L300 modifying X-over need speaker data

[jdelange]

The other day I decided to measure my L300's with- and without subwoofer using REW and a USB MiniDSP mic. I am measuring at approx 3 m from the speaker at listening position. I notice that the mid and high section generally operate at a reduced volume relative to the low end. It may also be that the (mid) bass is operating at an elevated level due to (floor) reflections. This cannot be altered with the L-pads. I was thinking that perhaps changing some filter components could make a difference. There are various programs allowing modelling of filter and speaker components, but for this I need speaker data. I have obtained the necessary frequency response and Thiele and Small parameters for the 2235H, but I am unable to find this data for the 2425j (16 ohm) and the 2405h. I found a very basic frequency response of both on a JBL spec sheet, but no impedance curve and no Thiele and small parameters required by the software.

I have electronically measured the 2235H with Woofer Tester 2 to tune the bass port, but I am not sure if I can use it to measure the mid and high drivers.

The network I am using is the charge coupled design by Giskard http://www.audioheritage.org/vbulletin/showthread.php?166-4333-S300-L300-equivalent-bandbass-circuit&p=1044&viewfull=1#post1044

I know that a MiniDSP unit would allow me to straighten the curve, which in essence would come down to lowering the (mid) bass, but there is so much reserve in the mid and high drivers that it makes more sense to reduce the dampening / roll off of those.

Anyone went through a similar exercise? Anyone with suitable driver data? Or perhaps an advise which components to change?

Attached some in-room curves I measured. Left, right, with- and without subwoofer. The subwoofer is located next to the right speaker, so it's phase behavior is different compared to the left and right main speaker. Room resonances in the low end are obvious.

[grumpy]

at approx 3 m from the speaker at listening position. I notice that the mid and high section generally operate at a reduced volume relative to the low end.

This is expected/normal from what is essentially a room-response measurement.

http://www.audioheritage.org/vbullet...n-Target-Curve

You might have some work to do (as I believe you mentioned) in the 200-700Hz area if it sounds too forward/vocals sound "chesty",
but you may risk killing some of the fun of such a system. I would suggest not trying to micro-manage the problem
and perhaps use the DSP to temper the range with a fairly wide filter (low Q) and in small amplitude steps.