lots o detail
XRK ,
Here is the 5db per division graph from Akabak for the 40hz Tri-chamber Karlflex loaded with a PA-310 ..... Hopefully the real world results won't be quite so volatile but it will likely be somewhat midrange dominant ....
HolmImpulse and DATS measurements to be posted later today
XRK ,
Here is the 5db per division graph from Akabak for the 40hz Tri-chamber Karlflex loaded with a PA-310 ..... Hopefully the real world results won't be quite so volatile but it will likely be somewhat midrange dominant ....
HolmImpulse and DATS measurements to be posted later today
Attachments
Last edited:
Some early testing results
So here are some pictures and graphs .... I ran some tests on the Tri-chamber Karflex loaded with PA-310 without the K-aperture attached, and i can already see that it tuned below my target ... The fundamental really needs to land around 44 to 46hz because the K-slot aperture will add more path length/loading and drive down the fundamental to around 40hz .... Attempting to tune below 40 with this driver in a small box is counterproductive, so i am going to stick with my ending target of 40hz, which means i will have to make a few adjustments to the box ....
NOW FOR THE GOOD NEWS!
The parasitic chamber seems to be functioning properly and it's resonance landed right around 75hz which is exactly what i was shooting for .....
I attached impedance curves measured by the DATS unit (without aperture attached to front of cab) , with S1 fully stuffed and a small amount of stuffing in the parasitic chamber/pipe ...... I also attached the simulated impedance generated by Akabak ..... You can see that the measured impedance closely resembles the classic DCR but it may change a little once i add the aperture ...
Using HolmImpulse and my Dayton EMM6 mic i measured the output at each port and also the cone ...... Green=Cone .....BLACK=main vent .... RED=Parasitic chamber's vent ...
Output shows signs of being highly overdamped right now which is to be expected since without an aperture it is basically operating as just an undersized Offset-Driver-ML-TL with parasitic chamber, the K-slot aperture will add two orders and makes the system behave more like a series tuned 6th order box with some tapped effect and should alleviate the overdamped condition to some extent allowing for more output at fundamental ...
The phase plot shows that all ports and the cone become fairly phase coherent around fundamental which is GREAT! Then they deviate around the parasitic resonance which is to be expected in a DCR-ish design, then above that they become fairly coherent again out to almost 500hz which is actually pretty awesome But once again , things will change some more once i get this box tuned, and a K-slot slapped onto the front ...
So here are some pictures and graphs .... I ran some tests on the Tri-chamber Karflex loaded with PA-310 without the K-aperture attached, and i can already see that it tuned below my target ... The fundamental really needs to land around 44 to 46hz because the K-slot aperture will add more path length/loading and drive down the fundamental to around 40hz .... Attempting to tune below 40 with this driver in a small box is counterproductive, so i am going to stick with my ending target of 40hz, which means i will have to make a few adjustments to the box ....
NOW FOR THE GOOD NEWS!
The parasitic chamber seems to be functioning properly and it's resonance landed right around 75hz which is exactly what i was shooting for .....
I attached impedance curves measured by the DATS unit (without aperture attached to front of cab) , with S1 fully stuffed and a small amount of stuffing in the parasitic chamber/pipe ...... I also attached the simulated impedance generated by Akabak ..... You can see that the measured impedance closely resembles the classic DCR but it may change a little once i add the aperture ...
Using HolmImpulse and my Dayton EMM6 mic i measured the output at each port and also the cone ...... Green=Cone .....BLACK=main vent .... RED=Parasitic chamber's vent ...
Output shows signs of being highly overdamped right now which is to be expected since without an aperture it is basically operating as just an undersized Offset-Driver-ML-TL with parasitic chamber, the K-slot aperture will add two orders and makes the system behave more like a series tuned 6th order box with some tapped effect and should alleviate the overdamped condition to some extent allowing for more output at fundamental ...
The phase plot shows that all ports and the cone become fairly phase coherent around fundamental which is GREAT! Then they deviate around the parasitic resonance which is to be expected in a DCR-ish design, then above that they become fairly coherent again out to almost 500hz which is actually pretty awesome
But once again , things will change some more once i get this box tuned, and a K-slot slapped onto the front ...Attachments
-
overdamped 40hz DC-ML-TL-PHASE.png -
Tri-Chamber-Karlflex-stuffedS1-ModerateParachamber-stuff-Black=main-vent__ Green=cone__ Red=para.PNG -
Tri-Chamber-Karlflex-stuffedS1-ModerateParachamber-stuff.PNG -
Tri-Chamber-Karlflex-stuffedS1-ModerateParachamber-stuff--AKABAK-IMPEDANCE.PNG -
IMG_2640.jpg -
IMG_2639.jpg
Last edited:
FREDDI !
Good afternoon!
INDEED! , wild and wacky stuff ..... I suppose i could have kept it simple and left out the parasitic section and it would have been just a regular Karlflex which would have worked fine, but with this particular driver you gain an extra 5db (virtual) at FB according to Akabak (+10v drive and +30L para-section size) made possible by the extra cone control that this DCR-ish modification provides around the excursion hump above FB (the PA-310 really needs tight cone control in order to safely take higher drive levels)...... So i guess i was tempted by the extra output, and knowing that my friend's DC-TL towers turned out to be such a great success helped me to feel confident enough to give this a try, an experiment, FOR THE SAKE OF SCIENCE and art!
...
I wish i would have made more of the panels easily removable (not just the side panel) but i have already managed to cut out and eliminate that little shelf extension in front of the main vent, and then I rounded over the end of that vent which altogether shifted the FB up about 1.5hz so that is progress .... I also clamped on a temporary aperture-esque panel across the front of the cab so i could take a few measurements, and the loading only shifted things down by a few hertz, so that is good news, im actually not too far off the mark here ....... I can say for sure now that this thing definitely sounds a heck of a lot better with some sort of aperture on the front
Not nearly as overdamped, warmer, almost acceptable ..
Gday team-transflex,
New to the party, I'm interested in building using one of the lab15-4ohm that I picked up. Just wondering if I'm on the right track using 12mm ply. I know the internal tilted portion didn't come out too well on sketchup, but you get the idea. Design was utilising the attached plan by MMJ and the 'fire-in-the-middle' karlflex apeture.
I have an inuke 3000 and dcx2496 for power and xo, finish exams mid-june so should begin construction there-abouts.
Quick conversions:
H = 24" = 609.6mm
W = 19" = 482.6mm
D = 18.25 = 463.55mm
A = 6.937" = 176.2mm
C = 2.5" = 63.5mm
3.75" = 95.25mm
Regards,
Matt.
New to the party, I'm interested in building using one of the lab15-4ohm that I picked up. Just wondering if I'm on the right track using 12mm ply. I know the internal tilted portion didn't come out too well on sketchup, but you get the idea. Design was utilising the attached plan by MMJ and the 'fire-in-the-middle' karlflex apeture.
I have an inuke 3000 and dcx2496 for power and xo, finish exams mid-june so should begin construction there-abouts.
Quick conversions:
H = 24" = 609.6mm
W = 19" = 482.6mm
D = 18.25 = 463.55mm
A = 6.937" = 176.2mm
C = 2.5" = 63.5mm
3.75" = 95.25mm
Regards,
Matt.
Attachments
Hello Mattcalf!
Glad you were able to pick up some of those Lab15-4s , i keep telling folks about them because they are one of the best values in the subwoofer driver market right now, it is a very good dollar-per-decibel deal
(they are still available on ebay for the moment, but who knows for how long, so people should grab some while they can!)I can see that you made the box a little deeper to compensate for the extra front chamber depth and double-thickness baffle
.... No problem at all, looks good , your slightly larger front chamber will increase the midbass presence a little bit and lower the bandpass cutoff, depending on the size of the aperture's opening which are all effects that are totally acceptable for that driver .....Your sketch and numbers get my stamp of approval
However this box will need some bracing of course, fairly similar to how i braced the Tri-chamber Karlflex that i posted the pictures of recently with a brace on top in the main vent , one in the S1 area , and one on the bottom panel connecting to the S1 panel ... If you would like i can sketch it up for you.... You may also want to put a small vertical brace behind the front panel (below the pyramid) , of course it could not have too much depth or the cone will smack it at high excursions..The Fire-In-The-Middle aperture is a good choice (as is the inverted K-slot) for this heavy driver because it adds some parallel loading and adds gain in the midband and midbass (below cutoff) .... I would suggest that the Pyramid shaped mouth/terminus have a cross sectional area which is no less than the SD of the Lab15, if you make it much smaller than that it will likely tune the box too low and you could lose performance .... The Pyramid's vertical placement will also effect the fundamental tuning of the box (placing it lower tunes the box lower) .... 35hz is a good target FB for this cab with this driver ... Do you have a way to measure your FB? If not just message me in private and i can give you a few methods ...
Of course since you are in Australia you will have to do all of this upside down
but the same principles apply 
Last edited:
Freddi,
Yes, definitely, i can block off the path to the parasitic section turning the box into a standard Karlflex to get some measurements .... I will have to try that
So here are some pictures and graphs .... I ran some tests on the Tri-chamber Karflex loaded with PA-310 without the K-aperture attached, and i can already see that it tuned below my target ... The fundamental really needs to land around 44 to 46hz because the K-slot aperture will add more path length/loading and drive down the fundamental to around 40hz .... Attempting to tune below 40 with this driver in a small box is counterproductive, so i am going to stick with my ending target of 40hz, which means i will have to make a few adjustments to the box ....
NOW FOR THE GOOD NEWS!
The parasitic chamber seems to be functioning properly and it's resonance landed right around 75hz which is exactly what i was shooting for .....
I attached impedance curves measured by the DATS unit (without aperture attached to front of cab) , with S1 fully stuffed and a small amount of stuffing in the parasitic chamber/pipe ...... I also attached the simulated impedance generated by Akabak ..... You can see that the measured impedance closely resembles the classic DCR but it may change a little once i add the aperture ...
Using HolmImpulse and my Dayton EMM6 mic i measured the output at each port and also the cone ...... Green=Cone .....BLACK=main vent .... RED=Parasitic chamber's vent ...
Output shows signs of being highly overdamped right now which is to be expected since without an aperture it is basically operating as just an undersized Offset-Driver-ML-TL with parasitic chamber, the K-slot aperture will add two orders and makes the system behave more like a series tuned 6th order box with some tapped effect and should alleviate the overdamped condition to some extent allowing for more output at fundamental ...
The phase plot shows that all ports and the cone become fairly phase coherent around fundamental which is GREAT! Then they deviate around the parasitic resonance which is to be expected in a DCR-ish design, then above that they become fairly coherent again out to almost 500hz which is actually pretty awesome
MMJ,
Great work here!!!
you are one of the few who are actually using the ability of Akabak to model wild and wacky stuff not possible with other sim programs. Your agreement between model and measurement on impedance sweep is impressive and shows that you built according to the model. Nice work. I am really excited to see this great development and look forward to the final results with K aperture in place. My experience is that it pulls the tuning a few (3 to 5Hz lower) and adds come motion control for higher power without hitting xmax so soon. Along the way, you may get a W shaped dip around 250Hz to 500Hz. I don't think you want to use this above 500Hz as the mids will have delay and sound like they are echoing.
oddly enough, Fane's double whizzer cone 12" sounded very clear in a Karlson Twelve/"K12" - there's a chance (?) MMJ's cabinet can reach a K-tube - Acoustic's 115BK does well imo.
are there any general observations on what layout of port and driver position in the path can fill or choke the output in the middle of the "W" portion of response?
(apparently Bo Diddley didn't heed this message)
are there any general observations on what layout of port and driver position in the path can fill or choke the output in the middle of the "W" portion of response?
(apparently Bo Diddley didn't heed this message)
Walking the Plank
Thank you XRK
The sim and measured impedance plots do have a strong resemblance, and i was very happy to see that Once again , thanks a bunch to you and the other awesome folks on here that got me started with Akabak ..
Over the last couple of days while i was doing some testing with different amounts of stuffing i also experimented with a simple clamp on plank-in-the-middle sort of aperture (something i could easily slap on the front while the side panel is still just temporarily sealed with weatherstrip) ..... I attached a picture for everyone to see , and you can also see some of the heavy items i placed on top to apply pressure to the weatherstrip ... hehehe ...
The plank style aperture allowed for a rather large mouth area at 860cm sq , and despite the large mouth area there was still a noticeable bandpass cutoff around 600hz and FB was shifted down by about 2hz ...... I attached the measured response graph .... Still looking somewhat overdamped , but not sounding too bad really (I do however notice a small amount of that muddy delayed or echoed mid sound that you mentioned XRK) .....
I have no doubt that the actual K-slot styled aperture will probably behave a little better, but i still need to trim up the main vent a little more yet (FB needs to come up by a few more hertz) before i glue the cabinet side on permanently ..... 40hz output should come up a little more with the additional tapped effect i will get with a proper tuning & aperture (and smaller mouth area) ....
NOTE: I have found out that stuffing is a necessity with this cabinet using the PA-310 , sound quality drastically improved with fine poly-fiber stuffing added to the S1 area (and also a little in the parasitic chamber section where there was room for it without choking the path) , it really cleans up the midbass and midrange and makes this box much more pleasant to listen to. The difference to my ears was dramatic, but oddly enough the measured amplitude/frequency response didn't really change much at all, although the measured impulse response did look cleaner , so maybe that is what i am hearing (and or something phase related)......
MMJ,
Great work here!!!
Thank you XRK
The sim and measured impedance plots do have a strong resemblance, and i was very happy to see that
Once again , thanks a bunch to you and the other awesome folks on here that got me started with Akabak ..Over the last couple of days while i was doing some testing with different amounts of stuffing i also experimented with a simple clamp on plank-in-the-middle sort of aperture (something i could easily slap on the front while the side panel is still just temporarily sealed with weatherstrip) ..... I attached a picture for everyone to see , and you can also see some of the heavy items i placed on top to apply pressure to the weatherstrip ... hehehe ...
The plank style aperture allowed for a rather large mouth area at 860cm sq , and despite the large mouth area there was still a noticeable bandpass cutoff around 600hz and FB was shifted down by about 2hz ...... I attached the measured response graph .... Still looking somewhat overdamped , but not sounding too bad really (I do however notice a small amount of that muddy delayed or echoed mid sound that you mentioned XRK) .....
I have no doubt that the actual K-slot styled aperture will probably behave a little better, but i still need to trim up the main vent a little more yet (FB needs to come up by a few more hertz) before i glue the cabinet side on permanently ..... 40hz output should come up a little more with the additional tapped effect i will get with a proper tuning & aperture (and smaller mouth area) ....
NOTE: I have found out that stuffing is a necessity with this cabinet using the PA-310 , sound quality drastically improved with fine poly-fiber stuffing added to the S1 area (and also a little in the parasitic chamber section where there was room for it without choking the path) , it really cleans up the midbass and midrange and makes this box much more pleasant to listen to. The difference to my ears was dramatic, but oddly enough the measured amplitude/frequency response didn't really change much at all, although the measured impulse response did look cleaner , so maybe that is what i am hearing (and or something phase related)......
Attachments
Last edited:
made me think of Swinyard's 1935 patent http://patentimages.storage.googleapis.com/pages/US2020166-0.png
To boldly go where ....... someone already went back in 1935! =-O
I see the similarity! Laid back baffle and the plank is extra cool because it is shaped like an upside down Star Trek insignia badge! hehe ..... Cool stuff , and from 1935 even!
Looks like the Star Trek insignia shaped structure is a hollow waveguide sort of device firing into the mouth/baffle area for additional loading?
I see the similarity! Laid back baffle and the plank is extra cool because it is shaped like an upside down Star Trek insignia badge! hehe ..... Cool stuff , and from 1935 even!
Looks like the Star Trek insignia shaped structure is a hollow waveguide sort of device firing into the mouth/baffle area for additional loading?
Eek! My speakers have parasitic stuff inside, they must have eaten too much sushi
Thanks TB !
I am gaining more and more respect for the ancients and their great big brains! It amazes me that they were exploring all of these things without highly capable drivers , without T/S parameters (before the 60s right?), and without any software to guide them, at least nothing like we have today ....... The amount of intuition used in the engineering and the vast amount of sawdust created during the trial-and-error process must have been fully mind boggling .......
I just took a good look at the those H-PAS patents and it seems to me that it would not be too difficult to simulate that sort of parasitic chamber style concept in Akabak.... I will just have to attempt a sim now out of curiousity
Hi MMJ,
Lots of interesting work, keep it up.
Freddi posting a patent made me think too (you just can't beat the ancients), have you taken a look at H-Pas, and the Clements Pat4373606 and US2011/0058700A1? Good stuff.
Regards,
Thanks TB !
I am gaining more and more respect for the ancients and their great big brains! It amazes me that they were exploring all of these things without highly capable drivers , without T/S parameters (before the 60s right?), and without any software to guide them, at least nothing like we have today ....... The amount of intuition used in the engineering and the vast amount of sawdust created during the trial-and-error process must have been fully mind boggling .......
I just took a good look at the those H-PAS patents and it seems to me that it would not be too difficult to simulate that sort of parasitic chamber style concept in Akabak.... I will just have to attempt a sim now out of curiousity
Last edited:
BeauB,
Took me forever to get started too ..... It feels good to make sawdust though, i promise!
Well now , this looks good
TB46
I have good news! I did some simulating with the H-PAS style parasitic chamber using a chamber that is 30-35 liters in size added to a 60 liter Karlflex (the same ratio used in a DCR) giving us a total of around 90 liters net and i think the results are extremely noteworthy!!
It looks like the excursion control is comparable to what my labyrinthine parasitic section accomplishes (either type allows for a little more than 33% additional drive voltage with the PA310) and output is also very similar with either style... This is all very exciting because this H-PAS style chamber would greatly simplify the cabinet... The easier & faster build process appeals to me
It would still be a Tri-Chamber Karlflex and can still use the K-slot aperture, basically the same layout other than the simplified parasitic chamber and also with the exception of a shorter main vent in order to get the FB back up to around 40hz according to the akabak model...
I will post the code, graphs and a rough sketch here soon
TB46
I have good news! I did some simulating with the H-PAS style parasitic chamber using a chamber that is 30-35 liters in size added to a 60 liter Karlflex (the same ratio used in a DCR) giving us a total of around 90 liters net and i think the results are extremely noteworthy!!
It looks like the excursion control is comparable to what my labyrinthine parasitic section accomplishes (either type allows for a little more than 33% additional drive voltage with the PA310) and output is also very similar with either style... This is all very exciting because this H-PAS style chamber would greatly simplify the cabinet... The easier & faster build process appeals to me
It would still be a Tri-Chamber Karlflex and can still use the K-slot aperture, basically the same layout other than the simplified parasitic chamber and also with the exception of a shorter main vent in order to get the FB back up to around 40hz according to the akabak model...
I will post the code, graphs and a rough sketch here soon
Last edited:
fwiw I once added a cavity to the top of a K's front chamber - Clements H-Pas
seems to be a re-naming of his earlier patent - guess these days easier to simulate
seems to be a re-naming of his earlier patent - guess these days easier to simulate
An externally hosted image should be here but it was not working when we last tested it.
