Hi MMJ,
Hornresp assumes that the value of driver voice coil inductance (Le) is constant at all frequencies. The inductive reactance still increases with frequency, though.
Kind regards,
David
MMJ - can you fudge K15's response from 200Hz down with HR?
http://classicaudio.ru/forum/uploads/monthly_08_2011/post-329-1385471361,079.png
http://classicaudio.ru/forum/uploads/monthly_08_2011/post-329-1385471361,079.png
Hi MMJ,
Unfortunately it would require much too much work to integrate the necessary changes into Hornresp - most parts of the program would be affected, one way or another.
AkAbak is better suited to the "weird & wacky stuff".
Kind regards,
David
Ok, I guess i will just have to stick to a reasonable level of weird & wacky then
... heheI suppose i will eventually give Akabak a try but i am being stubborn and putting it off because i am having so much fun with Hornresponse , i probably wont move on until i feel that i have explored and exhausted all possibilities & configurations with HR , and im not there yet.
I really appreciate the versatility of your software David ... 
I think i have figured out how to sketch this quasi-8th order sim (the one that uses the "rear vented" chamber in TH mode) ..... Seems to perform similarly to a Tapped Horn (TH with expansion) with the right drivers but with a much simpler build , no angles and no complex geometry, might be little bigger though ...Seems to work well with Saba's SWS12..... Could be a nice option for those like myself who possess just basic woodworking skills ...
Freddi ,
I have never heard a classic K box in my life, much less tried to sim or build one but from looking at the diagram that you posted it really looks like there could be a lot going on inside of that box .... It appears to be be as many as 4 vents or constrictions (if we include the mouth) and 4 chambers , making something like a 10th order if all chambers & vents are actually functional .... I cannot imagine the amount of mathematics involved or trial & error sawdust that was required to come up with this thing all those many decades ago, wasn't that thing developed sometime in the middle of the last century? Thats quite an achievement considering that there was no software used in it's development!
Any approximation that i could make in HR would be an inaccurate attempt , but i would use HR in TH mode with the throat chamber followed by a short port opening up little bit around S1 & S2 then another squeeze at S3 followed by flared segments at S4 & S5 , which would be 8th order at best , not 10th order.. There is no guarantee that all resonances will be accounted for in sim because i don't think (im assuming) that HR wasn't designed for such a thing which actually leads to a question that I had for David revolving around the extra midbass resonance (seemingly) developed by Brian Steele's DFD constriction in his PC#2 box .... He tuned it by sliding the constriction around until the additional midbass resonance fell right between the 3rd and 5th harmonic of the main pipe resonance effectively filling in the dip ....
Brian made his TH into an unquestionably 6th order series tuned alignment in the traditional sense with his mod creating a long chamber feeding a shorter chamber in series , divided with a constriction .... In the helmholtz model (large chamber feeding smaller chamber in series) the second chamber is responsible for generating the upper resonance ... I am thinking that Brian just came up with a really crafty way of making this happen in a TH which will of course outperform the helmholtz version (in both SPL and bandwidth) by a significant margin thanks to it's quarter wave loading .... This solution doesn't take up any additional space so the box remains the same size which i think is kind of exciting from a design standpoint ..
My question for David is: should HR acknowledge that additionally generated resonance, or could it?
Last edited:
K is considered 8th order? ok then , i could try to model it as an 8th order .... The contours of the throat chamber and second chamber wont be quite right in the sim but maybe it wont matter too much since the effective path length on that box is pretty short and it may be operating largely in helmholtz mode .... I am willing to give it a shot if you can give me the exact net box volume, chamber volumes and lengths, also the vent areas in sq cm along with the area of the K-vent...
The big K-vent is pretty interesting because when it comes to bass frequencies it will be seen as a flared segment with some path length sacrificed , or at least thats how i would attempt to emulate it in HR . .
Anyway GM, if you can get those figures for me i can create a sim and it could be fun to compare the sim's response results to any measured response charts that people may have for this box ..
Last edited:
You can't model the K aperture in HR. If you take a Karlson K15 and strip off the aperture you might get close with an 8th order bandpass. But small changes in the K aperture narrow slot portion have large impact on the bass extension. As far as I know, my AkAbak model is the only one that accounts for the K aperture. If you want to model it, the complete dimensions are available. Just search this forum for Karlson K15 and Freddi. Plans also available on Job Ulfman's Karlson website. Pretty straightforward geometry. Details of my model are in the Karlson thread http://www.diyaudio.com/forums/full-range/213594-karlson-10.html#post3528706
and Speaker that Kicks Butt thread http://www.diyaudio.com/forums/full-range/237948-speaker-kicks-butt-large-spaces.html
and Speaker that Kicks Butt thread http://www.diyaudio.com/forums/full-range/237948-speaker-kicks-butt-large-spaces.html
Last edited:
Hmm, I don't remember it that way, but nowadays that doesn't mean much.
Bill Woods [RCA-fan] did some measurements with the K-slot Vs just a square vent. I didn't save them, but I bet Freddy did, so hopefully he'll still have them for us. They use to be these and much more on his HE forum Gallery, but most are gone now.
Speaking of Bill.....all the links I have for his horn company, various plans he posted are all either gone or empty ghost sites, so what happened?
GM
That's what I figured. I suppose it could be possible to approximate the effect of box losses for a properly sealed and braced enclosure by including a very light percentage fill in the model - I think I might try that out this weekend to see if the results give a closer approximation to the measured impedance curve for my POC3.
Well, here's the results. I basically fiddled with the filling until I got the best approximation to the measured results for the two of the three major peaks in the impedance response. The corresponding FR curve shows that the passband has been smoothed a little and there's a projected loss of just under 1dB in the middle of the passband.
Perhaps this might be a good enough means of more closely approximating what the real-world results might be like if you build a properly-braced TH following a HornResp model. I emphasize PROPERLY-BRACED. If your build is not properly braced and the panels are resonating like drums within your TH's passband rather than having a more "broadband" effect, then you're on your own, LOL.
Of interest is the first response peak in the impedance curve. Real-world measurements have placed this of course lower in amplitude but interestingly enough higher in frequency than the HornResp model. As this peak is noticeably impacted by even a pinhole leak around the driver, I think this variation is due to both box losses and driver losses, primarily due to the driver's suspension, which is lossy linen cloth for most pro-audio drivers. If this is true, it may be possible to model the impact on the impedance curve and therefore the FR curve by inserting a very lossy port between the throat and rear chambers in the model.
Finally, while examining the measured impedance curve of my POC3, I didn't see any evidence that HornResp's use of Le in its model was incorrect - the measured impedance curve at higher frequencies was a near match for the one predicted by HornResp.
Attachments
Splendiferous! This is some smart software
I just downloaded the new version and the filling feature in TH works great, and the throat chamber prompt in main/wiz behaves perfectly now
I really like the feature that you built in which allows an adjustment of where the stuffing can be placed within the segment ... very nice..Will the filling ever be shown in main? or will it always be a feature for just the wizard specifically?
Last edited:
Great investigative work Brian!
By the way, i am curious, you said you had modeled your POC#2 DFD box in akabak, did it show the proper filling in of the midbass dip in that simulation? and if so was the constriction placed about 2/3rds of the way down the pipe in the model?
Last edited:
Hey guys for those that are interested. This is my (very) rough sketch of the Dual 12 TH I'm most likely going to build with the sws12. Just going to determine the angle required to get the distances and set the baffle length to set the path length. What negative effects will I get from slightly pinching the air near the top of the cabinet (between the top panel and the baffle pieces) I believe that the model assumes constant expansion but in order to get a little more path length I was thinking that I would just make the gap a little smaller (still leaving it ~4-5 inches tall gap-wise) to get the path length up to 186 cm (L23).
Attachments
Hmm, what you've drawn doesn't match up with the HR alignment, i.e. you show a straight divider board, but your sim has [3] different flare rates, so to correct it to the drawing requires using the LW to auto correct the S2, S3 expansions by clicking on the [2] 'manual' options, then 'save'.
GM
GM
Saba made a good choice in alignments though, 2 of those SWS12D drivers look really nice in a tapped horn with the drivers spanned ...Mutiple expansion rates are not necessary, one rate of expansion works fine ..... I was able to make it work well in about 265 liters ... Peaks at just 1/2 decibel short of 130 @ 41hz and is only a few db down at 35hz ...
Here is the set of inputs that i ended up with when trying out this alignment to see if it had potential for these Alpines ...
It ends up being very close to this quasi-8th order box i was working on , same size package , one just uses expansion and the other doesn't (it uses and additional chamber + port instead , will post sketch and inputs for that box here in a bit) ...
Here is the set of inputs that i ended up with when trying out this alignment to see if it had potential for these Alpines ...
It ends up being very close to this quasi-8th order box i was working on , same size package , one just uses expansion and the other doesn't (it uses and additional chamber + port instead , will post sketch and inputs for that box here in a bit) ...
Attachments
Last edited:
8TH ORDER SIMULATABLE IN HR?!?! FASCINATING!
Ok guys , what do you make of this? More reasonably weird & wacky stuff that HR is capable of simulating for us ..
I know the inputs and sketch could use a lot of refinement, and perhaps there is a better fold for this but i just wanted to post these for the sake of discussion and for the novelty of the concept .... This box has equal output to a tapped horn (of the same size & tuning ) with expansion, but there is no expansion in this pipe whatsoever, the gain (over a standard tapped pipe) is generated from the way the added chamber interacts with the rest of the system ....... Could be a good option for someone who wants Tapped Horn performance out of an easy build that has the simplicity of nothing but a few 90 degree angles!
If i were to use a constriction like a port or a vent of some sort on the main pipe (the one tuned to the fundamental)placed somewhere before the tap it could allow us to shorten the path length a bit and could generate an additional resonance as well ... Configured in this way it could be considered a 10th order box!
Ok guys , what do you make of this? More reasonably weird & wacky stuff that HR is capable of simulating for us ..
I know the inputs and sketch could use a lot of refinement, and perhaps there is a better fold for this but i just wanted to post these for the sake of discussion and for the novelty of the concept .... This box has equal output to a tapped horn (of the same size & tuning ) with expansion, but there is no expansion in this pipe whatsoever, the gain (over a standard tapped pipe) is generated from the way the added chamber interacts with the rest of the system ....... Could be a good option for someone who wants Tapped Horn performance out of an easy build that has the simplicity of nothing but a few 90 degree angles!
If i were to use a constriction like a port or a vent of some sort on the main pipe (the one tuned to the fundamental)placed somewhere before the tap it could allow us to shorten the path length a bit and could generate an additional resonance as well ... Configured in this way it could be considered a 10th order box!
Attachments
Last edited: