Removing a notch on point source drivers

[AllisterMcRae]

Just looking at designs around the Morel Integra 624 and seeing a notch in the tweeter that brings the SPL down more than the woofer can bring up at that point. On modelling it in VituixCAD and pushing the XOver frequency up to 8K (seems high!), the notch is still down by about 10dB. (And who wants a sensitivity of 82!?) It is interesting to compare it to the Tang Band W4-1879, which only sports a single graph for the entire driver.

Though I haven't modelled it, check out the notch in the Integra 524 at 7K.

Its easy enough to bring the upper peak down. But why would they create a point source driver where there is such a dip in a noticably audible location? I haven't figured out how to create sound without adding another driver. It just doesn't make sense -does it?



Photo order: Morel Integra 624, Tan Band W4-1879, Combined Integra graph, Morel Integra 524

[Billet]

Every speaker design has some compromises. The tweeter dip is probably a diffraction effect from the coaxial design and might be greatly outweighed by other positive attributes. It probably sounds just fine and likely much better than just fine.

In my opinion, it is impossible to know what a speaker actually sounds like by looking at graphs. Unfortunately, a dip like this might scare some people away from buying what is likely to be a very nice sounding driver.

[AllisterMcRae]

Every speaker design has some compromises. The tweeter dip is probably a diffraction effect from the coaxial design and might be greatly outweighed by other positive attributes. It probably sounds just fine and likely much better than just fine.

In my opinion, it is impossible to know what a speaker actually sounds like by looking at graphs. Unfortunately, a dip like this might scare some people away from buying what is likely to be a very nice sounding driver.

I do love my Morel Drivers. Though I did notice that between the 638ND and the UW1258, they both had a dip at around 300 (if I recall correctly) that I'm trying to combat. Come to think of it, I probably shouldn't have put my x-over point right there...Hmm.

Thanks for you input!

[Wolf]

The 300 dip if measured in room is likely just a floor bounce cancellation. Ignore it.
The 7k dip in the coax is due to guess what- being a coax! Measure it off axis and the dip will likely be much reduced or go away.

Wolf

[dynamo]

Yup, my guess is the dip is the relationship of tweeter diffraction on the woofer surround. Dips like that are less of a concern than they look. If it was a peak that big then you’d have something to concern yourself with. The dip should get better off axis