I'm no EE, but I don't believe reversing the polarity of an AC signal can have any
effect on a speaker beyond changing its phase.

 

Also check to see if your subwoofer's amp has a phase switch. Many do.

 

sorry, been enjoying life

oth, reversing phase 180 degrees acts kind of like a time delay/advance, and can cure many mid bass phse cancellation issues.

Jad, I'll try to write one up within this week and post it in the FAQ if you guys want it really simple math actually

 

I agree that reversing the polarity, by changing the phase, alters the timing
of the sound, just not that "this means the subwoofer's cone would be firing
towards itself instead of firing outwards" and risks damaging the speaker or
requires changing of the amp gain.

 

Sputniks, we really are on the same page on slopes and their effect--I just think my current speakers won't last nearly as long as yours!
My "system" is also not really a system yet, so maybe it's not really applicable anyways.

Right now I have a HU with stock speakers, so low volumes to me often mean less than 5W RMS. Maybe that's the real issue. And I could care less if I blow the stock speakers since I have new ones and an amp sitting under the xmas tree. I don't listen to much bass-heavy music. Lots of rock, some DMB type stuff, ...basically drums and notes from a bass guitar are all the lows my speakers see.

Here's roughly what I've been doing now:
-At <~40% volume I sometimes run the speakers full range
-At ~50% volume I need at least a LPF at 63Hz@-6dB/octave
-At ~60% volume I need at least 80Hz@-6dB/octave, and on some types of music I have to set it to 80Hz@-12dB/octave or turn it down to get it clean.
-At ~70% volume I generally need 80Hz@-12dB/octave or 125Hz@-6dB/octave for clean sound. I don't often play it that loud.
-For bass-heavy music, 125Hz@-12dB/octave is sometimes the minimum LPF point where it sounds clean. 80Hz@-18dB/octave would probably work fine too, but for no good reason I rarely want the really steep -18dB/octave or -24dB/octave slopes. But it's not fun to listen to bassy music with no bass, so I pretty much stick with rock or DMB type stuff in the car. With no sub I have improper equipment to play things like the crystal method or tiesto.
___

Maybe I'm just asking for trouble... perhaps at higher power things will change and some kind of filter-aggresiveness/volume relationship won't hold anymore?

And Claus, you have no sub now, how do you deal with it?

 

Ace: As you increase power, your slopes/xover points will have to shift - in this case, steeper slopes/higher xover points. I agree on steep slopes, I prefer more shallow slopes as they seem to result in better mixing.

Just keep in mind that even if you're adjusting your slopes for what sounds like no distortion, it's still pretty borderline. If I were you, I'd simply adjust the slope/freq to the cleanest point at 75% max volume, leave it there, and just cry b/c I'd miss my sub so much

In all honesty, the best response to "What should the guys without a sub do for bass" question is: "buy a sub ".

 

This slope/htz@db stuff makes my brain hurt. Anyone care to explain? I'm assuming that it will drop at output level in db, at X rate (6 or 12) after Y Hertz (LPF).

We're talking about the rate of drop off the woofers will have, to transition into the subwoofer, at a rate of 6 or 12 db per (I don't know what the intervals would be) after it hits the Low Pass Filter. Correct?

This is to ease the work on the woofers, and send it to the subwoofer, correct?

 

^^That's a simple way to put it, but yes. That's what it does.

Speakers can handle any frequency you throw at it. How much power it can handle at various frequencies varies based on the speaker's design.

Generally speaking, a sub is designed to handle 20Hz-120Hz at it's rated power. So we use a LPF to block all frequencies above (that's the xover freq part).

We use slope to ease the transition to the speaker with the next set of frequencies while ensuring that we're not putting too much power to the sub at a freq it's can't handle.

 

Sputniks, thanks--until bringing it up, I hadn't considered the fact that I am changing the input power by an order of magnitude, and distortion/etc won't necessarily stay the same at higher power like I had inadvertently been assuming. Perhaps I can only tolerate the same filter settings 10dB down, or basically not at all... especially since I don't want to destroy the new speakers.

I'll miss having a sub.

Oh well, I'll get one next winter, and my car will sound 10x better in the meantime even if it's not complete.

___

jaeden, the interval these slopes act on is octaves. I'll try to explain octaves and the crossover parameters:

Filters are shown in the form:
Decibels down @ start frequency / octave change
Ideal LP = low passes, high gets attenuated/reduced, no change at the crossover point
Ideal HP = high passes, low " " " " "

Decibels (dB) are done in a logarithmic scale, which is a hard concept to grasp. A difference of 3dB is equal to a difference of 2x, 6dB is 4x, 10dB is 10x, and 12dB is 16x.
To get a factor from a rating of XdB, take 10^(X/10). For 3dB: 10^(3/10)=1.9953 which is pretty much 2.

So if I add 3dB of signal, I double the signal, and if I subtract 3dB signal, I cut it in half. Adding 3dB twice is the same as adding 6dB, or multiplying by 2 twice is the same as multiplying by 4. And for reference, a difference of 1dB (factor of 1.26x) is just a barely audible difference.

Start frequency is straightforward--just the beginning point of the filter.

Per octave: musical octaves are steps of a factor of 2 on the frequency (equals one octave). Meaning that going from 80Hz up one octave goes up to 80*2 = 160Hz, and going from 80Hz down one octave means going to 80 / 2 = 40Hz. So they're pretty easy, just multiply or divide the frequency by 2 to go up or down an octave.

Putting it together:
-12dB / octave @ 80 Hz means that
at 80Hz, things are normal, and logarithmically we weaken the signal as we go downward (High passes, low gets attenuated).
At 40Hz, we are attenuated 12 dB, or in other words reduced by a factor of 16.
At 20Hz, we are attenuated 24 dB, or in other words reduced by a factor of 10^(24/10) = about 250.

Using a different slope, say 6dB, changes the amount of reduction (it would be a factor of 4 instead of a factor of 16 and a factor of 16 instead of 250).

Using a different crossover point is pretty self-explanatory.



Did that make it easy to understand, or did I do a poor job trying to explain it?

And does anyone know, is a half-octave step a factor of sqrt(2)? That seems right...

 

Excellent explanation, Ace.

BTW, I recently spoke with Nick from SI...he mentioned that the first shipment of BM mkIII's are very close to being done. Something like 20 of the baskets had to be retooled to meet his specs (he's that on top of it), so they had to wait for that before the shipment could be completed. Hopefully we'll see ordering open up some time in the next month or two