question on active crossovers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
If I were to set up a satellite/bandpass sub system, would I want to use a 3 channel stereo active crossover or a 2 channel stereo active crossover with the sub having it's own crossover which summed the signals to one mono signal? It seems to me that the 3 channel active crossover would work fine, but would I not get a low enough signal for the sub or something? Thanks.

Dave
 
Dave,
I'm not quite clear on where you're trying to go with this. Are you trying to biamp the satellites or do you just want the sub signal broken out?
For what it's worth, I'd go with stereo subs, but if you're set on a mono sub, it's not all that difficult to sum the channels by feeding them into a resistive network.

Grey
 
What I want to do is triamp. I want the system to consist of a satellite/subwoofer combination. I just wasn't sure if the subwoofer needed it's own crossover, or if I could get away with using a 3 way crossover as oppossed to a 2 way crossover.
Could you tell me what the advantages of using stereo subs would be? Thanks.

Dave
 
Dave,
Once upon a time, low frequencies were routinely mono, but it's been years and years since that was true (yet people still act as though it's the case). The logic went--well, if the bass is mono, why bother having two subs? Even then, the logic was shaky, because the sub crossover point was frequently different from the point where the two channels were summed together. Nowadays, there's no validity to it whatsoever.
The other reason you'll hear given for mono subs is that human hearing can't detect the directionality of low frequencies; they're omnidirectional. This is somewhat more valid, but...
There are five counterpoints that I'd like to make:
1) People confuse the idea that sufficiently low frequencies launch from a sub in an omni pattern (rather than cardioid) with lack of directionality in human hearing. They're two different things.
2) Even if you regard human hearing as unable to detect directionality below, say, 30Hz, you'll generally find that they set the crossover point at 70-100Hz, thus losing an octave or two of stereo. This is the same point I made above.
3) Although I've never made a formal experiment of it, I can tell directionality of some pretty low frequencies. Although I'm certain that there's a limit to this ability, I feel pretty confident that I can tell directionality down to at least 30Hz--well into subwoofer territory. I'd like to raise the possibility that this *may* be a learned ability, as I've been playing bass for quite some time. I've never put this idea to the test, so it may be utter bull.
4) If you sum two channels together, there will be a certain inevitable loss due to phase cancellations, etc. It will vary from one recording to another, and from one frequency to another. Even if all of the above were to turn out to be untrue, I'd want to get that last tiny bit of bass information out of the recording.
5) Low frequency drivers have a very difficult job to do. The larger the cone excursion, the higher the distortion. If you double the number of drivers, the cone excursion is reduced for the same SPL, hence lower distortion. Once you have some multiple of two drivers, you might as well split them out and provide them with separate signals.
In short, there are only two disadvantages to stereo subs--money (you knew that was coming), and floor space (which may get into SAF factors). To me the advantages clearly overwhelm the disadvantages.
As far as triamping goes, go for it. Each channel gets a highpass and midrange bandpass. For the subs, you'd either provide stereo subs with a lowpass for each channel or sum to mono for a single sub.

Grey
 
Andy,
You tryin' to say I gotta big head, bub? (Hey, if it works that way, maybe I should inflate my noggin another inch or two...)
Yep, I've heard that one, but I don't think it holds up under scrutiny. That would lead to a loss of directionality below about 2kHz or so (rough calculation based on one wavelength), which clearly isn't right. A better hypothesis is phase relationships between the ears (yes, also related to the distance between the ears), but that one's a lot more slippery, because you're left trying to decide how many degrees of phase rotation the ear/brain are able to detect. Which still leaves me asking whether this is an educatable sort of thing.

Grey
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.