I've got 4 x 15" drivers out of Jamo D7 subwoofers.
Normally these drivers use MFB (Motional Feed Back... ie: servo-like technology) however on these woofers, the MFB is defunt.
The woofers work fine as is.
I have measured 2 of the drivers with Dayton Woofer Tester 3, and both measured similar enough.
Here are the parameters of one of the drivers:
Fs = 20.86 Hz
Qms = 7.995
Vas = 9.719 cu.ft (this kind of scares me - tells me it's going to need a big box!)
Mms = 214.9 g
Xmax = 15 mm (from one piece of literature, this has been specified as xmax... could be half this?)
Qes = 0.598
Re = 3.3 ohms
Le = 2.694 mH
Z = 4 ohms
Pe = 400 watts
Qts = 0.556
I modelled these parameters in the original sealed enclosure, approx 100ltrs sealed and it comes close to a q of .707... Which tells me these measurements are probably quite accurate.
I'm wanting to build these into a coffee table subwoofer (as I can place this more optimally in the room), of dimensions approx 450h x 1100w x 750d mm, giving me approx 300ltrs of volume. I know this might be a tad small for these drivers, so I am contemplating isobaric if I can't make them work otherwise.
I am also wanting to have the drivers wired in push-pull, as to cancel out vibrations (and allow me to use it as a coffee table 😛).
Am open to suggestions on what sort of enclosure to use...
Thanks 🙂
Normally these drivers use MFB (Motional Feed Back... ie: servo-like technology) however on these woofers, the MFB is defunt.
The woofers work fine as is.
I have measured 2 of the drivers with Dayton Woofer Tester 3, and both measured similar enough.
Here are the parameters of one of the drivers:
Fs = 20.86 Hz
Qms = 7.995
Vas = 9.719 cu.ft (this kind of scares me - tells me it's going to need a big box!)
Mms = 214.9 g
Xmax = 15 mm (from one piece of literature, this has been specified as xmax... could be half this?)
Qes = 0.598
Re = 3.3 ohms
Le = 2.694 mH
Z = 4 ohms
Pe = 400 watts
Qts = 0.556
I modelled these parameters in the original sealed enclosure, approx 100ltrs sealed and it comes close to a q of .707... Which tells me these measurements are probably quite accurate.
I'm wanting to build these into a coffee table subwoofer (as I can place this more optimally in the room), of dimensions approx 450h x 1100w x 750d mm, giving me approx 300ltrs of volume. I know this might be a tad small for these drivers, so I am contemplating isobaric if I can't make them work otherwise.
I am also wanting to have the drivers wired in push-pull, as to cancel out vibrations (and allow me to use it as a coffee table 😛).
Am open to suggestions on what sort of enclosure to use...
Thanks 🙂
Here is a suggestion of 4-driver T-TQWT = 2 x paralleled isobaric drivers series connected in a 406L box. 😀
b🙂
PS: Check Driver Sd: If it's far off adjust at the HR input screen and fiddle the enclosure dimension parameters when in the Loudspeaker Wizard.
Attachments
Wow, that response looks exceptional!
Will that rise in group delay at 20hz have a large audible impact?
Is it necessary to have it going down to 13hz? While I no doubt will use it for movies as well as music, how much smaller would it be if the corner freq was closer to 20hz?
Now I just got to figure out how to fit them neatly into a "coffee table" lol, any examples of how to fold this up?
Thanks Bjorno!
I'm relatlively new with TWQT's...
Going to try and figure out Horn Resp now, wish me luck 🙂
Will that rise in group delay at 20hz have a large audible impact?
Is it necessary to have it going down to 13hz? While I no doubt will use it for movies as well as music, how much smaller would it be if the corner freq was closer to 20hz?
Now I just got to figure out how to fit them neatly into a "coffee table" lol, any examples of how to fold this up?
Thanks Bjorno!
I'm relatlively new with TWQT's...
Going to try and figure out Horn Resp now, wish me luck 🙂
Hi fury,
IMO, if the gldy is at or lower 1/f at 20 Hz the room ETC will probably be a greater problem if any.
http://www.diyaudio.com/forums/atta...e-sheet-challenge-bass-gldy-extrapolation.jpg
http://www.diyaudio.com/forums/subwoofers/186556-single-sheet-challenge-bass-2.html#post2531224
I didn't choose 13 Hz: Its a result appearing from HR manipulation when targeting a smooth FR.
Why not use my posted HR input screen and DIY tweak the entered dimension parameters until the response ends at 20 Hz ?
b 🙂
Last edited:
Thanks Bjorno.
I had a play around with it in Horn Resp, and I can get it a bit smaller however my reducing of "numbers" is quite random.
My understanding on T-TQWT's limited, so I guess I need to do some reading up on it.
Do I reduce all sections by the same %, or do I just want to reduce the last segment?
I had a play around with it in Horn Resp, and I can get it a bit smaller however my reducing of "numbers" is quite random.
My understanding on T-TQWT's limited, so I guess I need to do some reading up on it.
Do I reduce all sections by the same %, or do I just want to reduce the last segment?
I would first decide on how to fold the layout and measure the driver frame width,the mounting depth(add 2-3 cm to allow for clearance to the internal divider).
The CSA labeled S2 should then be about = (driver frame width) x (mounting depth+(2-3 cm)). If not: You need to increase the baffle-divider thickness until enough clearance is achieved.
There are no rules, only consequences if changing the suggested dimensions...
b🙂
Nice work Bjorno!
Looks like you are getting about 93 db sensitivity with 13 Hz extension. Pretty impressive. The box is a bit smaller than my T20.
FYI, fury's system is on my blog here:
Red Spade Audio: Fury's JBL waveguide speakers
Will the box fit?
I wonder if you will have some localisation challenges also. How about an isobarik bandpass? I'd be thinking 6th order. Shrinking any kind of 1/4 wave sub is going to give you a bit more out of band peaks. Bjorno has done a good job knocking up a quick design, but I wonder if the peak above 100 Hz will make this issue more difficult. My suspicion is that while this kind of design is more interesting and impressive, a bandpass might work out more workable. I'd be working on 20 - 50 Hz 6th order isobarik. You could model it for two drivers, then half the net volume and adjust the port tuning to suit the new volume. Then go for downfiring slot ports and keep them away from the listening position as much as you can. If you can get it to model well then I'd go for something like that.
Looks like you are getting about 93 db sensitivity with 13 Hz extension. Pretty impressive. The box is a bit smaller than my T20.
FYI, fury's system is on my blog here:
Red Spade Audio: Fury's JBL waveguide speakers
Will the box fit?
I wonder if you will have some localisation challenges also. How about an isobarik bandpass? I'd be thinking 6th order. Shrinking any kind of 1/4 wave sub is going to give you a bit more out of band peaks. Bjorno has done a good job knocking up a quick design, but I wonder if the peak above 100 Hz will make this issue more difficult. My suspicion is that while this kind of design is more interesting and impressive, a bandpass might work out more workable. I'd be working on 20 - 50 Hz 6th order isobarik. You could model it for two drivers, then half the net volume and adjust the port tuning to suit the new volume. Then go for downfiring slot ports and keep them away from the listening position as much as you can. If you can get it to model well then I'd go for something like that.
Thanks Paul.
The T-TQWT Bjorno posted is a wee bit too big.
Fiddling with dimensions I can get closer to my goal of 300 ltrs, but it's then a matter of folding it in a way that fits....
I considered a 6th order bandpass enclosure, but I haven't been able to get a FR and GD as flat as Bjorno's T-TQWT...
The T-TQWT Bjorno posted is a wee bit too big.
Fiddling with dimensions I can get closer to my goal of 300 ltrs, but it's then a matter of folding it in a way that fits....
I considered a 6th order bandpass enclosure, but I haven't been able to get a FR and GD as flat as Bjorno's T-TQWT...
Decided to do a quick sim in WinISD, keeping in mind that it's not entirely accurate. From memory it tends to exaggerate the peak on the top end. Example:
Red Spade Audio: Black widow prototype - 6th order bandpass
Just a little EQ and it came up nice and flat.
Red Spade Audio: Black widow test box measurements and impressions
Try this. 200L rear tuned 16 Hz, 100L front tuned 40 Hz. By 100 Hz it's down 16 dB acoutically. In this position I'd be more interested in that than smoothness. You should get smooth enough to get it into shape with a little EQ, if that is needed at all.
Assuming 7.5mm xmax, 500w > 114 dB with -3 db points at 21 and 60 Hz.
If you aren't doing a bandpass, try a little test with your existing subs in that position. I've never had good results with subs in a coffee table position, or various "nearfield" positions. Even with low distortion subs with a low crossover point and steep filters, it messes up imaging when up that close. I'd want to try it out before building something big! Imagine you fire it up only to find that you can't live with it.
Red Spade Audio: Black widow prototype - 6th order bandpass
Just a little EQ and it came up nice and flat.
Red Spade Audio: Black widow test box measurements and impressions
Try this. 200L rear tuned 16 Hz, 100L front tuned 40 Hz. By 100 Hz it's down 16 dB acoutically. In this position I'd be more interested in that than smoothness. You should get smooth enough to get it into shape with a little EQ, if that is needed at all.
Assuming 7.5mm xmax, 500w > 114 dB with -3 db points at 21 and 60 Hz.
If you aren't doing a bandpass, try a little test with your existing subs in that position. I've never had good results with subs in a coffee table position, or various "nearfield" positions. Even with low distortion subs with a low crossover point and steep filters, it messes up imaging when up that close. I'd want to try it out before building something big! Imagine you fire it up only to find that you can't live with it.
Paul, these were similar settings as what I got when simulating with BBP6.
It gives me -3db of 28hz and 60hz...
Bottom end is not quite the same as from the T-TQWT although it is 100ltrs smaller.
I will try with my current subs in the "coffee table" position, however it will mean my mains will have to sit on the floor for this test.
Will be interesting though.
It gives me -3db of 28hz and 60hz...
Bottom end is not quite the same as from the T-TQWT although it is 100ltrs smaller.
I will try with my current subs in the "coffee table" position, however it will mean my mains will have to sit on the floor for this test.
Will be interesting though.
paulspencer:
Will the box fit?
You cannot draw conclusions from looking at HR T-TQWT plots as for an TH. Well, you can, but only if you are very experienced or have done the required math or actually built/measured one where damping materials are involved.
A simulation using the full extent of adding damping materials can be done in AKAbak/or by using similar acoustic Cad programs permitting this) or by using a (semi-)simulation in MJK:s programs are IMO what's required to understand the acoustics involved in a T-TQWT.
IMO, My experience with properly done TL-sub's in respect to localization is that noticeable issues will only appear with a substantial change of the listening spot,where a tonal shift in the mid-bass register can be heard due to different far-field summation from the terminus and driver cone.
This does never appear for a T-TQWT as at the termination acts like a pressure point source and at onset within the ears non-ability to resolve time at lower frequencies in a normal room.
As the pre-pulse IR and the main-pulse IR levels are very similar: This result in system independence when integrating with main speakers: if the sub is phase aligned with the pre-pulse or with the main-pulse is just a matter of convenience when adjusting XO (and signal delay unit if used).
Any remaining residual gldy or amplitude ripple is IME at a 'blameless' level and not noticed especially when listening to music.
From solely localization point of view: If correctly designed:The IMO superior T-TQWT is second to none: As not suffering from attention-localization even when placed located beside you at your sofa close as possible..
And of course not forgetting the prerequisite: The use of a sub signal delay to line up with the main speakers and a suitable LPF: A LR2 if XO at 80~Hz or a LR4 if at maximum 145Hz.
Now we could talk about severe localization issues that IME always would appear.
I haven't build a 6:th order BP since the early seventies due to the always lousy SQ.
Yes but the peaks are never heard (hear-able as the are at JND levels for a stuffed T-TQWT) opposed to BR types using minimal stuffing in the main enclosure volume and none in the port(s).
If you look at my MJK plots in the first of submitted pictures and compare to (1) the peak appearing in HR and at (2) at the MJK plot where the OD-TL FR in reality is stuffed, ( T-TQWT inherit the ~ OD-TL FR above 80 Hz) it's evident that this is not an issue at all, not in theory as well as in practice if trying to measure or if you slowly sweep a sinus you will at most hear a dB dip located just above 100Hz, if in reality is occurring: depends on how well the stuffing is distributed and placed in the T-TQWT.
Single ported 4:th order BP BR('s) with very low off-band peaks (using very short ports like for an T-TQWT, i.e. no port resonances(driver to port) appearing at all) can IME be good assets in a multi sub system when placed at lateral positions in a room where the ports are pointed perpendicular to the listening position(s) and only if covering a flat critical BW above ~70 Hz + partly the mid-bass octave below. Note: Here the mid-bass per definition is the mathematical= 40 – 80 Hz.
IMO 6th order Br's, BP or not, that I've heard were all flawed in one way or another and I believe they only would work blameless if tuned very low, below the mid-bass BW, preferably below 20 Hz thus incorporating at least more than a half octave of the infra region,but that's my opinion.
b 🙂
Will the box fit?
You cannot draw conclusions from looking at HR T-TQWT plots as for an TH. Well, you can, but only if you are very experienced or have done the required math or actually built/measured one where damping materials are involved.
A simulation using the full extent of adding damping materials can be done in AKAbak/or by using similar acoustic Cad programs permitting this) or by using a (semi-)simulation in MJK:s programs are IMO what's required to understand the acoustics involved in a T-TQWT.
IMO, My experience with properly done TL-sub's in respect to localization is that noticeable issues will only appear with a substantial change of the listening spot,where a tonal shift in the mid-bass register can be heard due to different far-field summation from the terminus and driver cone.
This does never appear for a T-TQWT as at the termination acts like a pressure point source and at onset within the ears non-ability to resolve time at lower frequencies in a normal room.
As the pre-pulse IR and the main-pulse IR levels are very similar: This result in system independence when integrating with main speakers: if the sub is phase aligned with the pre-pulse or with the main-pulse is just a matter of convenience when adjusting XO (and signal delay unit if used).
Any remaining residual gldy or amplitude ripple is IME at a 'blameless' level and not noticed especially when listening to music.
From solely localization point of view: If correctly designed:The IMO superior T-TQWT is second to none: As not suffering from attention-localization even when placed located beside you at your sofa close as possible..
And of course not forgetting the prerequisite: The use of a sub signal delay to line up with the main speakers and a suitable LPF: A LR2 if XO at 80~Hz or a LR4 if at maximum 145Hz.
Now we could talk about severe localization issues that IME always would appear.
I haven't build a 6:th order BP since the early seventies due to the always lousy SQ.
Yes but the peaks are never heard (hear-able as the are at JND levels for a stuffed T-TQWT) opposed to BR types using minimal stuffing in the main enclosure volume and none in the port(s).
If you look at my MJK plots in the first of submitted pictures and compare to (1) the peak appearing in HR and at (2) at the MJK plot where the OD-TL FR in reality is stuffed, ( T-TQWT inherit the ~ OD-TL FR above 80 Hz) it's evident that this is not an issue at all, not in theory as well as in practice if trying to measure or if you slowly sweep a sinus you will at most hear a dB dip located just above 100Hz, if in reality is occurring: depends on how well the stuffing is distributed and placed in the T-TQWT.
Single ported 4:th order BP BR('s) with very low off-band peaks (using very short ports like for an T-TQWT, i.e. no port resonances(driver to port) appearing at all) can IME be good assets in a multi sub system when placed at lateral positions in a room where the ports are pointed perpendicular to the listening position(s) and only if covering a flat critical BW above ~70 Hz + partly the mid-bass octave below. Note: Here the mid-bass per definition is the mathematical= 40 – 80 Hz.
IMO 6th order Br's, BP or not, that I've heard were all flawed in one way or another and I believe they only would work blameless if tuned very low, below the mid-bass BW, preferably below 20 Hz thus incorporating at least more than a half octave of the infra region,but that's my opinion.
b 🙂
Attachments
Hi Oliver,
Of course, I tested your Simulation by increasing the stuffing density from 0.35 to 0.5 lb x ft^-3:
b🙂
Attachments
Hi bjorno,
Let's see if I'm getting this correctly, you changed the L12 to 1.275m, and added the additional length to L23, then added stuffing, and-according to the MJK simulations-the stuffing levels out the big response irregularity that Hornresp predicts @ 65Hz?
That's pretty neat, thanks for taking a look at this. This would give fury a wide range of enclosure sizes to play with.
Regards,
Let's see if I'm getting this correctly, you changed the L12 to 1.275m, and added the additional length to L23, then added stuffing, and-according to the MJK simulations-the stuffing levels out the big response irregularity that Hornresp predicts @ 65Hz?
That's pretty neat, thanks for taking a look at this. This would give fury a wide range of enclosure sizes to play with.
Regards,
Quite correct,The third harmonic will disappear completely,IME you don't need to offset the driver precisely at third of the length because you normally stuff the closed end using denser damping material that the wave enters through two times:
This gives you the freedom to make the closed end somewhat shorter than MJK:s program predicts or even longer if stuffed densely at the far closed end.
In MJK you will notice that the terminus far field level will rise at 65 Hz when stuffing is introduced but soon hits diminishing returns as losses occurs if stuffed too much.
b🙂
As always.. but when you actually measure the sub and then adjust again, it's a piece of cake, though sometimes takes time to get right, but I can assure: is very rewarding...
b🙂
Last edited:
- Status
- Not open for further replies.