Active filter 2 ways, your opinions?

[franck1234]

Hello !!

What do you thing of this circuit?



Cut-off frequency of 150hz between the low register box and the satelite.

It's good or not??

Thanks

[Speedskater]

Disconnect pot "RV2"
Connect the two 8K resistors to the negative input of an op-amp.
Connect pot "RV2" to the output of this op-amp.
(I didn't check the filter values)

[AndrewT]

Hi,
Kevin is almost right using the inverting opamp as the adder with the two 8k1 feeding the virtual earth.
Now, instead of using the 20k pot after the adder, which will alter the source impedance of the following stage, take the output pot back to the virtual earth and use it as a variable gain/attenuation feedback resistor. This maintains a low output impedance which the next filter stage needs to maintain the Linkwitz Reilly crossover slopes.
I would change to a 10k pot between output and inverting input to reduce the gain in the bass filter route, or increase the 8k1 to 16k or even 20k.

The bass speaker must be connected out of phase with the trebles to correct for the inverting stage of the adding/summing opamp.

[Speedskater]

I'm not sure that I like the idea of a volume/level control (with long leads) inside the feedback loop. I try to keep the feedback circuit as physically small as possible.

[jcx]

simpler solution:

move the LF pot&buffer to the output of either of the low pass filter section outputs

split the lowpass filter's 1st section input 7.5K into two 15K, one to each of L/R buffered inputs - replacing/eliminating the 8K resistors

as for the design of such a crossover I'd look at Linkwitz' site:
http://www.linkwitzlab.com/Pluto/subwoofer.htm
shows an example of crossing over to a sub (actually 2 subs), with eq of the sub response

[AndrewT]

Originally posted by AndrewT
.........Now, instead of using the 20k pot after the adder, which will alter the source impedance of the following stage, take the output pot back to the virtual earth and use it as a variable gain/attenuation feedback resistor.........

Originally posted by Kevin Graf
I'm not sure that I like the idea of a volume/level control (with long leads) inside the feedback loop. I try to keep the feedback circuit as physically small as possible.

It's not a long route.
Pin6 (output) to pin2 (inverting input) is just 8mm and the pot is similar in size. Standard inverting opamp with variable gain.

pot=20k & R1=8k1, gain =+7.8db
pot=10k & R1=8k1, gain =+1.8db
pot=1k0 & R1=8k1, gain=-18db

[pinkmouse]

All good points so far, but the most important thing I would like to point out is that this crossover is merely a small part of the overall speaker package, and cabinet design and driver specs will have much more of an effect than circuit tweaks.

[franck1234]

Like this??

[Conrad Hoffman]

I'm confused. Not a rare event. But, if this is for a summed central subwoofer, it seems like level controls on the L and R are not what you need. They can come through as is (unless the gain is high), whereas it's the sub that needs the level control. I built one very similar to this, with one level control on the front panel, and it worked great.

edit- I'd also bring up that it's easiest to get good sound if your satellites are *not* ported, lest there be odd phase problems around the crossover point. IMO, if they're ported it's a good idea to experiment with stuffing the port. I just picked up a small pair of KEF "Q" speakers for the TV, and they actually come with foam plugs, but little explanation. My guess is when one adds a subwoofer, they'll come in handy.

[franck1234]

AndrewT say:

pot=20k & R1=8k1, gain =+7.8db
pot=10k & R1=8k1, gain =+1.8db
pot=1k0 & R1=8k1, gain=-18db


It is this formula which you use to obtain these results?

Vs = - (R2 / R1) x Ve

Which value give to the tension of Ve entry?