Designing a notch filter for the TB W4-1320 Bamboo driver...

[fjhuerta]

Hi all!

I just finished my TB bipolar tower. It's a simple bass reflex design, tuned at 65 Hz, 16L.

While bass is surprising (I wasn't expecting how powerful and accurate those little drivers are) I'm having some problems with a "shouty" upper end.

I'm not sure if it's my room or the speakers. So I used JustMLS. I took the measurement about 10 cms. away from the center of the cone, because reflections were awful.

I also took a measurement 1 meter away, smoothing it at 1/3rd octave. The 7 KHz spike was nowhere to be seen.

Considering I built a bipole, I'm not sure I should build a notch filter... maybe the room is as guilty as the slight break-up I see. But, if I decided to build a notch filter, considering I connected both drivers in parallel, should I go for a shunt RLC network, or a series R/L/C one? I'm a bit confused about this...

 

[Corloc]

I would play with room position before trying a notch filter. I have the 1320's in the PAWO cabnet, and they had a bad boom in the bottom end. I tryed several room poitions and the boom went away. I have'nt had any shouting with my drivers.

 

[Mike514ml]

<I have the 1320s in the PAWO cabinet....>

I'm thinking about doing the same thing. Did you use the same dimensions as the cabinet for the FR125S?

I've seen one for the JX-92, which has slightly different dimensions.

Thanks,

Mike.

 

[soongsc]

For a shouting midrange, try putting a 5ohm//1uF parallel cirtuit in series with the driver and see what happens. If the balance is about right, add a 0.01uf ERO or Wima MKP 100V cap in parallel to the 1uF cap.

 

[Corloc]

I used the dimmentions for the PAWO. It was designed fr the CCS FR125. I used the Dyflex panel behind the driver. Then Blackhole on two wall, but the Blackhole was removed latter on.

 

[fjhuerta]

For a shouting midrange, try putting a 5ohm//1uF parallel cirtuit in series with the driver and see what happens. If the balance is about right, add a 0.01uf ERO or Wima MKP 100V cap in parallel to the 1uF cap.

Hi Soong,

I cured the "shouting" midrange by toeing out the speakers (they were very toed in). I suppose listening directly on axis to the phase plug was the culprit. The speakers now are great... and the JustMLS measurement is very flat, almost to 20KHz.

I have very little experience with full range speakers (other than Zaph's Hi-Vi 3"), but I'm quite pleased with what I hear so far!

 

[soongsc]

Hi Soong,

I cured the "shouting" midrange by toeing out the speakers (they were very toed in). I suppose listening directly on axis to the phase plug was the culprit. The speakers now are great... and the JustMLS measurement is very flat, almost to 20KHz.

I have very little experience with full range speakers (other than Zaph's Hi-Vi 3"), but I'm quite pleased with what I hear so far!

That also works to some degree.😉
I would have expected loss in high frequency when you toe them in so that the center axis cross in frount of the listener.

 

[fjhuerta]

Thanks a lot for your help Soong!

One more question (I'm a newbie in all of this). I find my full-range speakers quite boring whenever I'm not seated in the "sweet spot". There's a very, very tight place where they sound balanced and nice. Everywhere else they are a bit dark and recessed. Is this normal?

 

[soongsc]

Thanks a lot for your help Soong!

One more question (I'm a newbie in all of this). I find my full-range speakers quite boring whenever I'm not seated in the "sweet spot". There's a very, very tight place where they sound balanced and nice. Everywhere else they are a bit dark and recessed. Is this normal?

This is normal for the kind of frequency response you show. Most drivers that use more toe in have a frequency response rise starting around 5K and up. So that when you toe them in and listen off-axis, the response will be flat, and then you will have a better sweep spot when the toe in cross point is in front of you. Then you will also have the advantage that the far speaker generates more high frequency energy to balance the increased distance to create the larger sweep spot. However, since there is no way to compensate for the time difference of signals, it is still best while sitting in the middle.