Project idea for Visaton Open Baffle Dipole 3 way.

This seems like a bug to me. Try to contact UweG regarding that issue.

Did you double-check using a driver with a Qts less than 1?

Also check the settings used here: http://www.boxsim.de/download/Petite%20Orgue%20Bassteil.BPJ

Nice project by the way.

I have been trying different baffle layouts where I can place the drivers all in the centre so their radiation patterns are symmetrical, and this is what I have come up with.



What do you think of this layout?



About the tweeter, do you think I am crossing the G20SC too low, or does it look fine?



For the crossover, I have used real values for everything of components I can source locally with correct OHM resistance for each. I used a 16g laminated steel inductor on the woofer, and 18g air core for the rest.

How accurate is Boxsim on open baffle designs like this?

Comments?

(P.S. I have cleaned up my crossover diagram, I didn't think of using three amps for a cleaner layout... )

Thank you,

David in Canada.

Three amps is a good option when it comes to simulating active filtering.

The response is astonishingly flat down to 40 Hz, but I'm not sure with the impedance.

I am also not familiar with the tweeter. The crossover is depending on the performance (possibly demanding a higher or steeper crossover) and the radiation and diffraction (often demanding a lower crossover). Check also the sono.

The woofers might be placed to low. It is better to get closer to the midrange. Boxsim doesn't show this as it assumes infinite distance, not regular listening or measuring distance. An open baffle should be placed at least 1 meter away from any adjacent wall, so the listening distance is often shorter than regular, hence the aspect of proper addition of the ranges in the nearfield becomes more important.

Room excitation (floor bouncing etc.) should still be okay with slightly higher woofers. See Linkwitzlab and dipolplus.de for details. The in-room simulation is not covered by Boxsim. So it's accurate, but only in its limits. You should have an idea of how this could sound in your room before starting Boxsim and then consider room gain, room modes etc. Simply said it wouldn't work if you attach the dipole speaker to a wall although the Boxsim simulation is correct and cannot be blamed.

Hello there,

I have made a few changes to the design. I changed to the G25FFL in a wave guide for a tweeter and changed the baffle to a 2.5 cm flat panel with a smaller narrow baffle above it for the mid and the tweeter.

Here are the measurements,

If the real speaker is anything like the simulation, this looks very very good.

All crossover values ​​are actual values ​​that can be purchased at a store with correct resistance.

The directivity is incredible
(Die Direktivität ist unglaublich)

Do you think the real speaker will be anything like this When built?

Hi there, as I said before, Boxsim is accurate but it might not work when a dipole is not properly placed.

Also, the imoedance is not homogenous and the midrange doesn't reach the top. I would improve on this.