Very well.
If you're comfortable increasing the internal width from 32.4cm to 39cm, and internal depth from 37.8cm to 42cm, height increased to 120cm. you can use a 15" + slightly more aggressive roll off for almost 93db/w at 30hz.
Here's a sim that keeps just within xmax of the Faital 15PR400.
Particle velocity is not a concern at this power.
Edit:
I forgot that we have to adjust the 15" a few cm's down, this should not make much difference for the results, but the xo probably need some modifications.
If you're comfortable increasing the internal width from 32.4cm to 39cm, and internal depth from 37.8cm to 42cm, height increased to 120cm. you can use a 15" + slightly more aggressive roll off for almost 93db/w at 30hz.
Here's a sim that keeps just within xmax of the Faital 15PR400.
Particle velocity is not a concern at this power.
Edit:
I forgot that we have to adjust the 15" a few cm's down, this should not make much difference for the results, but the xo probably need some modifications.
Attachments
Last edited:
I like that better. I downloaded a copy of hornresp and am starting to learn how to use it. As a beginning test I loaded in your parameters, however I can't seem to get the horn geometry like yours. When I can't seem to separate S3 on the two lines. When I change S3 in the "S2 S3" line it auto-reflects down to the "S3 S4" line and vice versa.
I clearly don't know how to drive hornresp just yet.... baby steps. By the way, I'm a "rock-n-roller," so if I spend ~$3000 USD on parts, it's got to have some low end punch! haha.
gabo
Last edited:
Different woofer? Different x-over! Different speaker!! Start from scratch!!! 
I am not going there. If you want an Open Source Monkey Coffin but in a tower format, stick to the tried-and-tested driver combo, front-baffle geometry and x-over design. The larger volume will be useful to extend the bass cut-off frequency without affecting the existing design.
Maybe take a look at Aattos open source Tower XL project? That would be a rather different design, though.
I am not going there. If you want an Open Source Monkey Coffin but in a tower format, stick to the tried-and-tested driver combo, front-baffle geometry and x-over design. The larger volume will be useful to extend the bass cut-off frequency without affecting the existing design.
Maybe take a look at Aattos open source Tower XL project? That would be a rather different design, though.
Different woofer? Different x-over! Different speaker!! Start from scratch!!!
I am not going there. If you want an Open Source Monkey Coffin but in a tower format, stick to the tried-and-tested driver combo, front-baffle geometry and x-over design. The larger volume will be useful to extend the bass cut-off frequency without affecting the existing design.
Maybe take a look at Aattos open source Tower XL project? That would be a rather different design, though.
Hahah, nah, not from scratch. The midrange and tweeter combo are great, the crossover isn't going to require much tweaking. Just taking the opportunity to explore different box/woofer ideas.
At the end of the day, we might be right back at the existing design with a bit larger volume. But it never hurts to explore.
I don't figure you're going to build them anyway, I can't imagine you wanting to spend that kind of money on a variation of something you already have. I'm not afraid of tweaking things, front baffle and crossover included.
It's all in good fun. If I build something I don't like, I'll rip it apart and start over. Once I have the components, the box and construction is cheap. NO fear young man! haha.
gabo
Last edited:
You need to specify stepped segments. See the attached extract from the Help file for details.
Attachments
FIY I did the stepped segment as a sort of compensation for the port volume taking up space in that segment.
Just eyeballing it.
Just eyeballing it.
Some of us are happy using DSP and lots of power ampsDifferent woofer? Different x-over! Different speaker!! Start from scratch!!!
Last edited:
FIY I did the stepped segment as a sort of compensation for the port volume taking up space in that segment.
Just eyeballing it.
Some of us are happy using DSP and lots of power amps
Hahaha, yes. If we wanted to just go with the "tried and true" we'd just buy a kit
Still haven't figured out how to get hornresp in "step mode" but learning how to use it anyway, without step mode. I like what the PR400 woofer does to the response. But now that I am learning how to drive hornresp a bit, I need to also put in the 12" PR320 so I can compare them a bit better with various ports.
gabo
Baffle and XO
To my eye, the 15PR400 in the larger cabinet looks much better. Much flatter down to around 38hz or so but of course the steep rolloff after that. So if we use the 15PR400 and change the width of the speaker....
I guess we would start by just keeping the mid/tweet position as they are and extend each side of the baffle out to match the new width?
I need to go learn Vituixcad to understand what all this would do to the reflections and xo. That might take a while
gabo
To my eye, the 15PR400 in the larger cabinet looks much better. Much flatter down to around 38hz or so but of course the steep rolloff after that. So if we use the 15PR400 and change the width of the speaker....
I guess we would start by just keeping the mid/tweet position as they are and extend each side of the baffle out to match the new width?
I need to go learn Vituixcad to understand what all this would do to the reflections and xo. That might take a while
gabo
To change "modes" in hornresp, click the area between the fields, IE where it says S1, S2, S3 etc.
The depth was also increased in my sim, not only the width.
I think it sounds like a good idea to just put the mid and tweeter in roughly the same area, at least if you're using DSP + calibrated measurement microphone.
If you're sticking with passive you should get into Vituix first.
Either way, a measurement microphone is a very good investment if you don't have one. Ears are easily fooled.
The depth was also increased in my sim, not only the width.
I think it sounds like a good idea to just put the mid and tweeter in roughly the same area, at least if you're using DSP + calibrated measurement microphone.
If you're sticking with passive you should get into Vituix first.
Either way, a measurement microphone is a very good investment if you don't have one. Ears are easily fooled.
Yea I would like to do a passive, I've downloaded vituixcad and will start learning how to do that.
I have a measurement mic, I have lots of mics, actually that's what I do! I'm a drummer and do a bit of session work. Although these days not as much as I once did, but I still do sessions. Just finished a drum part for a singer/songwriter last week.
gabo
I have a measurement mic, I have lots of mics, actually that's what I do! I'm a drummer and do a bit of session work. Although these days not as much as I once did, but I still do sessions. Just finished a drum part for a singer/songwriter last week.
gabo
If you change the width of the front baffle, you can't keep the driver positions as they are (left and right distances to the baffle edges cannot remain the same). Different baffle size means different driver positions, so you'll have to take a very close look at the xover design, if not re-design it from scratch. That would be a lot of work.
If you want to go with a larger woofer, you'll end up with rather large loudspeaker. Take a look at Aattos "Open Source Tower XL".
Yes it does make it quite large for sure, more to think about.
"To change "modes" in hornresp, click the area between the fields, IE where it says S1, S2, S3 etc" Thanks for that KaffiMann, amazing how when you know what to do... it's so simple!!
gabo
Last edited:
good idea. Maybe pick a different mid to save money and really for that matter, why the pricy tweeter with waveguide?
Here's a comparison using VituixCad of the OSMC with:
The same front baffle just increased for height, the same mid/tweet with 4 SB23MFCL45's. A bit of a tweak to the woofer crossover (you can see the changes in the schema).
This is using the original OSMC measured data, along with modeled data for the new woofers.
Mordikai - I already have the mid and tweet drivers, and the concept was to try to lean on the OSMC as much as possible to avoid re-work. So I was trying to keep the same baffle and mid/tweet crossover. But I've also played around with changing the baffle/width, so for others it's DIYaudio, so redesign as much as you like. I'm game to follow along and help out as much as I can. It's good way to learn!
The original OSMC on the left, the new woofers on the right.
gabo
The same front baffle just increased for height, the same mid/tweet with 4 SB23MFCL45's. A bit of a tweak to the woofer crossover (you can see the changes in the schema).
This is using the original OSMC measured data, along with modeled data for the new woofers.
Mordikai - I already have the mid and tweet drivers, and the concept was to try to lean on the OSMC as much as possible to avoid re-work. So I was trying to keep the same baffle and mid/tweet crossover. But I've also played around with changing the baffle/width, so for others it's DIYaudio, so redesign as much as you like. I'm game to follow along and help out as much as I can. It's good way to learn!
The original OSMC on the left, the new woofers on the right.
gabo
Attachments
It looks like you tried to make the on-axis response as flat as possible. This will give very bright sound with the waveguides. Just use Vituix to calculate the power response and make sure it is a smooth line with a linear decrease. Read the osmcdoc and the corresponding posts in the OSMC thread.
I would STRONGLY recommend you to leave the xover as it was with the OSMC for the midrange and tweeter. Then find a woofer with an efficiency of about 97 dB/2.83V/1m into a 2pi geometry (or higher). This will not be easy. Then determine a suitable bass tuning for that woofer, and finally design a suitable xover for this woofer, taking into account the baffle step.
Honestly your best bet would be to stay with the 12PR320 woofer. The whole OSMC design started by choosing the right woofer, and then we designed everything else on top of that. Going with a different woofer seems like going back to the start and redesigning everything again from scratch.
I would STRONGLY recommend you to leave the xover as it was with the OSMC for the midrange and tweeter. Then find a woofer with an efficiency of about 97 dB/2.83V/1m into a 2pi geometry (or higher). This will not be easy. Then determine a suitable bass tuning for that woofer, and finally design a suitable xover for this woofer, taking into account the baffle step.
Honestly your best bet would be to stay with the 12PR320 woofer. The whole OSMC design started by choosing the right woofer, and then we designed everything else on top of that. Going with a different woofer seems like going back to the start and redesigning everything again from scratch.
Ok, thanks. I'll probably just build the OSMC and forget the tower. Not much interest here anyway. Sorry for the inconvenience.
As you've pointed out, if anyone is interested in a tower, they should probably go over to Aatto's Open Source Tower XL.
thanks, gabo