Subwoofers - Grille off or on?


I know speakers always sound best with the grilles off. Does this apply for subs too, or does it not matter?
nemesis1218
We find it easy to hear cloth soften the attack of a kick-drum and reduce the percussiveness of a string bass. It is best described as hearing a 'mumble' or a 'stumble' in the 50-70Hz range.

Huh? The "attack" on a kick drum is around 5 KHz. Most sound engineers can confirm this for you. How this can affect a subwoofer would be a mystery unless it were run full range like a bass guitar amp/speaker.

The main issue (apart from some very slight attenuation in higher frequencies) with grills is the way the baffle is designed. Some baffles have very sharp corners and the grill acts to help reduce the severity of this baffle step and edge diffraction.
I would add that the baffle step is usually somewhere around 400 to 800 Hz and edge diffraction is something that affects imaging mostly in the midrange and treble. Neither of these frequencies are terribly important for a high quality subwoofer that should not output high amounts of harmonic distortion.
About that 'softening the attack of a kick drum'-- You are right about what makes up the attack, which indeed are high-frequency tones that follow the impact of the beater striking the tightly-stretched skin. I should have been more clear, but this does describe the impression.

In as small of a nutshell I can find, here is why:

Any drum-sound takes a few cycles to build-up to its main resonant waveform. That resonance could be at a low frequency from a kick drum or in the middle-range from a conga.

Resonance means taking time to build up to an even louder sound, called magnification. Resonance also means taking a while to stop.

At the very first moment though, what we see on the `scope is a near vertical rise in air pressure. This is the beater pushing the drumhead out towards the mic instantly and then holding it out there for a moment. The side of this rise can be thought of as the fastest (high-frequency) risetime. The 'risetime' for the low-frequency portion to fully build up is several far-longer cycles.

In between these two extremes are all the other resonances which come and go with particular timings and loudness'. These are determined by the resonances in the drumhead and drumshell which impulsive 'thrusts', such as from the beater, always trigger.

All together, these make up the sound of "whats-his-name on a maple Pearl 22" kick drum, with Remo heads and a beater from Drum Workshop, an AKG mic placed right there, run into a Neve mixer, Pulltech EQ, Altec compressor, and onto 1" tape at high speed on a Studer. Oh yeah, with some reverb added. And he was having a good day."

The presence of the grille alters the LF-relationship to all of those other tones/events, by delaying the maximum rise of the low-frequency waveform and its subsequent decay.

What we hear and feel can be described as a less-defined impact, perhaps a more-rounded sound depending on the drum and how it was played and recorded. We can often feel the pace is delayed- less 'drive'. The arrival of the LF-portion of that drum's sound has been blurred, and then told to hang on an extra moment.

This is a dynamic issue, not a steady test-tone issue, because the cloth's resistance to air motion changes non-linearly with the stroke of the woofer: Twice the cone velocity (=2x stoke) creates roughly four times the air drag. FYI, this drag can be translated into a total moving mass (cone + air) that is apparently changing with loudness = 'variable mass'.

If you can somehow put an accurate value for that change into the standard equation of motion for the air molecules, watch how fast a computer can choke! As a side note, a rocket loses moving mass as it burns its fuel. So it speeds up more rapidly each moment, from the same amount of thrust. But the increased speed increases the air's drag on the rocket by a lot more, a drag which also lessens with altitude, but then changes again when the speed of sound is first exceeded... Much computer-time for NASA before going to the moon.

When the grille reaches anything close to a steady back-and-forth motion, its resistive loss is then a) steady, and b) small compared to the cone's stroke. Therefore, its effect is not readily measurable on steady test tones, by MLS, or on pink-noise, but can be seen in an impedance curve.

Thanks for pointing this out. I appreciate the opportunity to clarify.

Best,
Roy

Roy, Do foam grills have the same or similar affects as the cloth grills?