Over the pat week I finished modifying my home-built Altec A7s. Based on discussions by Roland R on other boards, I enclosed the woofer/mid-horn driver (515-G) isolating it from the reflex space, leaving a back chamber of approximately 42 liters. The chamber is lined with acoustic (egg crate) foam. I'm running Altec 802s on 511 horns on top.

The reflex chamber, approximately 5-1/2 cubic feet, is dual-ported in the rear, 4-1/8" Dia X 3-3/4" long (40 Hz). The front baffles are two layers of 3/4" plywood (1-1/2" total). The drivers in the reflex chamber are JBL 2225.

The things are driven through a Rane MX-23 with Crown D-75As on top and a Crown Com Tech 200 on the bottom.

The "mid-horn" seems to integrate really well with the 511/802 combination at about 600 Hz. It's surprisingly seamless, even at a few feet away - you don't have the impression that either source is dominant on solo female voice, which I think is pretty good.

The reflex driver is XO at 175 Hz to the mid-horn. Again the integration with the whole seems pretty good, but not as good as with the mid and top horns.

Still fooling with level matching. I haven't done any measurements, principally because I don't know what to do. I have an SPL meter and I may do some investigations with it after some reflection.

Right now I think it's a vast improvement over the stock A7 in that the totally absent bottom is overcome in a pretty musical way. Having braced, damped and altered these cabinets to high heaven, I think this will be the end of the road for this approach. The next time out will be a fresh start with a different configuration altogether.

Can you post pics? I'd like to see your speakers.

• http://wun.smellthecolors.com/ModifiedA7/

Is that an Altec woofer in the horn, and a JBL 2225 down below?

:-P

Dave:^)

I'm thinking that if I were to add another 8 ohm driver and series cap in parallel with the top horn, the lowered impedence would not present a problem because of the limited power in that spectrum. There's no crossover impact since the system uses an active crosover.

Is this a reasonable assumption?

Thanks - and thanks to all for the positive feedback.

In any case, I suggest using a shunt resistor across he tweeter of about 1.5 times the advertised impedance to damp the circuit. The capacitor and tweeter voice coil form a resonator that will have a small peak around the crossover frequency, and a shunt resistor will help damp this peak.