Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[John L Nobile]

Been using a 24db/oct Butterworth for hipass on my subs. I have 48 db and different types available.
Wondering what the best slope and type is for hipass? I've read that steeper slopes introduce more phase but does that matter if used as hipass?

[Keith Broughton]

Been using a 24db/oct Butterworth for hipass on my subs. I have 48 db and different types available.
Wondering what the best slope and type is for hipass? I've read that steeper slopes introduce more phase but does that matter if used as hipass?

Danley lists 24Butterworth so stick with that.

[Ivan Beaver]

Been using a 24db/oct Butterworth for hipass on my subs. I have 48 db and different types available.
Wondering what the best slope and type is for hipass? I've read that steeper slopes introduce more phase but does that matter if used as hipass?

My "go to" filter for highpass on subs is 24dB/Butterworth.

It provides a good bit of protection while being the flattest before it starts to roll off.

The steeper slopes can tend to "ring", but do provide more protection against signals that the sub can't provide.

My opinion is that if you need more protection, either raise the freq a little-or get more subs, or turn it down.

Sometimes I use 24 Butterworths for mid and hF highpass, other times not.  Often Bessel filters are used.

It has been a long time since I have used a L/R filter for high or low pass.

Electrically in amplitude it may "appear" to be better, but the problem is that loudspeakers are not perfect and you have to consider the phase relationship as well.

[Merlijn van Veen]

Actually, all things being equal, Butterworth rings the most at the corner frequency, then comes Linkwitz-Riley and Bessel rings the least. These phenomena are proportional to the filter order. However, Butterworth has the least amount of attenuation at the corner frequency.

[Michael John]

If I recall correctly, for a vented enclosure the displacement just below port resonance rises at about 12 dB/oct at its steepest. Using a 2nd order (12 dB/oct) high-pass here will flatten displacement, for constant voltage, but since music tends to have rising energy at low freq, 3rd order (18 dB/oct) is probably closer to maintaining constant displacement and with 4th order (24 dB/oct), you're pretty much guaranteed to have protection!

Best regards,
Michael

[Merlijn van Veen]

According to Patronis et al. vented enclosures exhibit 4th order Butterworth HPF behavior (electrical analogy) at the tuning frequency by design. Something I have measured numerous times.

Combined with an electronic HPF, typically 4th order Butterworth, to prevent over excursion, the compounded phase shift is 720 deg in the lower operational range with group delay in order of 25 ms to as much as 40 ms.


Verzonden vanaf mijn iPad met Tapatalk

[Ivan Beaver]

According to Patronis et al. vented enclosures exhibit 4th order Butterworth HPF behavior (electrical analogy) at the tuning frequency by design. Something I have measured numerous times.

Combined with an electronic HPF, typically 4th order Butterworth, to prevent over excursion, the compounded phase shift is 720 deg in the lower operational range with group delay in order of 25 ms to as much as 40 ms.


Verzonden vanaf mijn iPad met Tapatalk

What it comes down to is "it depends".

And what is "most important".

If you are not pushing a system hard-with freq down below the tuning freq, then a steep cutoff is not needed.

But if you are going to "abuse" the system, then having a steeper cutoff could help the system "live to play another day".

I do agree that sonically a lower slope or no slope at all is best.  BUT ONLY up to the point that the speaker dies.

At that point, it is NOT good sonically

There is no "one size fits all" answer.

It is a tradeoff of output vs sonic quality.  And most users don't understand the compromises.

So helping to keep the speakers working is a bit closer to the "top of the list" for me.

[Mark Wilkinson]

Actually, all things being equal, Butterworth rings the most at the corner frequency, then comes Linkwitz-Riley and Bessel rings the least. These phenomena are proportional to the filter order. However, Butterworth has the least amount of attenuation at the corner frequency.

Hi Merlijn, I thought LR is just two cascaded BW.   How then can LR ring less?  Does the cascading help cancel ringing? Thx.

edit:  Hi again Merlijn, never mind please, I get it...2 cascaded have more like the slope of 1...

@John, never simple.....ain't it the damn truth 

[John L Nobile]

There truly is no "simple answer".

[dave briar]

There truly is no "simple answer".

Sure there are.  They're just, as Ivan notes, wrong.

   ...dave

[ProSoundWeb Community]