Questions and Answers, Part 1
> The F3 is the Frequency 3 db down, or the frequency at which output is 3
> decibels down from the rest of the audio signal. This is also referred to as
Oh, ok. I always knew that as the 3 db point(s). I never heard it called F3 before. Maybe that's how it's referred to in the audio specs world? But then again, I don't remember seeing it in any equipment specs either.
> the half-power frequency, though that's a bit inaccurate, as far as db. I
> believe the consensus is that 10db = double output perceived by human ears.
> Here's some useless info: above 90 db, sounds are twice as loud to women as
> they are to men. Guess that explains a lot.
Now that's interesting! My wife always complains that I listen to TV/stereo too loud. I thought that maybe I was going deaf. Not that I listen at above 90 db though...
> By the way, every time you double power in watts, you increase output by 3db.
> Hence, it takes 10x as much power to double perceived output.
I'm aware of the first statement, but I didn't follow how that leads to the second sentence. Does this have something to do with our hearing perception being logarithmic?
> The reason for firing the sub backward is wave folding. Some idiots (I won't
> mention any major chains) believe that low frequency sound waves (ones that
> have a wavelength longer than a foot or so) can not be heard in a cabin,
> because "they don't have room to unfold." This is a misnomer. In fact, the
> larger the wavelength, the greater the boost to the output. That's why when I
> build a box, I use a design that gives a slow to medium low-end roll off, with
> an f3 of about 55hz. With the boost from wave-folding (for lack of a better
> description) you will generally get flat cabin response down to 20hz or
> below. I had one customer who didn't believe me when I told him this, so I
> built him a box (about twice the size of my original design) that had an f3 of
> 22hz. Since the resonant frequency of the driver was below this, it was
> really no problem. However, the low bass overpowered everything, and the
> in-car response was flat to 12hz. The only thing the sub did was shake the
> living hell out of the car. He ended up putting it in his living room, and it
> works quite well there.
> F3 can also refer to the system's response in the cabin. While f3's are
> generally calculated in a large room, 1 meter from the speaker with a
> directional microphone, they may also (for car audio) be used to describe
> cabin-response (using omni-directional microphones) of the system.
It sounds like you have a great wealth of knowledge in acoustics. Ever consider writing your thoughts down and sending them to Brian to put on the web site? For example, you could describe how to plan/calculate what a sub and enclosure one may require, and then how to build it. I know that I'd love to see that.
I never mentioned anywhere in my post that one of my hobbies is woodworking. Things like furniture and stuff. I would really enjoy building a custom enclosure for my GP. Maybe I could make it less intrusive in terms of space/location sacrificed.
> You're right about the cost of cables. High suppression RCA's 12ft long would
> probably run a good $60-$80 dollars. If not more.
> One last suggestion. Always use 12 guage speaker wire with high current
> amplifiers. Being an EE, you should realize the effect that small wire will
> have on the signal due to voltage drops. Cheap 16 guage (depending on brand)
> can have a voltage drop of 1 or 2 volts in as little as 20 feet. This
> wouldn't be so bad, if only it wasn't 10% or more of the voltage. High
> current amplifiers react the worst, as they run lower voltages to maintain the
> same power. And if I'm not mistaken, that Soundstream of yours is technically
> a high current amp. If it's Reference, it should be.
According to the specs, the four primary channels are "high power", not high current. The sub channel is switchable for high power or high current. I'm running that channel in high power mode also because the sub is 4 ohms. The specs recommend to switch to high power for loads under an ohm.
I agree with you in general. Boy, you should see some problems I've seen in high speed backplanes in terms of attenuation, reflection, ground-bounce, etc. Try to picture this. I once saw a transient (only lasted about 5 ns) 2V drop across a one inch long gold plated connector pin.
But getting back to the issue, I'm not moving THAT much current through the cables. Remember, that my loads to the amp are all 4 ohm, not that low. The amp is rated (at 12V) to put out 25 W into a 4 ohm load for four channels. Soundstream uses an unregulated supply, so I'm probably getting more like 30 W. For a 4 ohm load, that translates to about 2.7 amps. Not too bad.
You're right to point out that the current going through the wires is the critical parameter to deal with. If the wires can't carry the current comfortably, it will kill the sound. What makes it even better is that the attenuation will vary based on frequency. If an amp of higher output was used (or if speakers with much less impedance that 4 ohms) then yes, 12 gauge wires would have been a necessity.
Now on the sub channel the amp I have will put out 100W, which again is more like 120 W with the engine running. For that I did use 16 gauge.
I used pretty decent wires from Streetwires. They have a lot of conductor strands that reduce the wire's impedance for AC signals like we're running. You're 100% right. Good wires are critical, and their cost is usually peanuts compared to the cost of a decent upgrade.