Asymmetrical crossover question

[gainphile]

I'm new to asymmetrical XO. Usually I just use textbook XO.

Below is a picture of P13WH woofer measured nearfield (ignore the SPL). It follows either bessel or LR 2nd order highpass filter, almost perfectly.

The question is: Is it reasonable to simply let the driver rolloff by itself (no xo) and for the woofer I apply 2nd order lowpass LR (red curve on left).

What about the phase issue?

[CLS]

Somewhat confused here.

My understanding would be: you'll use this P13WH "woofer" as a mid, and add "another woofer" by <2nd order lowpass LR (red curve on left) >.

If so, I'd like to have at least the simplest HP filter to help reducing the cone excursion of the mid driver and forget about the symmetry of xo slopes.

[gainphile]

Hi CLS,

Correct. I'm actually happy with the excursion issue as I don't play music that loud.

I guess I just have to build and measure the response and see if it's summed properly.

[ttan98]

Try this Thread,

http://techtalk.parts-express.com/sh...20&postcount=1

you may find it useful.

Cheers.

[ttan98]

Try this Thread,

http://techtalk.parts-express.com/sh...20&postcount=1

you may find it useful.

Cheers.

[rabbitz]

For that to work the best IMO is to use a 1st order on the woofer with an oversize zobel to help with the roll off (shape) a couple of octaves past the crossover point. It's sort of how 0.5 woofers work but in this case with different sized drivers.

I've used this method several times with great effect and helps with overall BSC. Room gain on the woofer tends to fill in the gap below the P13 roll off say below 80Hz. The downside is the impedance will drop due to the 2 drivers in parallel. I've had no excursion problems doing it this way and can be assisted by the enclosure design restricting driver excursion.

Using a 2nd order on the woofer would give worse results IMO and would not have as smooth a transition and would have worse phase and summing.

Using 2nd order on both may require reversed polarity on the P13 to ensure correct phase and summing but technically would be the best for a true 3-way design.

Depends on the behaviour of the woofer (mid capability) and how quick you want to get it to dive. It's nice not to have a large (say 120-250uF) cap in front of the P13 by allowing a natural acoustic roll off.

[gainphile]

Thank you all for the advise. That link is very interesting read indeed.

The P13 is on open baffle and textbook crossover did not materialise nicely due to driver rolloff. It resulted in somewhat thin sound.

I implemented the xo tonight and it actually worked!. P13 is able to cope with quite loud SPL, and even with shelving filter down to 60Hz for dipole compensation. It does work hard

The phase did create issue, quiet tiring to listen. After few read I calculated that the induced -90 degrees phase shift on the woofer is worth about 60cm. I did not have the space and moving back my h-frame woofers 45cm seems to be good enough. I know it's not coherent due to no XO on the mid, but we'll see whether this stands longer listening test.

[rabbitz]

OB

Yes that changes the game. Have to allow for additional 1st order roll off under Fe and a rise of around 6dB to Fp.

I remember having to play around with polarity of the mid and woofer in an OB I did. IIRC the woofer ended up reversed with the mid normal.

[tinitus]

Surpricing that your small midwoofer doesnt seem to have the dipole related rolloff