2-Ways speakers for 22W per channel amplifier

Hello!

I wanted to ask an opinion. I'm doing a 22W per channel amplifier with stereo controls, which will be connected to a 2-way speakers per channel.

In the next picture is the circuit that connects the amplifier to the speakers.

[IMGDEAD]http://img225.imageshack.us/img225/8321/dibujo3ab.th.jpg[/IMGDEAD]

Uploaded with ImageShack.us

In the following, are the speakers I can get.

[IMGDEAD]http://img33.imageshack.us/img33/4194/agudos.th.jpg[/IMGDEAD]

Uploaded with ImageShack.us

[IMGDEAD]http://img593.imageshack.us/img593/5945/medios.th.jpg[/IMGDEAD]

Uploaded with ImageShack.us

I want to make a filter of 1st order with 6000Hz court frecuency , it gives me some results of C = 4.42uF and L = 0.21mH

What do you think? Would it work well? Should I add anything else? For example, Zobel network?

Thank you for your attention!
 
When you are in possess of the woofers ,put them on running for at least 5 hours ,preferably in the box . Then you can give it a listen ,also unfiltered ,and you can judge its sonic behavior - specially in the mid-trebles ,and also in the lows ,if the box chosen is the final .
The frequency response in the curve given in the brief datasheet does not tell the sonics of that woofer : it may tell you that it might reach 3 or 4 KHz ...but it is better to cut at least an octave below . Also the speaker-in-a-box response will change dramatically and some changes will lead to change component values in the crossover (if not well calculated) .
Zobel networks ,that are used after the amplifier to prevent oscillations/dependence from cables ,can be used here to make the amplifier see a constant load trough the crossover , i.e. 'corrects' the impedance rise at a certain frequency ,the R determining the Q of the intervention ,associated also with the presumably high (for a coil at low frequencies )resistance of the inductor .