Very nice design Jennygirl. Thanks for sharing. Reminds me of this design which has two mouths and common push pull center.
http://www.diyaudio.com/forums/subwoofers/258706-study-dipole-cardioid-bass-horn-18.html#post4009376
http://www.diyaudio.com/forums/subwoofers/258706-study-dipole-cardioid-bass-horn-18.html#post4009376
An externally hosted image should be here but it was not working when we last tested it.
Wow, that was fast. I've never seen anyone go from being unsure of t/s parameters entry into Hornresp to a fully complete folded and CAD'ed design in less than four days before.
Are you going to let me check the fold for accuracy? If so I just need a couple of things from you, I'll let you know what they are if you are going to let me. The info I require won't take you more than a couple of minutes to collect and I'll have an answer for you within 15 minutes or so of receiving it so it won't slow down your building schedule at all.
I'm off to work now so let me know if it's ok and when I get back I'll let you know what info I require.
Jennygirl,
Yes, the symmetrical dual path uses up more wood for a given path length, which is why I suggested the simple fold pattern in post #19 that also has a mouth with a low aspect ratio, using a fraction of the parts with nearly the same path length as your dual path design. J.A.G's post #122 also shows another single path approach, though for your given size restraints and driver size my suggestion matches better.
Since I tried the Keystone (same fold pattern in post #19) out with dual drivers and with a variety of exit shapes, I am confident it would work well, with minor tweaks, the bracing was not shown for clarity, and though it is pretty close to your size restrictions, it is not exact. It also has the potential of easily adjusting the exit size and shape, which can adjust the F3 over a fairly wide range as well as change the upper response. That said, your fold will work well too, just harder to build using more wood, heavier, and not easily adjustable if you do want a lower F3. The compression ratio also seems a high considering your desire for clean low frequency reproduction.
Art
Last edited:
xrk- whoa okay, yeah those multiple folds are nuts! Makes me feel a bit better about my design, lol.
Guy- thanks so much for offering! I am curious to see how accurate my Cinema4D folding guide setup ended up. I was willing to give it the benefit of the doubt, but if you are offering to check I cannot turn it down. Let me know what you need and I'll happily post it.
Art- I just never understood how that will fit my size requirement, but the part about a wide range F3 and simple build are enticing me enough to give v2 a shot regardless. It is time to whip up a new sim for a possibly better alternative. Heck, if I could reach a 30hz F3 there is absolutely no way I can turn that down... pun intended 😉
Just wanted to say THANKS for all the help from everyone. I have had way more response for this project than I expected. It has been one of the main reasons I've been able to do this so quickly. Seriously you guys rule! Your comments and help are extremely appreciated.. I can't really thank you enough with words alone.
Guy- thanks so much for offering! I am curious to see how accurate my Cinema4D folding guide setup ended up. I was willing to give it the benefit of the doubt, but if you are offering to check I cannot turn it down. Let me know what you need and I'll happily post it.
Art- I just never understood how that will fit my size requirement, but the part about a wide range F3 and simple build are enticing me enough to give v2 a shot regardless. It is time to whip up a new sim for a possibly better alternative. Heck, if I could reach a 30hz F3 there is absolutely no way I can turn that down... pun intended 😉
Just wanted to say THANKS for all the help from everyone. I have had way more response for this project than I expected. It has been one of the main reasons I've been able to do this so quickly. Seriously you guys rule! Your comments and help are extremely appreciated.. I can't really thank you enough with words alone.
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?
The problem I am seeing is that the horn path length is limited severely with only 3 folds. In order to get decent F3 I need a ~340cm (134in) total length. That is just not going to happen with a box that has a 37x23 cross sectional area and low compression
Unless I am missing something?
Unless I am missing something?
Last edited:
Okay I am getting close, but only if the drivers are oriented 90 degrees, as in at the same point along the horn path instead of one after the other. It is closely resembling my drawing in post 16.
Why is that orientation of drivers any different from the one in post 147? Doesn't seem to make any sense to me why it would matter?
Why is that orientation of drivers any different from the one in post 147? Doesn't seem to make any sense to me why it would matter?
Alright, lol..
With the fold of the first drawing I had proposed (in post 16 similar to Art's suggestion), I am able to get a longer horn length and drastically lower F3, perfect parabolic expansion (S2/S3 set to Auto), fits within a smaller box with less than half the amount of wood panels (39w x 24h x 38.5d), and 2.94:1 CR.
I know this design would have issues like driver sag and it would need massive amounts of bracing... But look at that F3! And it would be so much easier to build. No doubled up panels, either 🙂
w/ HPF excursion limiter
FS215 v1 (grey) vs v2 (black)
**rough** fold pattern
With the fold of the first drawing I had proposed (in post 16 similar to Art's suggestion), I am able to get a longer horn length and drastically lower F3, perfect parabolic expansion (S2/S3 set to Auto), fits within a smaller box with less than half the amount of wood panels (39w x 24h x 38.5d), and 2.94:1 CR.
I know this design would have issues like driver sag and it would need massive amounts of bracing... But look at that F3! And it would be so much easier to build. No doubled up panels, either 🙂
w/ HPF excursion limiter
FS215 v1 (grey) vs v2 (black)
**rough** fold pattern
Last edited:
Not sure what you are asking - why what would matter? But if you want to know about the different L12 for the different configurations ...
When you specify L12 you are describing the distance from the beginning of the horn to the acoustic center of the driver(s). Since you can't sim dual drivers in Hornresp we have to take the consider the two drivers as one and the acoustic center is in the middle.
Some people have commented that this isn't as accurate as simulating it with two drivers but to do that you need Akabak. I haven't tried an Akabak sim of this configuration yet since I haven't been really interested enough so I don't know how much difference it makes.
This is fine, and there's other layouts that will give similar results. Your dip at 100 hz is getting deeper, if you are concerned about that you need to adjust the last segment's length.
It doesn't need massive bracing but the bare minimum is a single solid brace running down the entire path length to break up the side to side resonance and a bunch of stick or dowel bracing to finish up.
If the driver passes the driver sag test (5% or less of xmax is the limit I think) then it's fine, just reduce xmax by that much in your sim and every once in awhile flip the box over to counter long term damage.
I've mentioned this before, but if your S2/S3 was not set to auto for ANY of sims that you posted a CAD pic for, the fold was wrong. This is a picture of what it would look like with S2 on Manual - there would have to be a change in your cross sectional area at that point, and NONE of your drawings show a change in cross sectional area there.
An externally hosted image should be here but it was not working when we last tested it.
I thought you could sim multiple drivers, here?
Correct but you still have to set L12 according to the acoustic center of the driver(s). That's going to be the beginning of the flare to the midpoint of the driver(s) in any design because Hornresp can't simulate dual drivers as dual sources, it can only simulate a single source.
Last edited:
I realized that the model didn't perfectly match the sim, but considering how much human error goes into actually building a box it did not concern me that much. A few mm's here and there doesn't necessarily make or break a build, and that is all I was seeing when converting the path to single straight lengths. I have read in multiple threads that sometimes a bit of error results in a more ideal response, so I guess that's where I was basing that method of operation from.
The v1 design was very very close to the actual sim, but definitely not dead on. However nothing near the extreme you just posted. I got it as close as possible within the constraints. Unfortunately it looks like I won't be building that design so I won't be able to quantify my thoughts on this.
If hornresp had more sectional inputs it seems it would be possible to perfectly sim any design.. I'm assuming that is what akabak excels with?
Either way, v2 is auto all the way down the horn path. And the dip at 100hz does not worry me much. The PEQ on the minidsp can fix that right up 🙂 As well, I will build the mids that cover that small gap.
Onto CAD for v2!
The v1 design was very very close to the actual sim, but definitely not dead on. However nothing near the extreme you just posted. I got it as close as possible within the constraints. Unfortunately it looks like I won't be building that design so I won't be able to quantify my thoughts on this.
If hornresp had more sectional inputs it seems it would be possible to perfectly sim any design.. I'm assuming that is what akabak excels with?
Either way, v2 is auto all the way down the horn path. And the dip at 100hz does not worry me much. The PEQ on the minidsp can fix that right up 🙂 As well, I will build the mids that cover that small gap.
Onto CAD for v2!
With v2, I'll be placing the drivers at equal distances along the horn path (instead of one behind the other as Art suggested). Thus there shouldn't be any change in midpoint, makes perfect sense.
btw, the lower the compression ratio, the lower your LC is generally going to be for any givin design, but you sacrifice both sensitivity and output (higher compression usually means lower excursion for given PE).
Thanks sine, I did notice that. With these 15HP1060s they seemed to want a higher CR to keep them within usable FR (lowering compression put it into dangerous territory as far as excursion). 3:1 seems about the right trade-off area.
v2 has 2.94:1 CR
v2 has 2.94:1 CR
As long as you know the risks it's ok to have an optimistic viewpoint on errors. As long as you are prepared for the fact that it could go the other way too. As you mentioned there's human error in construction, sometimes imperfect materials (slightly bent plywood) and then there's a bunch of real world considerations Hornresp doesn't account for. If you add all that up there's a potential for far less ideal response than expected. Or you could get lucky and see an improvement.
I know my drawing was accurate or to scale, I exaggerated the bend quite a bit to make it stand out. Since you were still using the Manual S2 I just wanted to make sure you knew what a bent flare at S2 looked like and why your drawings didn't match the sim.
Yes, and it's got a few other features too. Unfortunately it's not perfect though, and it's a slow process to make a script.
Last edited: