Padded cone for time alignment

[MarkMcK]

Just a thought. . . Yes, the acoustic center of a driver varies with frequency. There is little you can do about this. Yet, at the crossover frequency you can match the acoustic centers of the drivers. This way you will have a smooth transient transition. The only way to determine the acoustic center at crossover frequency is to measure it. The acoustic center of the drivers must be determined as part of the crossover design. If, at a later time, you change the crossover frequency or slopes, the acoustic centers will change.

If you are using impulse testing (assuming a postive going impulse input to loudspeaker under testing, polarity conservation, and the crossover network already close to summing), a good rule of thumb for determing acoustic center at crossover is to match first negative going spike of the tweeter with the peak of the first positive going spike of the woofer. The improvement in sound quality of transiently aligned and magnitude summing crossover is well worth the effort. It provides a quality of sound and spatial imaging you cannot find in non-aligned loudspeakers.

For a more detail explanation of the impulse response testing (what are spikes, how filters affect the shapes), please see my thread on a Simple MFB Woofer Project.

Good luck,

Mark

[nuppe]

Thanks for the reply Mark!

I'm going to have to think this through again.
I'll definitely try to measure the acoustic centers now that i've got a clue how to. I haven't read up on your thread yet, a little too tired to take anything in right now, but i will tomorrow.

As long as the offset doesn't measure too big i'll try and make the cone the right depth. If it gets to deep i'm going to run into trouble, an 11,5 degree cone gets big pretty fast...

I wonder what would happen if i oversize the cone and cut it along the baffle edges? Maybe that could work...

As these are protoypes for a pentangle speaker, i really want to keep baffle width down, they're going to be huge even as they are. O well, i'll have to think of something.

regards

Andreas

[rabbitz]

Just a note from the LDCookbook that you can use the centre of the voice coil as the radiating centre for calculations. Voice coil centre is the midpoint of the driver motor frontplate.

Just a thought.