Hi all,
I've been working on a design for a really, really big set of subwoofer horns as a little summer project. I want to prove to myself whether or not doing this is a feasible solution for a semi-permanent installation. I've really set myself up for failure in a lot of ways, but some pretty gnarly circumstances have allowed me to take that risk.
Each horn is going to measure 5'x5'x2', should weigh about 450 pounds, and they'll be used in stacks of 4. I chose the B&C 15NW100 for the drivers (one per horn).
This is the design:
And these are the hornresp inputs I'm using
Just figured I'd open myself up to some ridicule before starting construction next week.
If (and it's a big if) everything goes according to plan, four of these should be able to put out a pretty flat ~20 Hz to 140 Hz at 140 dB in 1/2 pi space off of 100w input each. If its pi space, that's down about 3 dB.
In reality, I'm probably going to rip my driver's cone to shreds the minute I plug it in. I found a startling lack of literature on that, so I'm really just going to have to take it slow and cross my fingers.
I welcome any and all feedback, and I'll update the thread once things get built and tested.
I've been working on a design for a really, really big set of subwoofer horns as a little summer project. I want to prove to myself whether or not doing this is a feasible solution for a semi-permanent installation. I've really set myself up for failure in a lot of ways, but some pretty gnarly circumstances have allowed me to take that risk.
Each horn is going to measure 5'x5'x2', should weigh about 450 pounds, and they'll be used in stacks of 4. I chose the B&C 15NW100 for the drivers (one per horn).
This is the design:
And these are the hornresp inputs I'm using
Just figured I'd open myself up to some ridicule before starting construction next week.
If (and it's a big if) everything goes according to plan, four of these should be able to put out a pretty flat ~20 Hz to 140 Hz at 140 dB in 1/2 pi space off of 100w input each. If its pi space, that's down about 3 dB.
In reality, I'm probably going to rip my driver's cone to shreds the minute I plug it in. I found a startling lack of literature on that, so I'm really just going to have to take it slow and cross my fingers.
I welcome any and all feedback, and I'll update the thread once things get built and tested.
Hi dragospk01,
You really need a true 0.5 PI Corner placement of Four Cabinets in order to reach 20 Hz at 140 dB..
Waiting for the forthcoming measurements of FR and Phase Response:
...The very minimum needed for me to return to this thread...
b
Attachments
Hi dragospk01,
You really need a true 0.5 PI Corner placement of Four Cabinets in order to reach 20 Hz at 140 dB..
Waiting for the forthcoming measurements of FR and Phase Response:
...The very minimum needed for me to return to this thread...
b
Yep, will do! Distortion is definitely something I'm worried about, as well as spatial loading. You really hit the nail on the head with that last comment; I'm doing this as a proof of concept rather than for any practical purpose. I'm expecting my fold to not match perfectly to the sim too.
bjorno, just a heads up, you entered "0.43" in the RMS field. OP used 8.43 there.
IMO, it makes sense to go smaller per driver, with dual loaded cabs.
50x52.5x31.5 external should fit (or you could split them into individual cabs, at 16.5 external width).
Heres a comparison of what I can only assume is your cabinet x4 with 60v input (2pi, which is akin to on the ground outside) at 1200 liters per driver, vs a quick sim I adapted from a "very large horn" I recently drew up for the BC 21SW152. It takes 60v for your cab to hit xvar "in passband" with a 24hz hipass My sim is about 600 liters per driver (half the size). Its dual loaded, so I've shown 2 cabs (4v4 drivers), fits in the previously described dimensions, and is shown with 65v, the input it takes to bring the driver to xvar at the low corner with a 28hz hipass.
You can see my cab is clearly tuned higher. but at half the physical size per driver, its no slouch. Its response is much smoother, and while it is less sensitive between 40-50hz, and below 28hz, I think the space savings are worth it, and honestly I think it will "sound better".
some notes on your design however.
Increasing compression ratio to nearly 4:1 (reducing S1 to 215), decreasing s2 to around 565, and increasing rear chamber volume to 100 liters appears to do WONDERS to "smooth response". 4:1 CR is approaching the "no fly zone", but it lowers the low corner group delay peak by almost 10ms.
Just a note, the group delay on your cab at the low corner is insanely high at 60ms. my cab peaks out at 33ms.
IMO, it makes sense to go smaller per driver, with dual loaded cabs.
50x52.5x31.5 external should fit
(or you could split them into individual cabs, at 16.5 external width). Heres a comparison of what I can only assume is your cabinet x4 with 60v input (2pi, which is akin to on the ground outside) at 1200 liters per driver, vs a quick sim I adapted from a "very large horn" I recently drew up for the BC 21SW152. It takes 60v for your cab to hit xvar "in passband" with a 24hz hipass My sim is about 600 liters per driver (half the size). Its dual loaded, so I've shown 2 cabs (4v4 drivers), fits in the previously described dimensions, and is shown with 65v, the input it takes to bring the driver to xvar at the low corner with a 28hz hipass.
You can see my cab is clearly tuned higher. but at half the physical size per driver, its no slouch. Its response is much smoother, and while it is less sensitive between 40-50hz, and below 28hz, I think the space savings are worth it, and honestly I think it will "sound better".
some notes on your design however.
Increasing compression ratio to nearly 4:1 (reducing S1 to 215), decreasing s2 to around 565, and increasing rear chamber volume to 100 liters appears to do WONDERS to "smooth response". 4:1 CR is approaching the "no fly zone", but it lowers the low corner group delay peak by almost 10ms.
Just a note, the group delay on your cab at the low corner is insanely high at 60ms. my cab peaks out at 33ms.
Attachments
Last edited:
bjorno, just a heads up, you entered "0.43" in the RMS field. OP used 8.43 there.
IMO, it makes sense to go smaller per driver, with dual loaded cabs.
50x52.5x31.5 external should fit
Heres a comparison of what I can only assume is your cabinet x4 with 60v input (2pi, which is akin to on the ground outside) at 1200 liters per driver, vs a quick sim I adapted from a "very large horn" I recently drew up for the BC 21SW152. It takes 60v for your cab to hit xvar "in passband" with a 24hz hipass My sim is about 600 liters per driver (half the size). Its dual loaded, so I've shown 2 cabs (4v4 drivers), fits in the previously described dimensions, and is shown with 65v, the input it takes to bring the driver to xvar at the low corner with a 28hz hipass.
You can see my cab is clearly tuned higher. but at half the physical size per driver, its no slouch. Its response is much smoother, and while it is less sensitive between 40-50hz, and below 28hz, I think the space savings are worth it, and honestly I think it will "sound better".
some notes on your design however.
Increasing compression ratio to nearly 4:1 (reducing S1 to 215), decreasing s2 to around 565, and increasing rear chamber volume to 100 liters appears to do WONDERS to "smooth response". 4:1 CR is approaching the "no fly zone", but it lowers the low corner group delay peak by almost 10ms.
Just a note, the group delay on your cab at the low corner is insanely high at 60ms. my cab peaks out at 33ms.
My biggest issue with going that route is that the 21" b&c is double the price of the 15nw100. I had played a bit with dual loading cabinets with the 15" driver, but I don't recall coming up with a sim that made it seem worth it. The mouth size that I chose for the combined 4 horns was to be 1/4 wavelength at 20hz, and the halved again for each halving of the steradians it was meant to operate into. Halving the size, as you said, rises the cutoff despite all the other benefits... Believe me, making cabinets this big isn't the most sane thing I've ever done, but I don't know if I'll have another opportunity any time soon
!The group delay is something I really should address... Honestly, its something I hadn't heard of previously, and I forgot to fix it after I understood what it was. I also had thought that it was something fixable through signal processing, but now that seems to not be the case.
Ill definitely see what raising the CR will do though, and how easily adaptable my design is. I do have to start building next week at the latest, so hopefully I'll find an easy fix (if such a thing ever existed)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.