Hi Jennygirl,
Post #151: "...fits within a smaller box with less than half the amount of wood panels (39w x 24h x 38.5d)..."
I'm trying to follow your thread, and I'm having trouble w/ some of the design parameters. Is the above the maximum box size? Is the FaitalPro 15HP1060 still the drive of choice? What is the source of the T/S parameters you are using?
By the way, I think the OP can edit the first post, and enter references and updates.
Interesting thread, Hornresp should be fine for this design.
Regards,
Post #151: "...fits within a smaller box with less than half the amount of wood panels (39w x 24h x 38.5d)..."
I'm trying to follow your thread, and I'm having trouble w/ some of the design parameters. Is the above the maximum box size? Is the FaitalPro 15HP1060 still the drive of choice? What is the source of the T/S parameters you are using?
By the way, I think the OP can edit the first post, and enter references and updates.
Interesting thread, Hornresp should be fine for this design.
Regards,
I just checked and if the distance between your parallel walls is 37.5 inches and if a half wave notch is the result it will occur at 180 hz. That might not be a bad thing, it might significantly reduce a problem peak in your response. I guess you don't need any solid bracing at all, you could use stick or dowel bracing exclusively if you wanted to, it would save a lot of internal volume.
The fastest way for me to code a horn with lots of turns in AkAbak is to print out a to-scale drawing and take a dial caliper to all the main dimensions (s-start, s-end, L) for each segment including the wedges that make up a turn. I have a basic wrapper script for horns that I just stick the dimensions in and usually it takes about an hour for horn like this. Let me know which drawing is the final design and I can give it a shot. Or even better if you can give me a table of all the dimensions of your advanced centerline path I won't need to use dial calipers and it will be even easier.
In post #19 I show the drivers side by side (fore and aft in relation to the truck), not one after another.Art- so when positioning the two drivers, they would be one after the other in the horn path... So how do I calculate L12 & L34? The midpoint between the two driver's centerpoints?
If you do lay them out one after the other, the midpoint is the virtual center, but with an evenly expanding horn path the two drivers will be stressed unequally, so I would not attempt it, unless doing a "stub" (a "V" contraction between drivers) as done in some of the DSL "in line" designs.
As far as driver sag, it won't be a problem if you rotate the cabinet once a year, or whenever you notice the cones dropping a bit.
Last edited:
Hmm... I think I've just identified my next spreadsheet project 🙂
I already have my spreadsheet dumping out the HornResp params corresponding to a horn layout. It shouldn't be that difficult for it to generate the corresponding Akabak script as well.
Yes and yes, and I got the t/s params from the spec sheet for the 15HP1060. Though I calculated cms & mmd within hornresp.
I didn't know I could edit the first post, that is helpful to know! Thank you 🙂
That's really interesting to know. I will stick with dowels for the unseen parts and do a slanted brace visible in the mouth.
Not quite done with the final look of the visible brace, but something like this:
I may add some holes or something too. I want it to look really neato 😉
I'll have the final done soon, I had to pick someone up at the airport so I was interrupted temporarily in my CAD process. Shouldn't take me long to get v2 done though as the C4D part is already complete (as seen above). Thanks for offering this up! Totally will be taking you up on it!
Okay sorry I must have been confused. Glad to know I am indeed doing it the right way, now! It's the only way it fits anyway!
Sounds like I am doing something similar, I have a spreadsheet that interpolates the data for entry into C4D. Then I have a very visual and reactive aid to see if something is going to fit because I can just drag the horn path spline around. I can't wait to see how accurate it is, but I have hopes that it will be just shy of perfect!
If you don't need some of the more advanced features of the 2x8 (like longer delays) using qnty 2, 2x4 miniDSP vertically integrated is less expensive. Each miniDSP goes to left or right channel and you now have 4 channels and 2 software selectable inputs. I am in a similar situation in that I just built a pair of 2-way tractrix synergy horns and want to run them with subs so I need the extra channel. A pair of 2x4's is $160 vs 2x8 is $300.
I'm very interested to see how accurate it is too, and to find out if your software is handling it properly. There was a thread awhile ago that addressed my concerns.
http://www.diyaudio.com/forums/subwoofers/247110-horn-folding-couple-questions.html
The premise is that even drawing out the straight horn geometrically (before folding) can introduce error. So you get your software to plot out the length vs cross sectional area data, so far so good. Then you get the software to draw it out as a straight unfolded horn and you get this.
Just the way it's drawn might be a problem.
There are a couple of different ways to draw it out, as shown here. The bottom one is drawn symmetrically, the top one is not.
An externally hosted image should be here but it was not working when we last tested it.
Both drawings have the same physical length from end to end but they don't have the same centerline length. This is a tricky but important distinction.
If your software lays it out in a symmetrical manner like the bottom pic the centerline length should be correctly corresponding to your Hornresp sim. But if the software draws it out like the top pic and uses the flat bottom of the horn flare as the path length the centerline length will be a bit too long. So there is a potential for error there.
But then in post 140 you said "Had to go back to make a small adjustment: once CAD'd I rechecked my advanced centerline and it was a few cm short." Which is the opposite of what I would have expected.
So I'm not sure how your software is dealing with this issue. And I'm also not sure if the way it lays out bends is accurate. This is why I draw everything out panel by panel and measure as I go. If you want the software to give shortcuts you have to be absolutely sure what it's doing.
I'm very interested to see if your process is accurate or not and we'll find out when you finalize your design. I'll need exactly the same info xrk971 needs for the Akabak sim. Then you'll know whether the plans match the sim and you'll have a more detailed sim than Hornresp can do, done with inputs directly measured from the plans.
Last edited:
Not sure exactly what you are up to from that description but it sounds interesting.
xrk- here is the advanced centerline
not sure if the path dims help that much, let me know if there's something you'd like to see.
Or I'd be happy to shoot you a DWG, OBJ, or STL file. Not sure what akabak can import as far as geometry
not sure if the path dims help that much, let me know if there's something you'd like to see.
Or I'd be happy to shoot you a DWG, OBJ, or STL file. Not sure what akabak can import as far as geometry
guy- here is the v2 version path scale in cinema, you can see that the Auto S2/S3 translated perfectly from hornresp
here is a more uncluttered version of the horn path if it helps
And the final v2 hornresp sim parameters
And the final v2 hornresp sim parameters
Last edited:
We need the cross sectional area (or at least the width) at each purple line as well as the beginning and end of the horn. Cross sectional area including subtraction of area taken up by bracing.
Also we need two more purple lines to show the driver tap points, with lengths and cross sectional areas for those.
Each of those purple lines are segment markers and we need to know the cross sectional area at those points.
Akabak is 20 years old, it can't import much of anything.
Last edited:
Yes but that's not what I was talking about. Your X axis horn length is 1.0. Your latest Hornresp sim has a length of 352.9 cm. If 1.0 on the X axis = 352.9 cm then your centerline length (drawn here in red) can't be 352.9 cm. If the red line doesn't equal 352.9 cm the drawing doesn't match the sim. Like I said, I don't know how your software is dealing with this but we'll find out if your plans are accurate.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
don't worry guy 🙂 I've got you covered!
What I do is interpolate/scale the 0-352.9cm horn length to a 0-1 "percentage", which cinema's sweep nurbs object scale parameter reads.
Here is the spreadsheet cell formula.
I am also converting from cm to inches first for some reason
What I do is interpolate/scale the 0-352.9cm horn length to a 0-1 "percentage", which cinema's sweep nurbs object scale parameter reads.
Here is the spreadsheet cell formula.
I am also converting from cm to inches first for some reason
Here is the build sheet for v2
Exterior box dimensions 39" x 24" x 38.5"
response @ 49V with 27hz 48dB BW HPF
Exterior box dimensions 39" x 24" x 38.5"
response @ 49V with 27hz 48dB BW HPF
Last edited:
response with 27hz 48dB BW HPF
What does that frequency response curve look like when you increase the driver's Re by about 50%?
Driver Re set to 4.35 instead of 2.9
Actually looks a bit flatter to me, and still louder & lower F3 than the vented box
xrk- here is the advanced centerline
not sure if the path dims help that much, let me know if there's something you'd like to see.
Or I'd be happy to shoot you a DWG, OBJ, or STL file. Not sure what akabak can import as far as geometry
Thanks - looks great. Can you get me the dimensions for all the purple lines representing the areas at each segment?
Oy, a little congested in there sorry about that
Here you go
And thank you! Very gracious of you to offer doing this 🙂
Let me know if you need the other lines too, I only included the ones where the horn path notches