Some additional AR4x stuffing info. and off axis response testing

[Zilch]

Carl's first curve is of the unmolested AR4x; I am unfamiliar with the measurement methodology.

The second curve shows the nearfield bass response after restuffing with fiberglass, which, being 2-Pi measurement, should be equivalent to AR's 180° anechoic curve. If the AR curve is the standard, Carl's AR4x is overdamped, meaning overstuffed with too much fiberglass....

[Carlspeak]

I don't know if a Q of 0.95 is considered overdamping. I thought the textbook definition was a Q below 0.7 was getting overdamped. However, I was a bit surprised at the significant loss in the characteristic hump at resonance with the 12 oz. of FG. Johnieo's Q at 12 oz. FG was 1.02, not much different from mine. However, he didn't publish any response tests to show what the effect of stuffing on Q had on response. He was only trying to show the contrast in box Fc at differernt Q's between FG and polyester.

[Zilch]

Pete's trying to tell us something, but I don't know what it is.

I've never done much with Qtc measurements, but I'll make a point of paying more attention in the future; I rely on the nearfield response. It has never been necessary to jam the box max full and then some with fiberglass, in my experience. Seems it would be relatively straightforward to gather the data experimentally.

I believe Qtc ~1.0 should exhibit augmented bass response, certainly when measured nearfield.

[tysontom]

Carl's first curve is of the unmolested AR4x; I am unfamiliar with the measurement methodology.

The second curve shows the nearfield bass response after restuffing with fiberglass, which, being 2-Pi measurement, should be equivalent to AR's 180° anechoic curve. If the AR curve is the standard, Carl's AR4x is overdamped, meaning overstuffed with too much fiberglass....

Yes, you're right, the second curve shows attenuation and slight over-damping; I thought we were talking about the first curve. After looking at the first curve more closely, the rise at resonance seems to be pretty much in line with the factory curves, and I think the "Q" for that woofer was set at 1 or slightly higher, as with the AR-2-series. The 12-inch woofer has a Q more in the .7 range with later models, although the original AR-1 had a Q of 1. A Q of 1 is usually fine and it does not in any way significantly increase "ringing" of the woofer, but does cause a small rise of 1 dB or so at resonance, which is actually desirable.

Close-miking or in-cabinet miking is very close, but not exactly equivalent, to the standard 1 meter 2Pi measurement.

[kkantor]

I don't know if a Q of 0.95 is considered overdamping. I thought the textbook definition was a Q below 0.7 was getting overdamped. However, I was a bit surprised at the significant loss in the characteristic hump at resonance with the 12 oz. of FG. Johnieo's Q at 12 oz. FG was 1.02, not much different from mine. However, he didn't publish any response tests to show what the effect of stuffing on Q had on response. He was only trying to show the contrast in box Fc at differernt Q's between FG and polyester.

Hi, Carl, et al.

According to the formal engineering definition, a second-order system with a Q of the square root of two, (0.707...), is called "critically damped." A Q above this is under-damped, below is over-damped. Thus, a Qtc of 0.95 is considered under-damped.

In general, most sealed box designers tend toward a slightly over-damped response when the cutoff is low, and a slightly under-damped response when the cutoff is higher. There are no rules here, but maybe one might pick something in the ballpark of 0.7 for a 50 Hz tuning, 0.6 for a 35 Hz tuning, and >1.0 for a 100 Hz tuning. The results are probably don't arise from ringing issues, just the shape of the frequency response.

(However, step response overshoot CAN present tricky issues in terms of the voltage clipping of amps and filters.)

-k