Tweet
Re: So - why DON'T people go with open baffle?
Just use Bagby's software and it's seriously as easy as monopole design. I'm serious. The ONLY difference is that when you are preparing your files in the Diffraction Simulator spreadsheet you need to specify that you are simulating a dipole design, not monopole. If you want to you can add in the effects of the walls and floor if you like, I choose to incorporate the floor bounce in all my designs (mono or dipole) but no other boundaries. Aside from that there's no special dipole room considerations, there's no room gain. So with the exception of toggling to "diople" instead of "monopole" in the Diffraction Simulator, there's NO DIFFERNCE to monopole or dipole design. Once you toggle to "dipole" you will see that the rolloff below the step frequency doesn't stop at 6db and the diffraction is about 2x worse, but all this is easily handled in the xo design.
Once you have your Diffraction files saved, import them into Response Modeller and proceed exactly the same as you would if you were doing a monopole design. Namely, sum the diffraction .frd with the driver .frd.
Once you have the .frd and .zma files finished in Response Modeller you enter them into PCD and design the xo in exactly the same fashion you would for a monopole design.
All that is the long way of saying that the only difference is modeling is that you need to toggle to "dipole" instead of "monopole" while creating your .frd files. That's really all there is to it.
To save a lot of frustration and trial and error it's best to understand which drivers are likely to work well BEFORE you go to all that trouble. Starting with the bass driver, you need to choose a driver and baffle width that complement each other to achieve smooth response down to your desired LF cutoff, no matter how low you want it to go. This is extremely easy to do and doesn't require a computer, you can do it in your head. You take the driver's IB response and sum it to the rolloff of a finite sized baffle. There's a page on my website that tells you how to do this, it's very easy. It's called "simple ob design" and it's in the theory section of my website. I've already linked to it at least twice this week so I won't link it again, google doesn't seem to like it when I constantly link to my own site.
Once you have chosen a suitable bass driver and baffle width then you can choose a mid and a xo point between them. The first big dipole diffraction hump is an excellent place to xo, and this is the same practice that we regularly use in our monopole design process as well, so nothing new with that concept. The mid driver does not have to have any special qualities whatsoever, it doesn't need more displacement than it would need if the design was monopole. In other words, just about any mid will work just fine as long as you use it above the first big diffraction peak. The difference is that the diffraction bump will be 2x bigger than if it were a monopole design but that's no problem to correct in the xo, assuming you are crossing over right at the diffraction peak, as you normally would anyway.
The last consideration is the tweeter. You can choose to go full dipole all the way up or you can use a regular tweeter. That's simply a matter of preference.
As an example, check the measurement I posted. The driver and baffle width have been specifically chosen to complement each other to provide response down to fs with no contouring filters of any kind. If that driver was used as a woofer only in a better design, it ideally would be crossed over to a mid at 700 hz and the xo would take care of leveling the diffraction bump in the woofer and mid response, same as you would do if you were doing a monopole design.
I really hope this all helps, I realize I've repeated myself several times now but I'm just trying to be as clear as possible.
Of course if you use active eq you can just disregard this completely and just force the drivers to do what you want.
Originally posted by fastbike1
View Post
Once you have your Diffraction files saved, import them into Response Modeller and proceed exactly the same as you would if you were doing a monopole design. Namely, sum the diffraction .frd with the driver .frd.
Once you have the .frd and .zma files finished in Response Modeller you enter them into PCD and design the xo in exactly the same fashion you would for a monopole design.
All that is the long way of saying that the only difference is modeling is that you need to toggle to "dipole" instead of "monopole" while creating your .frd files. That's really all there is to it.
To save a lot of frustration and trial and error it's best to understand which drivers are likely to work well BEFORE you go to all that trouble. Starting with the bass driver, you need to choose a driver and baffle width that complement each other to achieve smooth response down to your desired LF cutoff, no matter how low you want it to go. This is extremely easy to do and doesn't require a computer, you can do it in your head. You take the driver's IB response and sum it to the rolloff of a finite sized baffle. There's a page on my website that tells you how to do this, it's very easy. It's called "simple ob design" and it's in the theory section of my website. I've already linked to it at least twice this week so I won't link it again, google doesn't seem to like it when I constantly link to my own site.
Once you have chosen a suitable bass driver and baffle width then you can choose a mid and a xo point between them. The first big dipole diffraction hump is an excellent place to xo, and this is the same practice that we regularly use in our monopole design process as well, so nothing new with that concept. The mid driver does not have to have any special qualities whatsoever, it doesn't need more displacement than it would need if the design was monopole. In other words, just about any mid will work just fine as long as you use it above the first big diffraction peak. The difference is that the diffraction bump will be 2x bigger than if it were a monopole design but that's no problem to correct in the xo, assuming you are crossing over right at the diffraction peak, as you normally would anyway.
The last consideration is the tweeter. You can choose to go full dipole all the way up or you can use a regular tweeter. That's simply a matter of preference.
As an example, check the measurement I posted. The driver and baffle width have been specifically chosen to complement each other to provide response down to fs with no contouring filters of any kind. If that driver was used as a woofer only in a better design, it ideally would be crossed over to a mid at 700 hz and the xo would take care of leveling the diffraction bump in the woofer and mid response, same as you would do if you were doing a monopole design.
I really hope this all helps, I realize I've repeated myself several times now but I'm just trying to be as clear as possible.
Of course if you use active eq you can just disregard this completely and just force the drivers to do what you want.
Comment