Using a Low Q driver

[PeteMcK]

I bought this 8" driver some years ago when I was naive enough to believe the published specs:
Altronics - Your One Stop Audio Visual & Electronics Supplier
Turns out Fs was much higher (50.5Hz), Vas lower - 32.8 L, Qts a bit higher but still low, .27, eff. a bit higher at 93dB.
Looks like it would be good as FLH mid in a small PA system, but I wanted to use it as a 'sub', well from 40-150Hz anyway.
I thought of trying it in BP enclosure which would have been nice & small, but it lost too much efficiency (-8dB). Like most low Q drivers, it models well in an EBS alignment, in this case 30L tuned to 42Hz got it down to 40Hz, but I didn't want that step up to the mids. Adding a 1st order 135Hz LPF in WINISD gives it a nice roll of from ~170Hz with only about 3dB loss in overall efficiency, which I can live with.
The LPF turns out to be 7mH, & I just happen to have a 6.66mh inductor on hand, the box is an old test box I had lying around cut down to size, It came out nearly a cube, so quite compact.
Should have it wired up this weekend, not sure what I'm going to use it for yet though....

[woody]

Is that 50.5hz before you broke the driver in?

[PeteMcK]

No, after.... :-(

[head_unit]

Originally Posted by PeteMcK

Like most low Q drivers, it models well in an EBS alignment, in this case 30L tuned to 42Hz got it down to 40Hz, but I didn't want that step up to the mids...

Well, if you play around with the slope of the shelf, and the positioning in the room, it actually sounds to me like you're on to something. More like a woofer with good bass extension than a SUBwoofer (which to me means capable of really high SPL for movies-your philosophy may differ).

Originally Posted by PeteMcK

The LPF turns out to be 7mH...

As I've ranted elsewhere, passive sub crossovers just don't really work. They don't work in the bass because of the impedance peak at resonance (twin peaks for ported) and they don't work at higher frequencies due to the semi-inductance of the voice coil. Yeah, the coil will cut out the highs a bit, but you really need an active crossover. I used to think "oh, then a 2nd order will work better due to the parallel capacitor" but that actually made things worse.

[PeteMcK]

re:'More like a woofer with good bass extension ' - I agree
I don't get your 2nd point though, the coil gives me flat FR where I want it, and the 1st order slope allows me to use the 'sub' to compensate for baffle step with a simple level adjustment...

[mashaffer]

You might try modeling a horn or tapped horn as that can give you some extension. Not sure about how that low Qts would effect it though. Another easy thing to try would be ripole. Again I think that Q is a bit low but it is pretty cheap to knock one up.

mike

[mavric]

bump, mavric.
quit guessing, and listen. no hard feelings, but bo not cross over at 170hz??? I would give more advice on this, howver it is late. your sub should be at or around 80hz @ 18db slope, use the your crossover to compensate your slope for your sub/high con fig.

[head_unit]

Originally Posted by PeteMcK

I don't get your 2nd point though, the coil gives me flat FR where I want it, and the 1st order slope allows me to use the 'sub' to compensate for baffle step with a simple level adjustment...

Ah, sorry, I just noticed you're in Australia, so I have to turn my frequency response simulation upside down!!


OK, for real now, I think there are 2 points:

1) Extended Bass Shelf:
When I think of this alignment, to me it means the port is really kind of too undertuned, so the response does not roll off smoothly. From the low frequencies up, you could say the response rises, then kind of flattens, then rises again. It sounds like you're trying to kill off the second rise maybe?


2) Passive crossovers for bass:
If I presume this 8Ω woofer has an Re of say 5.6, the 6.6 mH coil would make a first order lowpass around 130 Hz (not 170 Hz)-IF the impedance of the woofer was like a resistor. But in reality in the box tuned to 42, there will be an impedance peak below 42, and impedance null at 42, another peak maybe up around 60-80 Hz, then a broad flattening to down near Re maybe in the low 100s, but above that the woofer impedance will start to rise due to the voice coils semi-inductance.

So the coil's effect is:
higher frequencies: roughly -3 dB per octave (not -6 dB)
around 100ish: maybe -6 dB per octave
below 100: a wavy, lumpy looking mess due to interaction with the complex impedance due to the resonance peaks.

At least the lowpass coil kind of does a little something, albeit nothing like what a simple crossover calculator tells you. A "highpass capacitor" has much worse interaction, and really doesn't work well at all.


Of course, to REALLY see what's going on would require measuring the woofer's actual impedance curve, but I can say from experience the above will be fairly true.