DIY Sub (Project Slopzilla) in progress; Can someone check my math?

[bcycle]

Hi There,

So I have begun DIY sub project. I have ordered some parts including A DAYTON RSS315HF-4 12" HIGH FIDELITY SUBWOOFER (295-464) and a DAYTON SA240 240W SUBWOOFER AMPLIFIER (300-804 ) from Parts Express. Originally, I was going to use Dayton's 15" Quatro driver but that would have required a 6 cu.ft. box. The one I ordered is supposedly a higher end driver allowing for a smaller box one is (I'm thinking 4 cu.ft.) My plan is to build a box around 25"H x 18"W x 22"D with some simple bracing and a 4" flared port, locating the driver on the front of the box. I used some software to determine that this one models F3 around 21.15Hz which for my needs should be just dandy. The tech guy at Parts Express also suggested I "use as much damping (fill) as possible (without blocking the path from the rear of the woofer to the port). This will decrease group delay (boominess)". But, after adding more stuffing, the F3 jumps up to about 27hz! Now I don't know if my old ears can even hear 27hz, but I'm wondering if this is the right driver for the money. On the other hand, I'm still pretty close to my goal of spending around $200.00 for parts to build this thing.

My goals is to build a Sub that's not to big, sounds good for music and HT. I won't be playing at ear splitting volumes but pretty llou from time to time. If you had to configure these parts, what would you do? Would you use different parts keeping in mind the budget? Any other suggestions are welcome!!

Thanks,

Greg

Driver Specs:
Specifications: *Power handling: 400 watts RMS/700 watts max *VCdia: 2-1/2" *Le: .95 mH *Impedance: 4 ohms *Re: 3.3 ohms *Frequency range: 23 - 1,000 Hz *Fs: 23 Hz *SPL: 89 dB 2.83 V/1m *Vas: 3.00 cu. ft. *Qms: 3.00 *Qes: .52 *Qts: .44 *Xmax: 14mm *Dimensions: A: 12-3/8", B: 11-1/8", C: 5-3/8".

 

[WmAx]

Hi There,

The tech guy at Parts Express also suggested I "use as much damping (fill) as possible (without blocking the path from the rear of the woofer to the port). This will decrease group delay (boominess)".

Bad advice from the Parts Express guy. If you do as he instructed, you will reduce the Qa parameter factor by a significant amount, thus reducing the efficiency of energy transfer to the port, thus compromising the efficiency and low frequency response boost that a ported system would otherwise offer. The stuffing in a vented system should be used to control the resonances in that system related to the midrange(in a full range system with woofers that also extend into the midrange) or upper bass response(if it is a very big box). You should never fill a vented box with any sort of stuffing material.

Group delay is over-hyped, and I dare that Parts Express guy to point you to a credible perceptual study that demonstrates substatial audibility of group dealy as a lone variable in the amounts found in a properly designed ported alignment. The primary factor in audibility here is frequency response of the total system(that includes the subwoofer + room acoustics). You can go so far as to make your ported system sound like any sealed alignment that you wish, if you have a low enough F3 in the first place, and then use a powerful parametric equalizer to shape the final response to match a target curve.

-Chris

 

[bcycle]

Interesting article here http://www.diysubwoofers.org/talkshop/messages/41415.htm making a case to overstuff the box. Now I'm completely confused!

Bad advice from the Parts Express guy. If you do as he instructed, you will reduce the Qa parameter factor by a significant amount, thus reducing the efficiency of energy transfer to the port, thus compromising the efficiency and low frequency response boost that a ported system would otherwise offer. The stuffing in a vented system should be used to control the resonances in that system related to the midrange(in a full range system with woofers that also extend into the midrange) or upper bass response(if it is a very big box). You should never fill a vented box with any sort of stuffing material.

Group delay is over-hyped, and I dare that Parts Express guy to point you to a credible perceptual study that demonstrates substatial audibility of group dealy as a lone variable in the amounts found in a properly designed ported alignment. The primary factor in audibility here is frequency response of the total system(that includes the subwoofer + room acoustics). You can go so far as to make your ported system sound like any sealed alignment that you wish, if you have a low enough F3 in the first place, and then use a powerful parametric equalizer to shape the final response to match a target curve.

-Chris

 

[WmAx]

Interesting article here http://www.diysubwoofers.org/talkshop/messages/41415.htm making a case to overstuff the box. Now I'm completely confused!

I remember reading the original article. It's true that resonant frequency is lowered, therefor the impedance point, just as the article states. But the author of the article ignored other issues. But in the article you linked, he stated that the lower frequency impedance peak was damped when the box was stuffed. This should have raised suspicsion; if you reduce the resonance magnitude, you decrease the impedance peak of that resonance. In this case, the resonance is the port output.The article ignores the critical Qa parameter, which is part of Qb, which relates to low frequency efficiency of a vented system. Please read details concerning these parameters here:

http://www.geocities.com/kreskovs/Box-1.html

-Chris

 

[bcycle]

Questions for WmAx

WmAx (and anyone else who can help):

I'm liking your angle on box design and I'm hoping I can appeal to you with some specific questions. I should mention also I'm fiddling with these software package simply to establish a baseline but also trying to learn enough through reading, forums, etc. to come up with a design. I'm on the front end of the learning curve here but ultimately I want build the darn thing and soon! Since I'm certain my first try ain't going to be perfect, I have a few other resources to help tweak the sound:

1. My Yamaha RX-V2500 has a pretty good auto-setup parametric EQ.
2. There's tweakabilty in a DIY subwoofer
3. I've had years of exposure to some good stuff including B&W Mtrx 801, Martin Logan, Quad, Bryston, etc. (sadly, I could never afford it!). I have some perspective and trust my ears to fine tune things
4. My current setup is fairly humble (thousand dollar speakers, thousand dollar receiver, source music is mostly off the computer) so and I don't expect it to reveal flaws like some of the higher end stuff will.


QUESTIONS:

1. WinISD shows a group delay over 27 ms. I know you think group delay is over hyped. 27ms seems like a lot. What can I expect from that figure. How audible is it at the lower frequencies. (no way I can hear 20hz anyway)

2. What do you think about the box size?

3. The parts that I ordered are the Dayton RSS315HF-4 12" Woofer (295-464) and their SA240 Plate Map (300-804) Do these parts look okay for this design?

4. No stuffing?? Nothing? How about bracing. Any other construction advice?

Below, I've listed specific data from WinISD.
For the driver in an unstuffed box, WinISD suggests the following:

Number of drivers : 1
Box type : Vented
Box size : 2.917 ft^3
Tuning frequency : 23.20 Hz
Vent : 1 vent(s)
18.42 in length for each
4.02 in round

I've attached some graphs n' charts from WinISD which might help. The parts are coming today so I'm hoping you're not going to say "those parts won't work you idiot". Also, I'm not expecting flat down to 20 hz. I don't need the bass to interfere with my breathing. I'm not trying for concert volume levels; just a clean rich bottom end to augment my reasonably full (down to about 35hz) front speakers.

Thank You!

Greg



Some Specs:

Driver spec:

Driver : 295-464 dAYTON rss hf
Vas : 3.000
Qts : 0.40
Fs : 23.00
SPL : 89.00



Relative
Freq Gain SPL
[Hz] [dB] [dB]
20.00 -6.10 82.90
25.00 -1.61 87.39
30.00 -0.37 88.63
35.00 -0.13 88.87
40.00 -0.08 88.92
45.00 -0.08 88.92
50.00 -0.07 88.93
55.00 -0.07 88.93
60.00 -0.07 88.93
65.00 -0.06 88.94
70.00 -0.06 88.94
75.00 -0.05 88.95
80.00 -0.05 88.95
85.00 -0.04 88.96
90.00 -0.04 88.96
95.00 -0.04 88.96
100.00 -0.03 88.97

Bad advice from the Parts Express guy. If you do as he instructed, you will reduce the Qa parameter factor by a significant amount, thus reducing the efficiency of energy transfer to the port, thus compromising the efficiency and low frequency response boost that a ported system would otherwise offer. The stuffing in a vented system should be used to control the resonances in that system related to the midrange(in a full range system with woofers that also extend into the midrange) or upper bass response(if it is a very big box). You should never fill a vented box with any sort of stuffing material.

Group delay is over-hyped, and I dare that Parts Express guy to point you to a credible perceptual study that demonstrates substatial audibility of group dealy as a lone variable in the amounts found in a properly designed ported alignment. The primary factor in audibility here is frequency response of the total system(that includes the subwoofer + room acoustics). You can go so far as to make your ported system sound like any sealed alignment that you wish, if you have a low enough F3 in the first place, and then use a powerful parametric equalizer to shape the final response to match a target curve.

-Chris

 

[WmAx]

1. WinISD shows a group delay over 27 ms. I know you think group delay is over hyped. 27ms seems like a lot. What can I expect from that figure. How audible is it at the lower frequencies. (no way I can hear 20hz anyway)

This is insubstantial for frequencies at subject here. Group delay is an inherant property of any filter, and it will always be a higher value for a steeper/cascaded filter. But it the values you give are insubstantial. It will have little, if any, audible effect.

But I think you need to hear [the lack of difference] for yourself under control conditions. I have prepared some samples for you to play over a high quality headphone with low bass response(if you have such avaiable). You should use an ABX comparator such as PCABX to compare the samples to each other. For example, compare a 2nd order(equals sealed) to a 4th order(equals ported) curve. Or compare the original unaltered file to either one. Get the freeware compare program here:
http://www.pcabx.com/program/index.htm

Get the files I prepared here:
http://www.linaeum.com/downloads/misc/gdelay/

The files I have provided have these properties:

drums1 = original

drums1_4 = 4th order slope at 30 Hz
Group Delay: 40Hz=17ms, 30Hz=25ms, 25Hz=29ms, 20Hz=25ms

drums1_2 = 2nd order slope at 30 Hz
Group Delay: 40Hz=13ms, 30Hz=15ms, 25Hz=17ms, 20Hz=15ms

basssynth1 = original

basssynth1_4 = 4th order slope at 30 Hz
Group Delay: 40Hz=17ms, 30Hz=25ms, 25Hz=29ms, 20Hz=25ms

basssynth1_2 = 2nd order slope at 30 Hz
Group Delay: 40Hz=13ms, 30Hz=15ms, 25Hz=17ms, 20Hz=15ms

Be sure to use the PCABX software if you want unbiased evaluation(s).

2. What do you think about the box size?

What about the size? It's up to you to choose your response characteristics so far as volume/tuning.

3. The parts that I ordered are the Dayton RSS315HF-4 12" Woofer (295-464) and their SA240 Plate Map (300-804) Do these parts look okay for this design?

Does the amp have sufficient power? Is it stabile at the impedance of the speaker you are using? If yes to both, then it's good.

4. No stuffing?? Nothing? How about bracing. Any other construction advice?

No reason to use stuffing in a vented system unless you use the subwoofer at a crossover frequency where signficant upper band energy leaks past the crossover relative to the half wavelength distance of parallel walls in the sub cabinet.

-Chris

 

[bcycle]

Thanks for the help

Thanks WmAx for the files and your guidance. I'm not hearing big differences in the files which dramatizes your point. As far as box size, I'm working with WinISD and BBox Pro which are showing somewhat different box sizes.

The amp should be fine according to the folks at Parts Express. We'll find out when I hook it up.

Thanks again!

Greg

What about the size? It's up to you to choose your response characteristics so far as volume/tuning.

Does the amp have sufficient power? Is it stabile at the impedance of the speaker you are using? If yes to both, then it's good.

No reason to use stuffing in a vented system unless you use the subwoofer at a crossover frequency where signficant upper band energy leaks past the crossover relative to the half wavelength distance of parallel walls in the sub cabinet.

-Chris

 

[bcycle]

ACK! This should be my last question....

It's me again,

I need a couple rules of thumb to work with here, and than I can start building.

I think I have the box figured out, with an F3 around 23.33 HZ and an FB about 18.6 but I think I've heard that there's a formula with respect to how these 2 relate.

Assuming that I've got the driver and box VB resolved, what are the other critical, relevant calculations that I need to consider? I've heard something about the F3 in relation to the FB; that one should be higher or lower, or something... Please advise.

Thanks,

Greg

 

[WmAx]

It's me again,

I need a couple rules of thumb to work with here, and than I can start building.

I think I have the box figured out, with an F3 around 23.33 HZ and an FB about 18.6 but I think I've heard that there's a formula with respect to how these 2 relate.

Assuming that I've got the driver and box VB resolved, what are the other critical, relevant calculations that I need to consider? I've heard something about the F3 in relation to the FB; that one should be higher or lower, or something... Please advise.

Thanks,

Greg

You have been using a proper modeling program to determine air mach vent speed, driver excursion, power handling and SPL vs frequency. You are doing fine. The one thing you need to consider at this point, since you appear to be greatly concerned with sound quality, is a parametric equalizer to even out your bass response in the actual room setting.

-Chris