typical rise for the Fs of the driver. This will look almost exactly the same even if the driver were in free air. It looks great!

You guys have a good weekend.

BrunoB · « **Reply #25 on:** 24 May 2008, 05:55 pm »

Quote from: Danny on 23 May 2008, 10:36 pm

Well, I think you missed it, but don't worry about it. You don't have to be able to do the math to get it.

In his example: For a baffle of D = 20" (0.5 m) length and with L = 10" (0.25 m) estimated, the resonance peak in the dipole output is at F = 0.25*v/L = 343 Hz.

His baffle is 20" deep and the resonance was 343Hz.

If the depth gets shorter (like 13.5" deep) then the resonance gets higher. Also, if the height and width increase the amount of loading decreases. It is like the port stayed the same length but got bigger around. That also pushes the resonance higher up.

Quote
Could someone help me understand the significance of the impedence plot. I take it that the impedence plot shows the resonance of the driver, and if there is no significant effect on the driver resonance frequency on the impedence plot when mounted in the baffle, then it means the baffle isn't having a detrimental resonance. Is that even close???

Yep, you pretty much got it on that one, and that is why Bruno asked to see the impedance plot. If there was a resonance there then it should show up as a bump in the impedance.

Dany is right. His impedance plot is very clean. You can compare with my own dipole sub (ripole) impedance plot: there is a bump due to the 200 Hz resonance.

Bruno

TRADERXFAN · « **Reply #26 on:** 24 May 2008, 09:44 pm »

I wonder what impact the servo may be having....

I was reading the Martin J. King paper (http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf) on the open, u and h frame comparions for the eminence 15a woofer. He talks about the impedence spike at Fs moving to a lower frequency in the U and H frames, just as you did. His explanation:
"Assuming added mass is the only property contributed by the enclosure near the
system resonant frequency, the impact on the Eminence Alpha 15A Thiele / Small
parameters can also be estimated. These calculations are shown in Figure 4. The
conclusions drawn from Figure 4 are that the added mass drops the system resonant
frequency and the SPL/W/m while increasing the effective Qtd. This is consistent with the
results shown in Figure 1 for the three different dipole configurations and is consistent
with experiences adding weights to woofer cones for traditional speaker designs."

If the air cavity is like adding mass, and changing the TS parameters, but the servo system "virtualizes" the T-S parameters then maybe thats why the woofer acts like its in free air?

In that same paper, he showed there was a ripple on the impulse response for the H frame that wasn't present on the open baffle. I wonder if there would be one on Danny's or not?

Also it woudl be interested to see if the servo control could be disconnected, or run the box powered by a non servo amp, for a comparison of these measurement results.

Danny Richie · « **Reply #27 on:** 25 May 2008, 08:23 pm »

Quote

but the servo system "virtualizes" the T-S parameters then maybe thats why the woofer acts like its in free air?

Keep in mind that the servo amp is not connected to the woofers when making the impedance sweep. What it is, is what it really is.

nicksgem10s · « **Reply #28 on:** 28 May 2008, 10:38 pm »

Are drawings or plans available for the W frame box?

Do you have additional photos to share?

Is the servo amp in that box or is it separate from the enclosure with the subs?

Thanks,

-Nick

Danny Richie · « **Reply #29 on:** 28 May 2008, 11:42 pm »

I will have plans available.

I don't have additional photos to share right now.

And yes, the plate amp is separately housed.

WerTicus · « **Reply #30 on:** 29 May 2008, 04:50 am »

I'm going to use the H frame design i think... ie the driver baffle is right in the middle like a true H and using 3 drivers ill reverse one.
using this method the sub towers will be the same height as my main speakers, and i can add cloth to the front to make them look good.

Also H frames seem to give the lowest extension and least problems with resonances, and an even radiation pattern as far as my research has shown. Also simple to make.

WerTicus · « **Reply #31 on:** 1 Jun 2008, 10:18 am »

I've been playing with the H frame in max - basically with 3 woofers the H frame would end up only slightly larger than my main speakers. (WAR audio, reference 1 )

I'm going to play with other designs now, including a 3 woofer W frame design, oh and does anyone know the exact depth of this driver? I just made it up for now.

WerTicus · « **Reply #32 on:** 1 Jun 2008, 11:30 am »

The sub design in my room that I just did a pretty quick mock-up of - to see that it can actually fit

With couch walls and everything.... Behind the 4th camera it opens up to the rest of the house, which is a volume 3 times larger than the room itself.
Shown from a human perspective sitting at each end of the couch and the middle and then just one floating in space to give you a overview of how it will work.

WerTicus · « **Reply #33 on:** 1 Jun 2008, 12:37 pm »

Ill be able to give anyone who wants it cabinet design plans for any of these.

And any feed back / ideas anyone has would be most welcome

WerTicus · « **Reply #34 on:** 1 Jun 2008, 01:09 pm »

I gotta say the W pattern is pretty nice from an asthetic point of view

Here is a triple W design. Could easily put cloth in the front and it would look like an ordinary speaker.

WerTicus · « **Reply #35 on:** 1 Jun 2008, 01:21 pm »

This W is probably better with two firing down to the floor and only one firing up.

Daygloworange · « **Reply #36 on:** 1 Jun 2008, 03:01 pm »

Quote from: WerTicus on 1 Jun 2008, 10:18 am

does anyone know the exact depth of this driver? I just made it up for now.

WerTicus,

You can go to the RythmikAudio site and download the plan in PDF format, there is a drawing of the sealed version servo driver that looks pretty detailed. The driver should be very similar to the GR servo driver. Maybe email Brian Ding and ask him how faithful the dimensions are to the free air GR driver. I measured out the sealed version as 7.214" in depth, from the drawing. Should get you in the ball park.

http://www.rythmikaudio.com/download/DS12s_DIY_plans.pdf.

Building on Danny's sand filled sub, I quickly did up a 2 D dual (sealed box)version with roughly a 1.6 cu. ft. internal volume. With both the internal and external boxes in 3/4" MDF, with a 1" gap for sand, and a 1 1/2" thick front baffle, the dimensions are quite compact. 33 3/4" tall, 19" wide, 19 1/4" deep.

Cheers

WerTicus · « **Reply #37 on:** 1 Jun 2008, 03:47 pm »

Awesome, here is the w's with a quick mockup of the rythmic sub... till i can get hold of a similar plan for the GR driver.

FlorianO · « **Reply #38 on:** 1 Jun 2008, 08:26 pm »

Quote from: Danny on 28 May 2008, 11:42 pm

I will have plans available.

I don't have additional photos to share right now.

And yes, the plate amp is separately housed.

Danny,

A couple of quick questions:

1) Hand on heart: How high an XO are you confortable with when _selling_ this thing ? Brian mentioned (i.e. heartly recommended) an XO not higher than 150 Hz -- and that was for the 16 Ohm drivers. Which pretty much "sounds" (sic!) a reasonable and solid approach for an indep. "generic" bass unit for OBs. Moreover, the pics on the plate amps he posted show the XO up to 120Hz.
You claim clean output out to 300Hz.
So, what are the current plans ?

2) Avail: Any ETA ?

3) Shipments to Europe: Any issues with that ?

Regards,

Florian

Danny Richie · « **Reply #39 on:** 2 Jun 2008, 05:08 pm »

Quote

Hand on heart: How high an XO are you confortable with when _selling_ this thing ?

Despite the 120Hz crossover setting I have seen -6db down points in the 300Hz range using the 12db per octave setting.

Quote

Brian mentioned (i.e. heartly recommended) an XO not higher than 150 Hz -- and that was for the 16 Ohm drivers. Which pretty much "sounds" (sic!) a reasonable and solid approach for an indep. "generic" bass unit for OBs. Moreover, the pics on the plate amps he posted show the XO up to 120Hz. You claim clean output out to 300Hz.

You need to keep in mind that the crossover setting on the plate amp are in regards to an electrically controlled curve. It does not mean that is the acoustic output level in application.

Quote

So, what are the current plans ?

I will be offering the woofers and amps to the DIY market.

I will also be using them in some designs that will be offered as commercial speakers by some other companies.

Quote

2) Avail: Any ETA ?

Woofers and amp are in stock and ready to ship right now.

Quote

3) Shipments to Europe: Any issues with that ?

No problem. I ship stuff to Europe all the time. There are a number of carriers including the postal service that can deliver to Europe.