I was simulating 8 speakers by 2 speakers series parallel, hence double the voltage, apparently I forgot to swap it back before uploading.
I did. 😉
Then you didn't follow my instructions as S2 shrinks from 582 cm^2 to 452.3 cm^2 and S3 increases from 925.8 cm^2 to 1084.68 cm^2. 🙁
The sim changes some of course, but in this case not a big deal. I was just trying to educate you [and any newbie lurkers] a bit on how to use HR.
I guess the easiest way to ensure that it matches what you drew [assuming you fold it properly] is to choose some S1, S4 [or S5] dims and set the others at 1 cm^2, then use the LW's auto fill [clicking on the 'manual' options] to calculate the others. Note too that you should use the 'Par' flare.
GM
Last edited:
Since this thread has expanded to include basic THs, here's my default dual SWS with a max 3:1 CR:
GM
GM
These were the details that I needed to figure out my situation. What I did experimentally (after the comment you quoted) was adjust the sim until the schematic looked like a straight line (Not ideal but got me close)
Good to know how to do the calculate way as well.
P.S. Autofill still not working for me 🙁
Last edited:
XRK,
Ill be damned!
That is definitely familiaresque ! I had never seen this particular box before but it has some similarities to the boxes we were looking at earlier in this discussion ... Tapped 6th order series tuned, with a large throat chamber followed by a constricted path which is followed by the tap and mouth..
The amount of driver offset (about 1/3rd of total path) combined with the extra resonator section looks like it has almost entirely suppressed the third harmonic and provided some extended bandwidth .... Neat Stuff! Thanks X , Good find!
Last edited:
I had to modify it a bit because it wasn't a linear horn and I wanted the tune low. Thus I've reduced the taper and have brute forced a "constant expansion" horn (even though I'm pretty sure my model will pinch off a bit and I'll try and model this pinch off point at the top of the cabinet between the baffle and the top panel)
P.S. You're brilliant at autocad, wonderful work displaying the fold and everything. Working on workarounds at the moment to get path longer.
How would I model that 9.041 inch section you have at the top (distance from top of baffle to top panel) as being say 6 inches or so, thus increasing my effective path length but also slightly constricting at that point the air to 2/3 the total cross section (thus increasing velocity by 1.5x)
I realize that this hornresp file simulates the "constricted .75 inch section" right before the driver but how close would this compare to having the constriction at the top? Is there any way to simulate it the way I want?
Last edited:
Hi MMJ,
It all depends upon the nature of the feature being added. Some additions, such as the inclusion of absorbent filling material, are better suited to the Loudspeaker Wizard. Others, such as the System Design tool, are better suited to the main program. It is usually easier to add new features to the main program. The "behind the scenes" operation of the Loudspeaker Wizard is very complex, making it more difficult to seamlessly integrate new features into that tool.
Kind regards,
David
Hi sabaspeed521,
Post #628: "How would I model that 9.041 inch section you have at the top (distance from top of baffle to top panel) as being say 6 inches or so...
As in many of the other models in this thread you are running into the limitations of Hornresp - AkAbak would be the better choice. The enclosure designs are always an iterative process: you sketch, put numbers too it, simulate, sketch again..... 🙂 The only way I know to speed this up is to use AkAbak, and its math functions in such a way that changing one input will automatically change the other connected dimensions in the simulation. xrk971 is the master of this process. I'm just starting to try to use some of these capabilities.
More to the point, once you decide to build a Hornresp model that does not reflect the actual enclosure you have to make choices about where and what to fudge.
I'll attach a Hornresp file and a drawing that might make this more apparent. By using the throat chamber to, e.g.: the middle of the top board, you can add a port of any size to constrict this area. You must then use L12=.01cm, and asign a constant duct cross-section to S1/S2/S3/S4 (in this particular case).
Additionally, you could leave everything as is in the drawing, and rotate the bottom driver to give you a maximum S4, that may improve things a little more.
But, remember when you fudge the numbers in a simulation you may or may not get the simulated result when you start building the enclosure. In this case I think you would be fine, and you could place a litte bit of stuffing into the throat chamber to reduce the peaks. Don't forget the bracing.
Regards,
Post #628: "How would I model that 9.041 inch section you have at the top (distance from top of baffle to top panel) as being say 6 inches or so...
As in many of the other models in this thread you are running into the limitations of Hornresp - AkAbak would be the better choice. The enclosure designs are always an iterative process: you sketch, put numbers too it, simulate, sketch again..... 🙂 The only way I know to speed this up is to use AkAbak, and its math functions in such a way that changing one input will automatically change the other connected dimensions in the simulation. xrk971 is the master of this process. I'm just starting to try to use some of these capabilities.
More to the point, once you decide to build a Hornresp model that does not reflect the actual enclosure you have to make choices about where and what to fudge.
I'll attach a Hornresp file and a drawing that might make this more apparent. By using the throat chamber to, e.g.: the middle of the top board, you can add a port of any size to constrict this area. You must then use L12=.01cm, and asign a constant duct cross-section to S1/S2/S3/S4 (in this particular case).
Additionally, you could leave everything as is in the drawing, and rotate the bottom driver to give you a maximum S4, that may improve things a little more.
But, remember when you fudge the numbers in a simulation you may or may not get the simulated result when you start building the enclosure. In this case I think you would be fine, and you could place a litte bit of stuffing into the throat chamber to reduce the peaks. Don't forget the bracing.
Regards,
Last edited:
Saba , there are a few different ways to manipulate the path at the turn in a single fold cabinet ..... I grabbed this sketch from early in the discussion so you can see what i am talking about .... You could make the constriction mild to keep velocities within safe range but fashion the "port" to have some length for the purpose of increasing it's effectiveness ... The version of this with no expansion (like shown in the sketch) is really easy to sim in HR using the throat chamber as the first section & port, and may get the total cabinet height down to where you want it ...🙂
I haven't simmed this yet with your SWS12s but it may be an option worth looking into ...
I haven't simmed this yet with your SWS12s but it may be an option worth looking into ...
Attachments
I see .. .So it all depends ... Ok very interesting ....
I had some ideas for added features but im sure people keep you terribly busy with this HR project as it is (just going by what i have seen in the massive Hornresponse discussion here on DiyAudio)...
If things ever slow down for you on your end just let me know and i can run these ideas by you to see what you think ... .
By the way , i found a way to do 10th order sims in HR ! 😀 It looks like both chamber types (throat , and rear vented) can be used simultaneously as long as a port isn't needed on the rear chamber ... So i came up with a few sims that look good and then hand sketched them up to see if they can be turned into something physical, and they can! I will produce a digital sketch and post them here in the next few days ... Even though they are 10th order , the build complexity isn't increased by much, they still look pretty simple ... Fully,
MMJ in Arizona
This looks real familiar and pretty sure I saved the patent, but can't find it. 🙁
GM
edit: your sketch
GM
edit: your sketch
You really need to give AkAbak a try - up to 52 nodes or 51 elements. Just get a virtual machine running in your Win64 bit first.
INDEED ! You are right , i'm just seeing how far i can take things in HR first , then i will start learning Akabak ...... I have some computers around here that still run on the old XP pro ... should work on those ...
So guys, here are my numbers and semi-final sketch of what I'd like to build. The angle has been decreased in order to get a lower tune at the expense of spl and the SA segments are straight off the measured model. Including L23 aka path length.
I'm open for bracing suggestions (angled cuts aren't really a favorite of mine but I can do them... Had a few annoying braces to do on the T60, would rather avoid segmented braces if at all possible).
I'm open for bracing suggestions (angled cuts aren't really a favorite of mine but I can do them... Had a few annoying braces to do on the T60, would rather avoid segmented braces if at all possible).
Attachments
Last edited:
GM ,
are you referring to the T shaped baffle? or the 10th order thing?
Somebody has surely patented some version of either or both of them over the years ..... The high order pipe box is intriguing for sure, i am getting quite a bit of gain just by adding an extra 20 liter (2 foot long) strategically placed "rear chamber" to a 55 liter box ... Looks pretty useful so i am sure that someone has utilized it at some point ....
Here is a relevant quote:
"There is no such thing as a new idea. It is impossible. We simply take a lot of old ideas and put them into a sort of mental kaleidoscope. We give them a turn and they make new and curious combinations. We keep on turning and making new combinations indefinitely; but they are the same old pieces of colored glass that have been in use through all the ages."
- Mark Twain
Attachments
Last edited:
The 'T'.
I've seen 10th order DIY alignments, but no patents come to mind unless the Graham-Holliman velocity coupled Infra-bass speaker complies once coupled to the floor/wall junction. There was at least one long thread on it.
With a 36+% increase in net Vb plus in a resonant pipe, I certainly hope so.
GM
I've seen 10th order DIY alignments, but no patents come to mind unless the Graham-Holliman velocity coupled Infra-bass speaker complies once coupled to the floor/wall junction. There was at least one long thread on it.
With a 36+% increase in net Vb plus in a resonant pipe, I certainly hope so.
GM