I'm very interested in the notch filter and hi pass filter diagrams. Sounds like fun to work with. As you may recall, I supplemented the low end with powered woofers (up to 250Hz) and up/rear firing tweeters with simple capacitor hi pass at about 14kHz. Seems consistent with both of your impression of what they need.

Can't PM here though. Catch me on the Omega circle? Or post them here?

Thanks....

Thanks for the feedback!

Another alternative to the speaker-level notch filter is the Bottlehead "The Fix". This is a line-level tone control box (for $75!) that can selectively boost both the bass and treble. I would much rather use passive electronics at the line level than at the speaker level.

Does look a good option. Not as much fun though!

I'm sure Louis would not endorse the notch filter as equivalent to a passive crossover, Dennis obviously has no issue with it, worth experimenting IMO.

I know from my conversations with Louis that he has no issue with inductors, but has an aversion to speaker-level capacitors.

Personally, I have tried notch filters before on a pair of Triangle full range drivers. It killed the sound for me. I would rather explore other means if I am using wide band drivers, including adding additional helper drivers at the top and bottom end, using active bass reinforcement, or choosing better-behaved drivers.

Louis' drivers have their compromises, most apparently in the midbass. The 7" driver that Dennis played with likely had a falling response below 300hz. Its top end is also less extended than the RS5 driver.

I would argue that if we are swapping tubes and capacitors, contouring the response at line level by adding a resistor or two is not that different.

I agree with your last sentence. But we are talking about very little $ and maybe use of caps I already have to play with a minor DIY project. This is a hobby afterall!

In that Dennis keeps up with this thread I got an email from him today correcting me. The circuit he rigged up for my Omegas is a Contour Network, not a notch filter as I wrote. I will have to research the difference when back from vacation. I will also get you the circuits as well.

Thanks for sharing and I find your concept/thoughts on this very interesting as I am an Omega owner myself.

When the RS7 driver based speakers first came out there were 3 different flavors of drivers.

(1) the "Vintage" ones which did not have a whizzer cone,
(2) the "Standard" ones that did have a whizzer cone,
(3) the "AlNiCo" ones which also had a whizzer cone.

I was just wondering ... are your RS7 drivers the "Vintage" type (without a whizzer) or the "Standard"/"AlNiCo" type (with a whizzer cone)?

The "Vintage" ones that do not have a whizzer cone roll off a lot earlier (somewhere @ 17kHz) than the "Standard"/"AlNiCo" ones that roll off a bit higher (somewhere @ 20kHz).

I've owned and heard both the "Vintage" and "Standard" versions of the RS7 drivers and much prefer the "Standard" type drivers.

Thanks! Looking forward to them.

The difference between a Contour Network and a Notch Filter is not that great: a notch filter will parallel a capacitor, an inductor and usually a resistor in series with the driver. The values of the components will determine the frequency, bandwidth and depth of the trap circuit you're installed. In this case, a Contour Network comprises a paralleled resistor and inductor in series with the driver.

Louis uses a single inductor in series with a helper driver in his "1.5 way" models. This parallels a second driver below the crossover frequency (if this were the RS7 driver that would likely be 200hz). The main driver is left to run unimpeded. Its rising response is complemented by the additional bass response of the second driver.

If instead we installed a simple contour network comprised of a resistor and inductor, we could accomplish a similar effect albeit at the cost of lower sensitivity. The inductor in series with the driver acts as a short a low frequencies, allowing the full bass signal to reach the driver. Above the tuning frequency, the signal will be attenuated at 6dB/octave. But we have place a resistor in parallel with the inductor. So as the reactance of the inductor rises with frequency, more current passes through the fixed resistor. This pads down the output above the inductor's tuning frequency.

However, a speaker driver is not a simple resistor. In the RS7 driver, the impedance is likely about 8 ohms at 300hz, but rises to about 16 ohms at 20khz. That means the series resistor plays a smaller role as frequency rises and will attenuate high frequencies to a lesser extent than midrange frequencies.

Example: We want to knock down the level from 300 - 3,000 hz by 3.5 dB.

Our contour filter comprises a 0.53 mH inductor and a 4 ohm resistor. At 300hz, the inductor's impedance rises, effectively forcing current through the paralleled resistor. At 300hz, the impedance of the resistor and driver total 12 ohms vs. 8 ohms for the bare driver. However, at 20khz, the total impedance is now 20 ohms vs. 16 ohms for the bare driver. The added resistance will attenuate the midrange more than the high frequencies.

This option, a contour network comprised of an inductor and resistor, avoids capacitors in the speaker circuit. It may meet with Louis' approval. If you have the power and can handle the loss of 3-4 dB sensitivity it may be a valid choice.