Has anybody made a cardiod midrange?

[ghemink]

Hello all,

I know many people are experimenting with dipole woofers and midranges (also myself), however, is there anybody that has tried to make a cardiod woofer/midrange, say from 100 to 1500Hz or so. I know that people are doing U-frames for woofers to get something that is close to a cardiod response, however, for a woofer/midrange that has to work upto 1500Hz or so that would be far more complicated. There are commercial systems (Gradient) doing this, however, I`m looking of course for a DIY solution.

Best regards

Gertjan

[mhelin]

Another commercial cardioid speaker(s) also come from Finland, amphion's xenon and crypton. This is a very diyable solution, drill holes in triangle form (~ 3/4 of the area of the speaker cone, holes shouldn't be too small) and use some textile or fiber material to work as resistance for air (remember T=RC, R is the factor related to material used in opening, C is the filling or stuffing of the box, maybe don't be sure), usual cotton like used on bed sheets is fine as textile material. Experiment a lot and make measurements. The DIY2004 Finland winner used this kind of midrange and bass resistance boxes (
see this one ,
).image

[DSP_Geek]

Originally posted by ghemink
Hello all,

I know many people are experimenting with dipole woofers and midranges (also myself), however, is there anybody that has tried to make a cardiod woofer/midrange, say from 100 to 1500Hz or so. I know that people are doing U-frames for woofers to get something that is close to a cardiod response, however, for a woofer/midrange that has to work upto 1500Hz or so that would be far more complicated. There are commercial systems (Gradient) doing this, however, I`m looking of course for a DIY solution.

Best regards

Gertjan

Well, you really only need to worry about controlling directivity until the baffle step region, so that's about a few hundred Hz. Above that there's progressively less sound going to the rear, modulo controlled diffraction off the baffle edges of course.

As previously mentioned you could set up an acoustic low-pass filter for the backwave so you don't get a transition from cardioid to dipole above BS frequencies. The LPF will add a bit of delay, so you want to account for that when calculating your baffle sizes; it turns out that if you get everything just right you even get a couple of dB gain for an octave or so, since the backwave is delayed just enough to come back into phase with the frontwave.

The problem with running open baffle is that most drivers have a fairly low Qts, so you need to boost them up to a couple of octaves away from their resonant frequency to make sure the response is reasonably flat - think of it as a Linkwitz Transform, but with a higher Qts instead of a lower cutoff frequency.

Yes, I have been thinking about cardioids, why do you ask?


Cheers,
Francois.

[ghemink]

Originally posted by DSP_Geek


Well, you really only need to worry about controlling directivity until the baffle step region, so that's about a few hundred Hz. Above that there's progressively less sound going to the rear, modulo controlled diffraction off the baffle edges of course.

As previously mentioned you could set up an acoustic low-pass filter for the backwave so you don't get a transition from cardioid to dipole above BS frequencies. The LPF will add a bit of delay, so you want to account for that when calculating your baffle sizes; it turns out that if you get everything just right you even get a couple of dB gain for an octave or so, since the backwave is delayed just enough to come back into phase with the frontwave.

The problem with running open baffle is that most drivers have a fairly low Qts, so you need to boost them up to a couple of octaves away from their resonant frequency to make sure the response is reasonably flat - think of it as a Linkwitz Transform, but with a higher Qts instead of a lower cutoff frequency.

Yes, I have been thinking about cardioids, why do you ask?


Cheers,
Francois.

Thanks Francois and Mhelin for the tips and links. I`ll try to gather a bit more info (and finish my current project first) before starting to experiment with this. I use a DSP crossover and speaker correction system (DEQX PDC) so I don`t worry too much about low Qts of the drivers.

Best regards

Gertjan