Hi all,
It has been a long time (2-3 years) that I discussed my first steps to design an open baffle loudspeaker utilising a Mundorf AMT and an acoustic elegance Dipole woofer.
(old thread: https://www.diyaudio.com/community/...ng-bass-driver-and-u-frame-side-wings.384160/)
I made some progress on this project in the meantime but also could not really work on it due to several reasons (private and professional).
Now that things have settled I am back to it!
A quick summary on the current stage of the project:
Drivers are set:
I came to the realization that the crossover topic would delay this project significantly as I have never done this before and I did not want to go through the unavoidable simulation + trial and error for optimizations rounds.
So I decided to make them active with Hypex Fusion FA123 plate amps (with included actrive crossover). IMHO an ideal combination!
Next step: To optimize driver placement and test a revised prototype (I have measurement equipment & use REW).
Goal: to make it as small as possible without sacrificing design & performance.
The version 2 looks like this (without the nice frontplate and without the waveguide for the AMT):
The design is now made with FreeCad and fully parmaterized so that I can easily change settings and driver placements.
I already "converted" one of my prototpyes to these dimensions, the next one will be slightly taller and narrower.
I will post some measurements in the next days and hope you can give me some advice on how to optimize it.
It has been a long time (2-3 years) that I discussed my first steps to design an open baffle loudspeaker utilising a Mundorf AMT and an acoustic elegance Dipole woofer.
(old thread: https://www.diyaudio.com/community/...ng-bass-driver-and-u-frame-side-wings.384160/)
I made some progress on this project in the meantime but also could not really work on it due to several reasons (private and professional).
Now that things have settled I am back to it!
A quick summary on the current stage of the project:
Drivers are set:
- Mundorf AMT23D6.1-R
- Audio Technology C-Quenze 18H52
- Acoustic Elegance Dipole12
I came to the realization that the crossover topic would delay this project significantly as I have never done this before and I did not want to go through the unavoidable simulation + trial and error for optimizations rounds.
So I decided to make them active with Hypex Fusion FA123 plate amps (with included actrive crossover). IMHO an ideal combination!
Next step: To optimize driver placement and test a revised prototype (I have measurement equipment & use REW).
Goal: to make it as small as possible without sacrificing design & performance.
The version 2 looks like this (without the nice frontplate and without the waveguide for the AMT):
The design is now made with FreeCad and fully parmaterized so that I can easily change settings and driver placements.
I already "converted" one of my prototpyes to these dimensions, the next one will be slightly taller and narrower.
I will post some measurements in the next days and hope you can give me some advice on how to optimize it.
Last edited:
I would put the woofer as close to the floor as possible and put the midrange higher. Is that a calculated waveguide for the Mundorf AMT? Of not I would not take the risk of using one, it probably makes it much worse. I would put the AMT naked on top of the baffle. For proper dipole behaviour you want it on a baffle that is as small as possible or even non existent. The mundorf tweeters sound pretty good naked. This way you will get just as good constant directivity as with a proper waveguide but in a smaller package. Be carefull with the use of wings, to deep destroys the dipole behaviour and has a tendency to sound like cupped hands. Personally I only would use baffle likes this down to 60-70Hz or so with a good subwoofer underneath. For proper dipole lowbass look at a H-frame or W-frame with dual 12 or 15 inch drivers. A flat baffle with one 12 inch driver will make no impact at all in the lowest octave, it can be very good in midbass though.
Thanks for the suggestions. Here some comments:
Woofer placement:
I had it further down but after my decision to use the Hypex plate amp I needed some space for the amp below the woofer. This is more or less the lowest without interfering with the plate amp.
Waveguide:
Nothing calculated. Is ther a program to do so? I let Freecad create an approximated subtractive loft and just defined 3 planes. I tried to roughly mimick this design: https://www.selmoni-speakers.ch/produkte/lumina/index.html
This is the current cross section:
I will use a CNC router to make these....
Wings:
Here I was inspired by the NaO speakers: http://musicanddesign.speakerdesign.net/products.html
I will definitely play around with dimensions with my prototype and then decide.
Woofer:
My hope is that with the Hypex amp I can selectively increase the gain by a few dB for the woofer to compensate for the smaller area. I think the AE drivers are robust enough to handle this I am not a loud listener anyway.
Midrange:
That was my plan for V3 anway to move this one up as close as possible to the AMT so that it still looks good.
Woofer placement:
I had it further down but after my decision to use the Hypex plate amp I needed some space for the amp below the woofer. This is more or less the lowest without interfering with the plate amp.
Waveguide:
Nothing calculated. Is ther a program to do so? I let Freecad create an approximated subtractive loft and just defined 3 planes. I tried to roughly mimick this design: https://www.selmoni-speakers.ch/produkte/lumina/index.html
This is the current cross section:
I will use a CNC router to make these....
Wings:
Here I was inspired by the NaO speakers: http://musicanddesign.speakerdesign.net/products.html
I will definitely play around with dimensions with my prototype and then decide.
Woofer:
My hope is that with the Hypex amp I can selectively increase the gain by a few dB for the woofer to compensate for the smaller area. I think the AE drivers are robust enough to handle this I am not a loud listener anyway.
Midrange:
That was my plan for V3 anway to move this one up as close as possible to the AMT so that it still looks good.
I would strongly advice against a waveguide that is purely designed on astatics , chances are very high that you only make it sound worse with resonances that are hard to tame and a polar response that is all over the place. Ditch the faceplate and run them naked on top of the baffle, there a big
chance it sounds much better that way. At least it will have dipole behaviour in the crossover region to your mids.
As for the low end, you probably need around 15 to 20 db compensation, that's more realistic for open baffle bass. A single 12 inch is very minimal, specially on a flat baffle. The wings make it a monopole in the crossover region, a cardioid in the lower mids and dipolke in low end. It is not the best solution for dipole bass. Dual 12 inchens in a H-frame is more likely to work. For dipole use you simply need a lot of moving areas, small moving areas with high x-max only sounds like an airpump. Other possibiltiy is to make the baffle flat, or with very small wings and us it from 60-70Hz with a closed sub in the low end. I have done that a couple of times and it works pretty good, even with a small 8 inch closed sub.
The selmony loudspeaker is not a dipole, that a whole other story.
What you want to achieve is more or less the same polar response of your midrange and your tweeter at least in the crossover region. Unless you don't want to run your tweeter with an open back, then a waveguide probably works betterr, but with open back, ditch the waveguide.
chance it sounds much better that way. At least it will have dipole behaviour in the crossover region to your mids.
As for the low end, you probably need around 15 to 20 db compensation, that's more realistic for open baffle bass. A single 12 inch is very minimal, specially on a flat baffle. The wings make it a monopole in the crossover region, a cardioid in the lower mids and dipolke in low end. It is not the best solution for dipole bass. Dual 12 inchens in a H-frame is more likely to work. For dipole use you simply need a lot of moving areas, small moving areas with high x-max only sounds like an airpump. Other possibiltiy is to make the baffle flat, or with very small wings and us it from 60-70Hz with a closed sub in the low end. I have done that a couple of times and it works pretty good, even with a small 8 inch closed sub.
The selmony loudspeaker is not a dipole, that a whole other story.
What you want to achieve is more or less the same polar response of your midrange and your tweeter at least in the crossover region. Unless you don't want to run your tweeter with an open back, then a waveguide probably works betterr, but with open back, ditch the waveguide.
Last edited:
Thanks for the suggestions.
I am working on some modifications as removing the waveguide. I will move the AMT as close as possible to the front but I will keep the front panel as it is an important part of my concept (I have seen simialr things with e.g Spectralaudio open baffel speakers).
Adding a sub is not an option for me, I will work as best as possible with the drivers I have now. I definitely don't want to cut the Dipole12 low as they were rather expensive and I bought them specifically for their good performance in OB speakers (e.g. high xmax and low fs).
Updating my FreeCAD model takes a bit longer as I am now using the weekly builds which required me to change to a different assembly workbench...
I am working on some modifications as removing the waveguide. I will move the AMT as close as possible to the front but I will keep the front panel as it is an important part of my concept (I have seen simialr things with e.g Spectralaudio open baffel speakers).
Adding a sub is not an option for me, I will work as best as possible with the drivers I have now. I definitely don't want to cut the Dipole12 low as they were rather expensive and I bought them specifically for their good performance in OB speakers (e.g. high xmax and low fs).
Updating my FreeCAD model takes a bit longer as I am now using the weekly builds which required me to change to a different assembly workbench...