Print

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

[Ivan Beaver]

  I've had it make such a significant audible difference so many times I'm just frankly surprised to hear you say you havent been able to measure. ??  Hey, I can eq out a bit above 80 hz if i have too much. That's a cheap easy problem to fix.  Conversely, having sufficient energy at 40 to 70 hz to work with is critical. Have a great day!
Chuck

I never said it won't make the low freq louder-it will.

But when people talk about FREQ RESPONSE, THAT is the relation of the lower freq to the upper freq.

It is NOT about reproducing a freq.  HECK-ANY tweeter can EASILY reproduce 1Hz with NO problem what so ever.

But how LOUD that 1Hz is in relation to the REST of the freq response-THAT is a completely different issue.

If you wanted to use the "more subs makes it go "lower" then that would work with ANY type subs-not just horn subs.

The freq response (-3, -4 -5 -10 etc) HAS to be tied to the sensitivity-which "should" be an average SPL across the intended freq band (but in many cases it is not) and in some cases is simple a peak OUTSIDE the intended freq usage band.

IF you don't TIE the sensitivity to the freq response-then all you have is meaningless numbers that somehow people "want" to believe makes some sense.

But with no relationship-you HAVE to ask yourself "So the -3dB point is 3dB from WHAT?"

If you want to mean that is it is 3 dB down from "something" (that is undefined), then why not rate all subs down to 1 Hz?  or lower?

BTW I have seen some many subs that are rated at -3dB at some freq, and yet are actually 10dB down from the rated sensitivity.

I have seen some that are actually TWENTY 20dB down from the rated sensitivity.

Yet people still quote them as saying they have a freq response down to "whatever".

LOOK CLOSER AT THE SPECS!!!!!!!!!!!!!!!!!!!  And what they REALLY mean.

It DOES make a difference.  Of course you will EASILY hear the difference in a side by side.  But short of a side by side-all you have is the specs to go on-ASSUMING the manufacturer provides a MEASURED response-NOT just some "simple numbers" that you have no idea where they came from.

The marketing departments ideas and yours are often VERY different.

[Mark Wilkinson]

Of course the rest of the response will also be 6dB higher-so the "net" increase in lower freq in relation to the upper freq of the cabinet is next to nothing.

So it is not going any lower-just the more cabinets are getting louder.

Sure 30Hz gets louder-but so does 60Hz and so forth-so the ratio stays the same (give or take)

I hear all the time about how adding multiple horn subs together extends the low freq response.

HOWEVER I have never seen ANY measured evidence of this.

The horn subs I have MEASURED (in singles and multiples) do not extend the low freq response (in relation to the rest of the cabinet) by more than a Hz or two, when used in multiples.  Certainly not an octave.

What the multiple sub DO offer is a better impedance loading to the air-which smooths out the bottom of the response curve making it "appear" to have more low freq, but in reality it is not going any lower.
The ripples in the response are just smoothed out-giving more "overall energy" on the bottom-but not lower.

I would love to see some evidence that would prove me wrong-and I will take back what I say.  But as of now I stand by my experience and measurements.

What I would like to see is 1-2 and 4 subs measured using the same drive voltage with maybe 2-3Hz resolution.

Here's 1,2, and 4 labhorns all with 2.0v on each box, .......albeit at only 5 hz increments.  Steady sine waves.
For me at least, it sure seems to support what you are saying....

[chuck clark]

Thank you Mark! Amazing +14db at 65 hz.  I haven't fooled w/ Lab subs.  what's the mouth size of 4 Lab subs?  I don't suppose the dimensions would bear any relation to the 1/4 wavelength of 65 hz?  Pathlength to measurement mic variables and any major reflective boundary bounces notwithstanding. (or any other out-of phase arrivals from designs that utilize energy coming off both sides of the driver in SOME tuned way)
Plus 14 db - that is truly astounding, and roughly +10db thru the entire operating range! When you factor in the loudness curves below 80 hz narrowing significantly you can see why the actual live experience is nearly 3x the low end! Funstuff! May you all experience a bit of awesomeness on your way to a measurement!
Chuck
 

[Ivan Beaver]

Thank you Mark! Amazing +14db at 65 hz.  I haven't fooled w/ Lab subs.  what's the mouth size of 4 Lab subs?  I don't suppose the dimensions would bear any relation to the 1/4 wavelength of 65 hz?  Pathlength to measurement mic variables and any major reflective boundary bounces notwithstanding. (or any other out-of phase arrivals from designs that utilize energy coming off both sides of the driver in SOME tuned way)
Plus 14 db - that is truly astounding, and roughly +10db thru the entire operating range! When you factor in the loudness curves below 80 hz narrowing significantly you can see why the actual live experience is nearly 3x the low end! Funstuff! May you all experience a bit of awesomeness on your way to a measurement!
Chuck
 

Of course 4 subs over 1 would be a "technical" 12dB gain in output.

If the 35Hz happened to be in a "low ripple" of the single cabinet and when the others were added it simply "filled it it" (due to the better impedance coupling of multiple cabinets), then that would explain the additional gain.

However I have never gotten good meaningful data by using sine waves and trying to measure SPL.  The slightest little variance in mic position-position of the cabinets in relation to each other and boundaries, can give very different results.

It might work if the measurements were in the middle of a field with no reflections around (including the other cabinets that would be added later).

Even having cabinet "sitting around"  when measuring the single or double can affect the results-due to the tuning of the "unused" cabinets and this causing "suckout" or notches in the response, that if centered on a particular freq could give some "unrealistic" SPL readings.

Single tones are better than nothing-but I have spent a lot of time in years past trying to make sense of what I was measuring-and walked away more confused than when I started.

[Tom Danley]

Of course 4 subs over 1 would be a "technical" 12dB gain in output.

If the 35Hz happened to be in a "low ripple" of the single cabinet and when the others were added it simply "filled it it" (due to the better impedance coupling of multiple cabinets), then that would explain the additional gain.

However I have never gotten good meaningful data by using sine waves and trying to measure SPL.  The slightest little variance in mic position-position of the cabinets in relation to each other and boundaries, can give very different results.

It might work if the measurements were in the middle of a field with no reflections around (including the other cabinets that would be added later).

Even having cabinet "sitting around"  when measuring the single or double can affect the results-due to the tuning of the "unused" cabinets and this causing "suckout" or notches in the response, that if centered on a particular freq could give some "unrealistic" SPL readings.

Single tones are better than nothing-but I have spent a lot of time in years past trying to make sense of what I was measuring-and walked away more confused than when I started.

Hi All
I might be able to help here. 
Horns are complicated things, perhaps most confusing is you can have something that looks just like a horn, but is acting much more like a duct than an acoustic transformer. It all depends on the acoustic dimensions, not physical dimensions of what’s involved and not what it  looks like.

Part of what sets the low frequency cutoff of an actual horn, it how long the path length is, for bass horns, that is the most important factor I think as the mouth area or lack of effects the size of the peaks and dips in the response curve. The bass horn has it’s lowest peak at what corresponds to the acoustic quarter wavelength and so one cannot have a short bass horn (sort of like a Giant Shrimp?)

The acoustic length is how one can see a small change in the low corner if one stacks up the boxes the right way.  The acoustic length of a round horn is what one expects plus one finds that the “active” part of the horn extends slightly forward of the actual mouth, what has been called a “mouth bubble” at work as some of the designs “play” with that.   
The bubble is an apt way to describe it I thought because if one imagines a shallow bubble at the mouth of a round horn, that line is an equal pressure gradient where  the outer edges are still bounded by the horns physical walls and an arc formed by the natural expansion.   An old thumb rule was that it adds an acoustic length roughly equal to about half of the mouth radius.   So, even though the active part is protruding forward of the horn itself at the center, the pressure is still contained by the horn walls.  Hope that makes sense.

So if one has horns with the right internal shape (for example W bins don’t do this well) and you stack them up the right way, one can lower the low corner… a little bit,  because the bubble is larger and makes the path length longer.  Keep in mind, lowering the corner an octave, requires doubling the total path length and not at all realistic (short of a measurement).

A general comment on our industry
In the real world, we never or very rarely deal with anything that operates on the simplest principles and usually our world involves very complicated things mixed in.  Take sound for example, you can’t see it so one has to imagine or picture how it works and it is easy to wrong because there are few ways to prove ones picture of it.

For that reason, a proper measurement always trumps a computer model and the strength of computer modeling in design, is only as good as it’s ability to predict what you measure in the real thing and not make a cool looking prediction.     
Marketing at worst is all about what you can get people to believe while at best a proper measurement accurately reflects the acoustic reality you were in.
 
As demonstrated here, even simple sine wave measurements can tell you something useful.  In the olden  Servodrive days at Intersonics, I spent many afternoons of quality time out in the parking lot on my hands and knees with a an oscillator and sound level meter.
It has never been easier or cheaper to take good measurements than now; I wish more people were comfortable doing it.
Best,
Tom Danley
Danley Sound Labs

Fwiw, In the Loudspeaker chapter in the 5th edition of “Handbook for Sound Engineers” I tried to spend some extra time on horns in particular and how they work on the inside.

[chuck clark]

Hi Tom!  Interesting! I was unaware of the "bubble effect " extending the effective 1/4 wavelength of the horn slightly past the mouth.  As a lifelong Bass nut and early adopter of Danley technology I still have a pair of SDL-4's in my garage! Say, do you happen to know where I can get belts for them ? lol
Have a super day,
 Chuck

[Ivan Beaver]

  As a lifelong Bass nut and early adopter of Danley technology I still have a pair of SDL-4's in my garage! Say, do you happen to know where I can get belts for them ? lol
Have a super day,
 Chuck

I purchased the first SDL5's in the Washington DC area-back in the mid 80s or so.  So I have been around Toms work for awhile.

Belts are not easy to come by-especially for the ones (SDL4s and 5's) that used a "timing belt" going between the motor and the rotary to linear converter.

[chuck clark]

I used to talk w/ a guy named "Nehru" at Innersonic.  Wish I had bought more spare belts when you could get them!  I've gone to a vacuum cleaner repair shop and found some reasonable replacement belts but weren't sure of the exact length spec. We seem to burn belts up when feeding more than 400-500 watts to it.  So now they've been sitting in the garage for some 27-28 + years. I was always abit dissappointed that Tom didn't come up with a more dependable replacement kit- perhaps gear driven or something. Oh well,...hey, maybe I better patent that idea hmm?!
  I appreciate your dedication to accuracy Ivan, and actually enjoy your posts on here. I just don't have as much time to school the predators and victims of "better sound thru marketing" on here like you seem to. Lol!  And hey... and old know it all actually learned a couple things.!
Have a super day!
Chuck

[Mark Wilkinson]

Hi Chuck, the labs were all side by side when grouped together, with unused ones as far away as possible, and shorted. All I was really trying to test was how peaks and valleys smooth out with multiple boxes, and how close to +6db a doubling of boxes actually  produced. Measurements were really too close to my house to be taken too seriously. And as Ivan points out, sine waves can read funky even inch by inch with reflections around.

And Hi Ivan, a question if I may..... it seems readings I would get from pink noise should ultimately be the same as just a bunch of sine readings...iow, the exact same reflection problems should show up freq by freq in both?

And Hi Tom,  thank you for the explanation, and many thanks for encouraging simplistic testing!

[Ivan Beaver]

. We seem to burn belts up when feeding more than 400-500 watts to it. 

Well considering that they were rated for 400 watts------that might explain it

[ProSoundWeb Community]