Ok, I have a passive crossover that's rated for an 8ohm speaker.
If it's supposed to cut low frequencies at 800Hz and down 6db per octave,
but I have it going to a 4ohm speaker does that make the crossover point 400Hz or 1600Hz??? (does that make sense?)

Thanks

JEFF

 

400 Hz.

It sounds like it's a 1.5 mH coil.

http://www.the12volt.com/caraudio/crosscalc.asp

 

alright thanks!!

 

Hmmm... wouldn't that be a 100uF decap for the highpass instead of an inductor? Also, if you still need the xover point to be 800Hz (along with keeping the same phase response) then you might want to add a 4ohm power resistor in series with the speaker. You'll take a hit on the overall dB from the speaker but it might save you some time trying to figure out why certain frequencies seem to be "missing" from your soundstage. Then again, if ALL your speakers end up being phase shifted by the same amount (aka. using the same "altered" xover designs) then it might not matter anyways -- you're ear can detect frequency and amplitude but not phase.

But then you'd have those speakers being driven at 400-800Hz where they'll either be damaged or force your amp to work harder than it needs to.

Also, keep in mind that the speaker's rated impedence is usually wrong. I've done some sweeps on speakers before and found quite a few that are off by over a factor of 2!!! A pair of old JLs I had were rated at 8ohms and measured out to around 3.7 (from 30Hz to about 160Hz)!!! Just keep that in mind when you make your calculations.

 

I guess I misinterpreted his statment - if it's a HP 1st order xover it is indeed a cap. However, for a HP of 800 Hz and a Z of 8, it's a 25uf cap... and that 25uF cap WOULD have a HP frequency of 1600 Hz when connected to a 4 ohm driver. Sorry about my error in reading your original post.

Of course, Lego, you wouldn't be confusing DC Resistance (DCR or VC Re) with impedance. Since impedance varies with frequency, nominal impedance is an average.

And while you might be saying that we don't detect phase directly, we of course detect the effects of phase problems - because phase is just time, and in the time domain, waves can interfere with each other, and thus cause audible amplitude and frequency effects.

Or, more simply, you do hear it.

 

Yeah, sorry about that... even with my fancy calculator I still can get the wrong answer. So, you'd agree that he should use a 50uF cap to achieve an 800Hz rolloff with a 4ohm speaker load (C = 1 / (2pi * f * Z))?

As far as the DC resistance, most quality resistors are going to be within +/- 1% of their value over the course of the audible frequency range. Even though the speaker's resistance will vary (can be modeled as a complex load) you can get a good measurement of it at the xover frequency... and that's really all that matters.

If only one speaker is reproducing a certain frequency then the phase makes no difference. It only really matters at the xover point when two speakers could possibly be cancelling each other or causing a peak in the overall response.

I guess if he used the xover as is then he'd lose all the frequencies between 800-1600Hz, not muddling over the 400-800 range as I wrote before. My error. So either he can add the resistor in series (assuming the speaker really is 4ohms at 800Hz) or replace the 25uF cap with a 50uF one. With either solution he still has to figure out the phase response of the other lowpass (or bandpass) xover at it's 800Hz cutoff.

I apologize for the confusion; it's been a while since I've had to apply this stuff to the real world.