ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => Audio Measurement and Testing => Topic started by: Frank Koenig on October 18, 2018, 02:12:56 PM

Title: Interesting woofer response
Post by: Frank Koenig on October 18, 2018, 02:12:56 PM
Here's the on-axis ground plane response of a BMS 15N630 15 in. woofer in  a 90 l, 47 Hz vented box. I built these 8 years ago -- time flies -- and am revisiting the settings using newer tools. What's odd to me is the step in the response at 200 Hz. The driver should be well within its piston range at that frequency. I'm pretty sure this is not a measurement artifact, as it also shows up (smoothed because of the shorter window) in pseudo free-field (up on a 12 ft. ladder) measurements, as well as some other measurements I did a few years ago. I tried different mic distances with no effect. The measurement rig passes loop-back.

My best guess is that it's due to cabinet standing waves or some other resonance, but would love to hear from more experienced practitioners if this fits some known pattern. Thanks.

--Frank

(http://FTrapGndPlane.jpg)


Title: Re: Interesting woofer response
Post by: Marc Sibilia on October 18, 2018, 03:02:02 PM

My best guess is that it's due to cabinet standing waves or some other resonance, but would love to hear from more experienced practitioners if this fits some known pattern. Thanks.

I thought I knew what was going on until I tried to write something explaining it, and now I realize I don't understand either.  But I have seen the exact same thing.  I will wait for an answer from someone who knows acoustics better than me.

I built some subs, 27" internally from the baffle to the rear panel.  It is clearly a resonance.  I have SPL plot both undamped and with some internal foam damping, so you can see in the undamped that the notches are deep, and that there is attenuation below resonance and reinforcement above resonance.

Marc
 
Title: Re: Interesting woofer response
Post by: Mark Wilkinson on October 18, 2018, 03:42:39 PM
Hi Frank, Maybe the step is just part of the driver's natural response....??

Here's BMS's measurement of the 15n630 in a similarly sized box 
(albeit it's a closed box, which shouldn't matter much for the frequencies of interest.)
Title: Re: Interesting woofer response
Post by: Art Welter on October 18, 2018, 05:22:10 PM
My best guess is that it's due to cabinet standing waves or some other resonance, but would love to hear from more experienced practitioners if this fits some known pattern. Thanks.
Frank,

Looks like a natural response step, not a resonance problem, which would be more narrow band.

Also looks like the 15N630 has at least two quite different measured responses under what are stated to be the same conditions. The one Mark W. posted looks more similar to your driver's response than the 1/14/2005 version I found online.

Art
Title: Re: Interesting woofer response
Post by: Mark Wilkinson on October 18, 2018, 05:39:05 PM
Frank,

Looks like a natural response step, not a resonance problem, which would be more narrow band.

Also looks like the 15N630 has at least two quite different measured responses under what are stated to be the same conditions. The one Mark W. posted looks more similar to your driver's response than the 1/14/2005 version I found online.

Art

Hi Art, 

What's your take on why we see such different curves/specs from the same manufacturer on their same product?   
(All too often I might add....)

I don't mean to be questioning their integrity....I just don't get it.
Title: Re: Interesting woofer response
Post by: Art Welter on October 18, 2018, 09:56:31 PM
Hi Art, 

What's your take on why we see such different curves/specs from the same manufacturer on their same product?   
(All too often I might add....)

I don't mean to be questioning their integrity....I just don't get it.
Differences between half space and whole space measurements can account for large changes.
Sometimes the product is changed, yet the model number remains the same.
Sometimes a trace is mis-labled, and goes to the wrong spec sheet entirely.

The differences seen between the two "anechoic chamber" 15NC30 charts are pretty gross, whatever the cause.

Art
Title: Re: Interesting woofer response
Post by: Uwe Riemer2 on October 19, 2018, 02:25:03 AM
Here's the on-axis ground plane response of a BMS 15N630 15 in. woofer in  a 90 l, 47 Hz vented box. I built these 8 years ago -- time flies -- and am revisiting the settings using newer tools. What's odd to me is the step in the response at 200 Hz. The driver should be well within its piston range at that frequency. I'm pretty sure this is not a measurement artifact, as it also shows up (smoothed because of the shorter window) in pseudo free-field (up on a 12 ft. ladder) measurements, as well as some other measurements I did a few years ago. I tried different mic distances with no effect. The measurement rig passes loop-back.

My best guess is that it's due to cabinet standing waves or some other resonance, but would love to hear from more experienced practitioners if this fits some known pattern. Thanks.

--Frank

(http://FTrapGndPlane.jpg)

Search the web for Baffle Step or Loudspeaker Enclosure Diffraction
e.g. https://www.trueaudio.com/st_diff1.htm
If it would be a resonance ( happens in "small" enclosures with long vents ), it would show up in an impedance curve and the  undampened enclosure would enhance the effect
Title: Re: Interesting woofer response
Post by: Frank Koenig on October 19, 2018, 01:16:10 PM
Thank you all for the responses -- so to speak  ;) I agree that resonances usually show up as a peak or dip, not a clean step like this. When I built these boxes I spent a lot of time scrutinizing high-resolution impedance curves and did in fact hunt down a number of small box resonances that I fixed with the application of glass wool.

The baffle step theory is interesting (thanks, Uwe, for that link), indeed there is an upward  step of ~6dB,  but I see this with the speaker's tush firmly on the ground, which I would expect to behave as half-space at this frequency. Half-space and full-space frequency responses match quite well up to ~700 Hz. See attached. Green = half-space, 100 ms window. Gold = full-space, 13 ms window.

I hope to experiment some more when I get the time and energy and get to the bottom of this. For now I've EQ'd it out and will take them to a gig (mainly voice in a sucky, reverberant room) Saturday.

As for manufacturers' frequency response plots, I've pretty much given up looking at them. I rarely know all the conditions and what I measure never agrees anyway...

--Frank

(http://FTrapGNDvsFREE.jpg)


Title: Re: Interesting woofer response
Post by: Len Zenith Jr on October 19, 2018, 01:56:04 PM

The baffle step theory is interesting (thanks, Uwe, for that link), indeed there is an upward  step of ~6dB,  but I see this with the speaker's tush firmly on the ground, which I would expect to behave as half-space at this frequency. Half-space and full-space frequency responses match quite well up to ~700 Hz.


Your 1/2 space vs full space baffle step phenomena seen up on the ladder, now becomes 1/4 space vs 1/2 space on the ground. The baffle becomes a boundary for freq above ~(4560/shortest baffle dimension in inches). The graphs would look about the same with floor bounce adding in the nulls you see up high.
Title: Re: Interesting woofer response
Post by: Frank Koenig on October 19, 2018, 02:32:03 PM
Your 1/2 space vs full space baffle step phenomena seen up on the ladder, now becomes 1/4 space vs 1/2 space on the ground. The baffle becomes a boundary for freq above ~(4560/shortest baffle dimension in inches). The graphs would look about the same with floor bounce adding in the nulls you see up high.

Len, yes, I see what you're saying. By your heuristic the critical dimension for a 200 Hz step would be 23 inches (lamda / 3), which is a good bit larger than the shortest baffle dimension, while it is on the order of the longest baffle dimension. I was assuming (perhaps incorrectly) that the 5.7 ft wavelength at 200 Hz would just "swallow" the whole enclosure when sitting on the ground. The wave equation finds its way - as some famous engineer (don't remember who) said.  I'm just not so good at picturing it. --Frank

Correction: The shortest baffle dimension is actually 21.5 in. Pretty close. I should try extending the baffle and see if the step moves.
Title: Re: Interesting woofer response
Post by: Len Zenith Jr on October 19, 2018, 03:33:53 PM

Correction: The shortest baffle dimension is actually 21.5 in. Pretty close. I should try extending the baffle and see if the step moves.

That would be a very definitive way of determining the response anomaly. The equation I gave is a very rudimentary way of determining baffle step frequency, which is more of a range than a step. You would have to integrate every infinite point on your cone with the distance to every infinite point on every edge of your baffle to get a better result. All while taking into account the volume of the box that takes away from the space. Rounded baffle edges behave differently than sharp ones, etc, etc.

The baffle step is the whole reason many home speakers have 1 tweeter and 2 woofers. While the top woofer plays its whole passband, the bottom woofer only plays from the baffle step down which adds 6dB down there to flatten the response.
Title: Re: Interesting woofer response
Post by: Frank Koenig on October 19, 2018, 05:55:45 PM
That would be a very definitive way of determining the response anomaly.

I tried to do a quickie experiment using Smaart and its automagical windows but there are just too many reflecting surfaces anywhere around here. I need to wait until I get set up in a good measurement environment. I'll report back. Thanks for the help. -F
Title: Re: Interesting woofer response
Post by: Roland Clarke on October 22, 2018, 06:06:41 AM
Another vote for potential baffle step artefact.  Usually not a problem when people are crossing over subs around 80-100 hz.  If it were a resonance, I would expect it to be a sharper peak.
Title: Re: Interesting woofer response
Post by: Marcel de Graaf on October 22, 2018, 10:47:12 AM
Hi Frank,

Interesting topic.

I think its the baffle step. Linkwitz had investigate the baffle step in the past and well good described.

http://www.linkwitzlab.com/diffraction.htm.

In his reading there are two things going on. The baffle step AND the edge diffraction.

The baffle step isn`t part of the acoustic power (efficiency) response of the woofer. In basis it`s the acoustic dimensions that put extra forward directivity to the system. In theory a 6db/octave rise from low to high freq.

gr. Marcel
Title: Re: Interesting woofer response
Post by: Mark Wilkinson on October 22, 2018, 12:59:51 PM


The baffle step isn`t part of the acoustic power (efficiency) response of the woofer. In basis it`s the acoustic dimensions that put extra forward directivity to the system. In theory a 6db/octave rise from low to high freq.

gr. Marcel

Yes, a change in directivity. 

Best literature I've found says the directivity increase begins when box dimension is about 1/10th wavelength, and is fully in effect when box dimension is 1/2 wavelength (potentially 6dB).
So maybe it is better described as a baffle ramp, than as a step.
I think it got the name step because once it reaches +6dB, there is no further dB rise with increasing frequency.

I guess I'd look at off-axis response for directivity changes.... vs your box dimensions,.....to try to separate natural response from any baffle step response.