I'm in a process of building a 1,5 way center channel, utilising drivers that I have at hand. These are two Tang Band W4-1320SI fullrange drivers mounted vertically. One will run fullrange, the other I plan to cross with 1st order lowpass at around 250hz, to avoid awful combo filtering at higher frequencies.
Since these will overlap over a good part of mid band, I thought that I could use driver spacing to achieve some directionality but by accident I've made the center to center spacing larger that initially intended, at some 23 cm instead of some 17 cm. Now, a distance of 17 cm corresponds to 1/4 of wavelenght of ~450 hz, will the increased c-to-c distance give more directivity control at even lower frequencies or will it just make the sweet spot even narrower and increase combo filtering?
My goal is to achieve lowered off Axis output but I got confused if I should space the drivers as far or as close as possibile. So given that I've already routed the baffle cut outs I dont know if I should mąkę a new baffle or its ok the way it is.
Input much appreciated.
Since these will overlap over a good part of mid band, I thought that I could use driver spacing to achieve some directionality but by accident I've made the center to center spacing larger that initially intended, at some 23 cm instead of some 17 cm. Now, a distance of 17 cm corresponds to 1/4 of wavelenght of ~450 hz, will the increased c-to-c distance give more directivity control at even lower frequencies or will it just make the sweet spot even narrower and increase combo filtering?
My goal is to achieve lowered off Axis output but I got confused if I should space the drivers as far or as close as possibile. So given that I've already routed the baffle cut outs I dont know if I should mąkę a new baffle or its ok the way it is.
Input much appreciated.
Because of the low transition frequency this might not be a problem. Were you planning to measure and confirm?
I think the full range method will probably give the more important mid/high directivity.
I think the full range method will probably give the more important mid/high directivity.
Hi,
one can test this kind of stuff in VituixCAD pretty fast. Open new project, add another driver, hooke them up. Go to Diffraction simulator and input your baffle and driver, move mic to on axis. Check "feed driver" and click "Export" to have a simulated measurements attached to the driver. Repeat for the other one. Few minutes max.
Anyway, 170cm and 230cm c-c don't have much difference. 250Hz 1st order is fine, seems to compensate for bafflestep about right (I don't know your baffle dimensions, guessed).
The vertical directivity is a mess though and DI doesn't increase even a DB unless you remove the lowpass ( or put it above 1kHz) but then the bafflestep is not compensated.
This kind of setup seems to be good only to compensate for bafflestep, at a first glance. You could achieve same result with only single driver and bafflestep compensation circuit and have no lobing issues. Loses few db:s on the sensitivity though. So, if you need the few db sensitivity or have the parts at shelf, this is good design.
First two attachments are 170mm and 230mm c-c, last two demonstrate that DI doesn't change much. Bafflestep is compensated though.
one can test this kind of stuff in VituixCAD pretty fast. Open new project, add another driver, hooke them up. Go to Diffraction simulator and input your baffle and driver, move mic to on axis. Check "feed driver" and click "Export" to have a simulated measurements attached to the driver. Repeat for the other one. Few minutes max.
Anyway, 170cm and 230cm c-c don't have much difference. 250Hz 1st order is fine, seems to compensate for bafflestep about right (I don't know your baffle dimensions, guessed).
The vertical directivity is a mess though and DI doesn't increase even a DB unless you remove the lowpass ( or put it above 1kHz) but then the bafflestep is not compensated.
This kind of setup seems to be good only to compensate for bafflestep, at a first glance. You could achieve same result with only single driver and bafflestep compensation circuit and have no lobing issues. Loses few db:s on the sensitivity though. So, if you need the few db sensitivity or have the parts at shelf, this is good design.
First two attachments are 170mm and 230mm c-c, last two demonstrate that DI doesn't change much. Bafflestep is compensated though.
Attachments
Thank You very much for the simulation, may I Ask what are the increments on the off Axis simulation (lower left graph)?
I do have the drivers at hand and adding sensitivity is a good thing sińce these are only 4" and kinda struggle with clean bass. Cutting them aby higher would make a proper mess with HF comb filtering.
So off Axis response will be poor no matters the spacing, and it doesnt change the directivity unless I play both drivers fullrange. The upside is that with increased spacing I can fit a high directivity tweeter in between and play both midwoofers up to tweeter xo, which I guess would improve overall directivity as an MTM design.
Thank You do much for solving my dillema!
EDIT: Once finished I'll try to measure the response, just pit of curiosity.
I do have the drivers at hand and adding sensitivity is a good thing sińce these are only 4" and kinda struggle with clean bass. Cutting them aby higher would make a proper mess with HF comb filtering.
So off Axis response will be poor no matters the spacing, and it doesnt change the directivity unless I play both drivers fullrange. The upside is that with increased spacing I can fit a high directivity tweeter in between and play both midwoofers up to tweeter xo, which I guess would improve overall directivity as an MTM design.
Thank You do much for solving my dillema!
EDIT: Once finished I'll try to measure the response, just pit of curiosity.
It is 10 degrees per line. Attached is the VituixCAD project file and the diffraction simulator generated measurement files so you can download VituixCAD and test this (and million other things in the future) yourself.
If you sweep the low pass frequency the off-axis response change but has similar trend, some frequencies are boosted and some attenuated. This relates to the c-c and driver size and the low pass slope. You could change the low pass filter order and what not but it won't never be any better only different. By better I mean it can only compensate the bafflestep and give additional sensitivity compared to single driver but at the cost of some lobing compared to single driver. Your current plan seems to be fine and adding a tweeter for MTM config later on sounds like fun time 🙂
If you sweep the low pass frequency the off-axis response change but has similar trend, some frequencies are boosted and some attenuated. This relates to the c-c and driver size and the low pass slope. You could change the low pass filter order and what not but it won't never be any better only different. By better I mean it can only compensate the bafflestep and give additional sensitivity compared to single driver but at the cost of some lobing compared to single driver. Your current plan seems to be fine and adding a tweeter for MTM config later on sounds like fun time 🙂