GM ,
It is good from about 35hz-100hz , about the same as the dual driver TH did with the same drivers .. . Not a great deal of bandwidth for sure, I was shooting more for high SPL in a small package since Saba mentioned a while back that he didn't need anything over 100hz , so i was just going with that ...
You know how it goes , you can trade some bandwidth for SPL , or some extension for SPL , or some box volume for extension (or SPL) and so on, its all a bunch of compromises right?
Nevertheless, its a rough sim & sketch which was posted with the intent to showing what HR is capable of when it comes to some more obscure configurations ...... Just exploring the possibilities here 🙂
Last edited:
Hi MMJ,
The inclusion of absorbent filling material will remain a feature of the Loudspeaker Wizard only.
The filling model requires that the frequency range be limited to 2000 hertz, and that only parabolic, conical and exponential horn segments be specified. This is most readily done using the wizard, which has the added benefit that the effects of changing the amount of damping can be seen in real time, facilitating design optimisation.
See the post below for further information:
http://www.diyaudio.com/forums/subwoofers/119854-hornresp-393.html#post3747637
Kind regards,
David
Last edited:
Absolutely, not many 'free lunches' in audio design.
Understood, but you asked our opinion on these variants.
GM
Could you explain how I would sim what I did draw? How much different would it be from what I simulated? I'm just trying to figure out why the TH I drew isn't what I modeled based off of the "basic TH" thread over on hometheatershack. If it is just because of the non-constant angle of expansion then that is the lack in knowledge that I need to figure out.
Any advice appreciated 🙂 just trying to settle on a design and it's difficult with roadblocks like this coming up.
Only Issue I see with this TH is the L23 is way too long to keep this thing 4 feet tall. Other than that could you explain how I can calculate the S values such that I have a constantly expanding horn (someone pointed out a flaw in my drawing and I'm trying to fix it).
Looking at this graph I basically conclude that smaller in this case is better. With a max of 2 dB at the very most (38 Hz) difference it makes sense from a logistical standpoint to go with the shorter/smaller box. Can't be building this thing 60 inches long, all my wood is cut to 48.
GM,
True on both points....
Seems like some 8th order boxes can extend bandwidth as in Brian's POC#2 and other configurations sacrifice bandwidth for gain ....
I have also been experimenting with Subwoofer Simulator (by Isaac MCN) specifically for it's ability to sim 8th order boxes(helmholtz only), and even though 3 strong resonances are developed in the model it still has a fairly limited bandwidth because the resonances have to be grouped pretty tightly, or at least that's the case with these Alpine drivers ..... A sim adjusted for a PA style driver might do better in the bandwidth department ....
I suppose the ultimate example of high bandwidth 8th order boxes would be the Karlson styles ..
Thank you David ,
Is the Wizard
more flexible when it comes to added features like these?Much respect & appreciation ,
MMJ
Question: Why in the heck do you want a subwoofer box to go past 100? IT is so much easier and better to run a midrange driver to 100 and let the sub go below than it is to get a reasonably flat response to say 300 Hz in a sub design. Subs go in corners, mains don't sound good in corners most often than not. This and the whole localization issue is why I do what I do (cross at 100, never worry about sub response above that or main response below that).
While I wouldn't cross subs much above 100 Hz either, having frequency and phase response that is decent for an octave or more above the crossover point makes for a better transition between subs and tops, the 100 Hz range is important for "punch".
Personally, I prefer to keep the mains and subs together to keep integration in the critical crossover region as coherent as possible.
Saba,
Disabling S5 makes it really easy to create a constant flare rate, S4 then takes the place of S5 and becomes the new mouth of your TH, and yes L23 will then end up being relatively long especially when using expansion...
The only way to shorten the path while maintaining the box tune is to use no expansion, or reverse taper, or constrictions of some sort (more effective towards the end of the path , like the Alpine 10S4 boxes that you built 🙂 ).
Unfortunately i don't think there is a way to model a long flared pipe followed by a constriction or port before the tap in Hornresponse in TH mode .... At least i haven't been able to figure out a way yet, so that's why you see me using a lot of non-tapered straight pipes in these designs i have posted here ...
Another option for you is to use more than one fold if you want to have expansion and a long path without making the box too tall ...
Last edited:
I mainly am trying to find out what is wrong with the TH model I posted #595. GM notes something is off with my geometry but once that can be figured out I'm pretty sure I have found a design that fits my design constraints and kicks serious butt in the output department (sacrificing 2 Hz extension).
I think I get what he is saying, editing in LW now to get "straight parts" to represent the straight board.
Last edited:
Fixed it! I think. Used LW to make the "expansion path" straight lines to better represent the straight boards I'd like to use (mainly the baffle). Seems I get slightly better low end and slightly less 40+ now but honestly it isn't a big change.
There yah go🙂
I just imported the txt file into my HR .... Looking at it now ...
I see you have already disabled S5 , so thats cool .... Your L34 par will need to be at least 30cm if you are spanning two 12" drivers, Your L12 is already there so that is fine ..Voltage should be in the low-ish 70s to get your cone excursion within range...
I fixed a few things for you ... check out the txt file ..
Your idea for extending the internal panel (baffle) in a single fold box will work for you because it provides a pinch in the path and extends path length just a little bit too, the results will be a downward shift in FB (which you need) and also a closer to flat and damped looking response will result , so there are a few benefits 🙂
An even better way of executing the longer baffle idea would be to move the top of the baffle over a bit (towards the corner) like you can see in the original sketches that began this whole conversation, it extends the path by a little more and gives you some "flare" geometry around the constriction ...
I just imported the txt file into my HR .... Looking at it now ...
I see you have already disabled S5 , so thats cool .... Your L34 par will need to be at least 30cm if you are spanning two 12" drivers, Your L12 is already there so that is fine ..Voltage should be in the low-ish 70s to get your cone excursion within range...
I fixed a few things for you ... check out the txt file ..
Your idea for extending the internal panel (baffle) in a single fold box will work for you because it provides a pinch in the path and extends path length just a little bit too, the results will be a downward shift in FB (which you need) and also a closer to flat and damped looking response will result , so there are a few benefits 🙂
An even better way of executing the longer baffle idea would be to move the top of the baffle over a bit (towards the corner) like you can see in the original sketches that began this whole conversation, it extends the path by a little more and gives you some "flare" geometry around the constriction ...
Last edited:
I'm not running this sub past 60 v for thermal power handling reasons, I've standardized my output curves that way as 60 v = 500 watt peak on the curve and while I could eek out more as far as displacement goes I'm not exactly keen on frying a driver. Basically by being conservative I can safely assume-> this or better performance.
Ok , i thought you were conservative with voltages so it seemed strange when i imported the file (from post #595) into HR and it said 120v in the Eg field ..... Are you meaning to run your two 12s serially in the sim? If so then 120v works , thats 60v per driver ... The imported file had the drivers in parallel for some reason (which is how i would normally sim) ..
Last edited:
Does this look familiar? Kind of like a Karlsonator with Helmholtz resonator HR).
Kvalsvoll Design AS Compact Subwoofer Design Article
Fun with AkAbak - needed to sim the HR.
http://www.kvalsvoll.com/Designs/T138/T138_flyer.pdf
Kvalsvoll Design AS Compact Subwoofer Design Article
Fun with AkAbak - needed to sim the HR.
http://www.kvalsvoll.com/Designs/T138/T138_flyer.pdf
Last edited:
Hi sabaspeed521,
Here is a workup on your design from Post #595, it may not go quite as low as you want it to. You may want to change the Hornresp flare rate from Con to Par. If you want it to go lower you got to make it longer. Make sure the drivers fit.
Regards,
Here is a workup on your design from Post #595, it may not go quite as low as you want it to. You may want to change the Hornresp flare rate from Con to Par. If you want it to go lower you got to make it longer. Make sure the drivers fit.
Regards,