Ported/bandpass: how much group delay is too much?

[filtercoffee]

Hi, I was playing with WinISD to design a ported or bandpass box to get a lot of SPL between 24 and 70hz, with a woofer with an Fs of 37hz. Box size was fairly large, 90L for ported and more than that for bandpass. I noticed with both ported and bandpass designs, the group delay was quite high in some frequencies, over 60ms.

Is this a red flag? What does it mean and what effect does it have?
Also, is there such a thing as too much port gain?

 

[Brian Steele]

GD of over 60ms in the passband seems kind of high. Some more details of the two alignments may be helpful.

 

[GM]

It's a good question that we beat to death on the late basslist and the way I remember it the ~consensus was if it had decayed to < 35 ms by 50? Hz it was OK. I mostly remember having built what became known as EBS alignments with tunings as low as ~14 Hz that would later sim mass quantities of GD, but me and everybody that auditioned them never noticed anything but wonderfully 'tight'/'full' <sub>bass, so for me it's a 'much ado about nothing' down low, ditto compression horns higher up since the XO knocks it down.

GM

edit: just remembered JM saved a few posts that kinda jives with my dim memory; anyway FWIW: Discussion of Group Delay in Loudspeakers</sub>

 

[freddi]

I made some BP4 about 28 yrs back with 12" and ran them crudely passively to some AR M1 mini speakers - on a good recording of a subway train (maybe Pierre Verany CD?), it sounded pretty taut in the "bass" when the train screeched to a halt on the steel tracks. That was a long time ago - maybe if blended well, its not as much an issue as some other problems - ??

 

[TBTL]

Also, is there such a thing as too much port gain?

If you can live with the peaky response (ported box) or small bandwidth (bandpass box) then it is OK. A peaky response might sound boomy.


Bass tightness is primarily determined by in-room response and integration of subwoofer with main loudspeakers. The former because room modes can cause peaks in the response of 10+ dB, the latter because a lack of mid-bass will make low-bass sound boomy.

 

[filtercoffee]

Sorry I lost the winisd file and am having a hard time replicating it.
However, I think it's because I tuned the ported chamber to a lower frequency than the sealed chamber.

 

[filtercoffee]

GM probably because there's not much musical content below 30hz, it all just sounds like a really low "hrmmmmm" for lack of better term. I can't imagine being able to describe < 30 hz as "tight".

I played around with EQ and phase and I think I agree with TBTL about midbass and integration with the sub making a pretty huge difference, however what I'm not yet sure about is how much smoothly transitioning the phase in the sub/midbass crossover region really matters. I wonder if the phase matters a lot, or if smoothly transitioning the magnitude is all that matters.

 

[TBTL]

I wonder if the phase matters a lot, or if smoothly transitioning the magnitude is all that matters.

If you have access to a DSP, try listening to allpass filters. I find it surprisingly difficult to hear 720 degrees phase rotation around 100 Hz (which is twice the phase rotation of a fourth order crossover).
A pc + Equalizer APO can also be used instead.