Best cabinet for odd subwoofer driver

[Altec]

A couple of quick sims to show relative 100 W input comparison of series VC vented Vs relatively small, low Qtc single VC, especially its impact on the mid-bass. Note that the vent will be large, long due to acoustically small net Vb, low tuning.

GM

 

[Altec]

DSL glitch

 

[Sir.Byrd]

Remeasure to check what?

GM

To check relevant T/S parameters?

 

[TerryO]

Ahh...yes, my steel trap is showing its size again. :sigh: Here's a post from a while back on another forum with measurements of a different DVC with different hook ups.
http://www.diyaudio.com/forums/subwoofers/126275-dvc-has-me-all-wound-up.html

Aside from the loss in Power handling and efficiency, I'll look into the other changes and see what other box options might open up.

Dave,

Look for Ron Ennega's (Ron E, on diyAudio) calculations.
When I first suggested using a DVC with resistive manipulation back in the 90's while the Bass List and Dan Wiggins were working out the design of "The Official Bass List Subwoofer" (later offered commercially as the "Shiva"), both Dan and Ron were pretty enthusiastic and Ron worked out the math involved, which was "way" over my head.

Dan subsequently published a White Paper on what he termed RDO (Resistively Damped Operation?) and there's also some decent information there. I have, over the years, given Dan a bad time for not having credited me with the initial inspiration. But truth be told, without Dan and Ron the whole idea wouldn't have gone anywhere.

Best Regards,
Terry

 

[GordonW]

The thing is... Dan Wiggins also taught us... you can tune vented alignments to have a LOWER initial F3 than a sealed enclosure... as well as also having really low F10... it just takes low port frequencies. That's how most modern Tannoys are tuned.

Try taking the "optimum" vented alignment... and dropping the vent frequency. Start about 20% or so lower than normal...

Wiggins used to use alignments, which were about 20% smaller air volume than a standard vented alignment, with vent frequencies about 20% to 40% lower than standard. Created an alignment with not as much energy at the "corner", but much more at lower frequencies... as well as pretty darn good measured transient response...

So, don't write off the vented alignment... the advantage of lower cone excursion (and the resultant lower distortion) are quite appealing, if you do it right for the situation...

Same goes for bandpass! It's a little harder to tune (requires changing the cabinet volume ratios, between small and large chambers, usually, as well as vent frequency), but it can be done...

Regards,
Gordon.

 

[TerryO]

The thing is... Dan Wiggins also taught us... you can tune vented alignments to have a LOWER initial F3 than a sealed enclosure... as well as also having really low F10... it just takes low port frequencies. That's how most modern Tannoys are tuned.

Try taking the "optimum" vented alignment... and dropping the vent frequency. Start about 20% or so lower than normal...

Wiggins used to use alignments, which were about 20% smaller air volume than a standard vented alignment, with vent frequencies about 20% to 40% lower than standard. Created an alignment with not as much energy at the "corner", but much more at lower frequencies... as well as pretty darn good measured transient response...

So, don't write off the vented alignment... the advantage of lower cone excursion (and the resultant lower distortion) are quite appealing, if you do it right for the situation...

Same goes for bandpass! It's a little harder to tune (requires changing the cabinet volume ratios, between small and large chambers, usually, as well as vent frequency), but it can be done...

Regards,
Gordon.

Gordon's correct, that's exactly how Dan tuned the enclosures using what was termed the "Adire Alignment." It certainly worked well, I built a couple of "Shiva" subs using variations of this and they turned out very nice.

As far as Bandpass alignments, other than being interesting, I've generally avoided them for the reasons I've stated before. An addition problem is the fact that the "really interesting" higher order Bandpass configurations were (in commercial form) often subject to litigation initiated by the B**e Corporation's well-funded legal department.

Best Regards,
TerryO

 

[Altec]

Hey GW, long time no speak! Still working up on the Square? Speaking of Tannoy, did you get yours to market yet?

Yes, the Adire Alignment works well if there’s enough room gain and very similar to the optimum the pioneers of audio came up with, though back then of course, efficiency was everything, so strong motors to offset the system’s high output impedance and high compliance to gain back lost efficiency, ergo large cabs tuned to Fs ruled; but as you know, many homes don’t have a lot of gain, so when it’s not known, best to make it as large as practical to increase tuning flexibility, though I’m a firm believer in low tunings to at least Fs and lower in some cases.

Regardless, comparison of the low Qtc sealed plus if vented to 20 Hz, yielding a significant increase in LF response:

GM

 

[Altec]

To check relevant T/S parameters?

As the info in Dave’s link shows, once measured, the specs of different VC wiring combinations can be calculated from them plenty close enough for designing cab alignments.

GM

 

[Sir.Byrd]

Okay, that's good to hear.
So I could keep the cabinet and later upgrade the amp to accommodate a 2 ohm or 4 ohm load and get better SQ out of it (at least with the parallel configuration)?

I'm curious about TL, can you run some numbers for me? I'd like to get to the point where I could get some box dimensions and go to town.

 

[Dave R]

Thanks for the pointer to the RDO info and related info, Terry.

After looking at it again, I think I finally understand. When a single coil of a DVC is used (while leaving the other coil open) that combination yields the highest Qts -- as compared to using one coil and shorting the other. By adding a resistor to the unused coil, you can change the Qts to a value somewhere between those limits. (adding infinite resistance to unused coil is equivalent to "open", while shorting is equivalent to adding zero resistance.)

As the info in Dave’s link shows, once measured, the specs of different VC wiring combinations can be calculated from them plenty close enough for designing cab alignments.
GM

Thanks, GM
Without having the actual driver to measure, it is still useful to look at the highest Qts case, for when one coil is used, and the other coil is open. In that case, the changes relative to the coils wired in series (Ref posts #1 and #30) would be
Qes goes from .289 to .578
Qms stays at 3.573
So Qts = 1/[(1/Qes) +(1/Qms)] = 1/[(1/.578) + (1/(3.573)] = .50

BL goes from 24.34 to 12.17
Le goes from 6.26 to 3.13 @ 1kHz (2.08 to 1.04 @ 10kHz)
Re goes from 7.38 to 3.69

Using Qts = .50, instead of .268, a slightly larger vented box can be calculated. However due to the driver size (Sd), assuming excursion (Xmax) stays the same, and tuning frequency (fb) targeted at 20 Hz, the port diameter and length are still fairly large (4" flared x 24") may not be large enough diameter. Conservative diameter might be closer to 6". A slot or shelf port or other configuration may provide better results.

Attached is a graph of different sealed boxes, based on a Qts = .50. Again, graph doesn't show rolloff due to v.c.

I am also interested in what TL or TH possibilities would look like. :scratch2:
(I've been goaded for quite some time by TerryO to move in that direction, but have not begun to dive in yet.) :sigh:

And by the way, Merry Christmas to everyone. :wave:

 

[Altec]

Okay, that's good to hear.
So I could keep the cabinet and later upgrade the amp to accommodate a 2 ohm or 4 ohm load and get better SQ out of it (at least with the parallel configuration)?

I'm curious about TL, can you run some numbers for me? I'd like to get to the point where I could get some box dimensions and go to town.

Correct, all that will change is the frequency response and why only Fs, Vas, Qts is required to calculate net Vb.

That’s been my plan all along, but sealed, vented has been proposed and the options of DVC wiring plus the impact of high Le needed to be thrown into the ‘mix’, so I've only been providing frames of reference to show when/why certain resonant column configurations might be worth the extra design/construction effort.

GM

 

[TerryO]

Thanks for the pointer to the RDO info and related info, Terry.

After looking at it again, I think I finally understand. When a single coil of a DVC is used (while leaving the other coil open) that combination yields the highest Qts -- as compared to using one coil and shorting the other. By adding a resistor to the unused coil, you can change the Qts to a value somewhere between those limits. (adding infinite resistance to unused coil is equivalent to "open", while shorting is equivalent to adding zero resistance.)



Thanks, GM
Without having the actual driver to measure, it is still useful to look at the highest Qts case, for when one coil is used, and the other coil is open. In that case, the changes relative to the coils wired in series (Ref posts #1 and #30) would be
Qes goes from .289 to .578
Qms stays at 3.573
So Qts = 1/[(1/Qes) +(1/Qms)] = 1/[(1/.578) + (1/(3.573)] = .50

BL goes from 24.34 to 12.17
Le goes from 6.26 to 3.13 @ 1kHz (2.08 to 1.04 @ 10kHz)
Re goes from 7.38 to 3.69

Using Qts = .50, instead of .268, a slightly larger vented box can be calculated. However due to the driver size (Sd), assuming excursion (Xmax) stays the same, and tuning frequency (fb) targeted at 20 Hz, the port diameter and length are still fairly large (4" flared x 24") may not be large enough diameter. Conservative diameter might be closer to 6". A slot or shelf port or other configuration may provide better results.

Attached is a graph of different sealed boxes, based on a Qts = .50. Again, graph doesn't show rolloff due to v.c.

I am also interested in what TL or TH possibilities would look like. :scratch2:
(I've been goaded for quite some time by TerryO to move in that direction, but have not begun to dive in yet.) :sigh:

And by the way, Merry Christmas to everyone. :wave:

A tale of two Norwegians: The irresistible force and the immovable object!
(Some days you're one and the next day you're the other.)

Merry Christmas everyone!
Terry

 

[Altec]

Thanks, GM

Without having the actual driver to measure, it is still useful to look at the highest Qts case, for when one coil is used, and the other coil is open.

..........the port diameter and length are still fairly large (4" flared x 24") may not be large enough diameter.

You’re welcome!

Correctomundo! Unless I know all the particulars for the app, I always recommend a worst case scenario to maximize tuning flexibility since it’s a lot easier to shrink a cab than to stretch it to ‘fit’. Anyway, looks like you have a ‘handle’ on the DVC wiring options subject now.

Right, and why when the vent gets to be large, long, it’s better to just go ahead and morph the box/vent into either a mass loaded [vented or inverse tapered] TL or tapped variant depending on how wide a BW is required.

Tapped horns OTOH require somewhat different specs [mainly higher Fs] to properly handle the same gain BW as the others, so can’t directly compare it, but we’ll scope out what this driver can do as a true sub for pipe organ symphonies and HT LFE anyway.

GM

 

[Altec Best]

Very Interesting reads gentleman ! :thmbsp: Lots to learn here....

I also would like to wish everyone a "Merry Christmas" and a "Happy Holidays" !

 

[Dave R]

How much room gain can be expected for rooms of the following sizes? (Positioned away from walls, next to wall, or in corner)
8'w x 10'L x 8'h (640 cu. ft)
12'w x 16'L x 8'h (1536 cu. ft)
16'x 24'L x 10'h (3840 cu. ft)

Anyone have any good sources for info on amount of gain and related frequencies?

 

[TerryO]

How much room gain can be expected for rooms of the following sizes? (Positioned away from walls, next to wall, or in corner)
8'w x 10'L x 8'h (640 cu. ft)
12'w x 16'L x 8'h (1536 cu. ft)
16'x 24'L x 10'h (3840 cu. ft)

Anyone have any good sources for info on amount of gain and related frequencies?

Dave,

Ray Wagner (RayW) recently posted a simple calculation to determine the frequency at which room gain could be expected to begin. It was, as he noted, appropriate to enclosed rooms. I can see if I can find it.

EDIT : OK, here it is from 12/20/2011:

Start with your room's largest dimension, then double it, then translate that wavelength to a frequency. That will be the point where your sub stops producing a longitudinal traveling wave and starts pressurizing the room. From there, going down, you will gain 12dB per octave in room gain. Note that this is a simplistic way to do it and only really works for a sealed room but it'll give you an idea.

Best Regards,
Terry

 

[Dave R]

Thanks Terry -- a rule of thumb is often a good starting point for estimating. That's probably what I'm looking for here.

Now, if I can just figure out which of my (ten) thumbs to keep it on! :tongue:

 

[Sir.Byrd]

Okay, so can you guys help me pinpoint a TL enclosure that I can build for this?

 

[Altec]

Happy New Year and all that jive!

Have you been doing a bit of ‘due diligence’ studying the collected works of MJK, which includes his Classic TL Alignment Tables that has since been turned into an Excel SS, and the many others that have posted reams of theory, designs, etc., on the AVS, DIYaudio, and Quarter-Wave forums?

Learning how to use HornResp along with Soho’s companion Eazy Horn designer and especially plunking down $25 as soon as you can to get MJK’s software will go a long ways towards simming your way to successful TL, horn alignments and their tapped variants even if the various theories confounds you.

GM

 

[Sir.Byrd]

Happy New Year and all that jive!

Have you been doing a bit of ‘due diligence’ studying the collected works of MJK, which includes his Classic TL Alignment Tables that has since been turned into an Excel SS, and the many others that have posted reams of theory, designs, etc., on the AVS, DIYaudio, and Quarter-Wave forums?

Learning how to use HornResp along with Soho’s companion Eazy Horn designer and especially plunking down $25 as soon as you can to get MJK’s software will go a long ways towards simming your way to successful TL, horn alignments and their tapped variants even if the various theories confounds you.

GM

I can only use excel-based software, I'm on a Mac, not much speaker design software is supported. I'm not going to run parallels just to do this though.
I will read up on the MJK stuff. But if you have any ideas or suggestions to get me in the right direction for this specific driver, that would be great. I don't want to spend too much time on this project, just get something that I can build and listen to!

Also, is there something wrong with the TQWT alignment? It seemed to have promising results with the F3 of 18 Hz!