Is there a way to quantify extension?

Got speaker theory questions? This is where you'll find the answers.
Message
Author
ncgrove
Posts: 457
Joined: Wed Jul 06, 2011 7:38 pm
Location: Colorado Springs, CO

Re: Is there a way to quantify extension?

#16 Post by ncgrove »

I like to think of it this way:

Even if the driver can handle a 40Hz tone at the voltage limit, the summed horn mouth area of a stack of less than 4 T39s can't support a 40Hz wave. So there's no point in stressing the driver for content that isn't going to be present at any appreciable level relative to the rest of the pass band.

Or, you can relatively safely, but you're wasting power trying to reproduce something that isn't likely to contribute much.

User avatar
Charles Jenkinson
Posts: 1125
Joined: Sat Aug 06, 2011 3:25 pm
Location: Manchester, UK

Re: Is there a way to quantify extension?

#17 Post by Charles Jenkinson »

Don't give up Bruce. I'm not totally getting it either, ...particularly what EQ has got to do with it. It's still not clear what the relationship is between voltage and driver stress (excursion?) for either 1 cab or the magic 4+ cabs, i.e. whether they (voltage and driver stress) are in fact the same for the different HP limits due to the improved coupling as Grant said - I doubt this is the case because "driver stress" is more of an emotive term, not an engineering concept of linking known measurable or modellable concepts. Despite Missilecrises apparent epiphany, afterwards, he also +1'd what you said.

I like Nathan's view of it, and the concept of the acoustic pass band, i.e. what a horn can and can't reproduce acoustically. Maybe it's best to think of the book stops here point as being the 4+ cabs, so the higher HP limit is as Nathan says, not wasting power, or using up driver excursion fatigue life unnecessarily.

Charles

Edit: I think you should live a bit closer to the edge on HP point. There's some territory to explore, before you get to the bleeding edge, I believe.
2xJ12L (3012HO) switchable/melded
2xT30

Words&graphics - Audio&Acoustics - Hardware&DSP; 3 different paradigms.

Michael Murphy
Posts: 343
Joined: Sun Dec 30, 2012 2:17 pm
Location: Trinidad

Re: Is there a way to quantify extension?

#18 Post by Michael Murphy »

MC wrote

Reasoning: it is these borderline frequencies that will drive you into overexcursion and the less steep the slope the more likely that the pesky 35 Hz fully redlined bass that they "accidentally" drop doesn't pop your woofers. I haven't modeled the driver excursion but I'm sure there is a way to do it (that's how bill gets his voltage limits) based on how many T39's you have stacked

I came across this regarding voltage limits

rec wrote

Yes, If my calculations are right I limited to 48v. I dont think I ever got close to the 48v limit. Based on Bills recommendation of half power here is the formula for the 600w 4012HO driver. 300*8=2400, square root of 2400=48.98v

And to sum up everything said, I asked the same question about doubling cabs and the 5hz decrease to 40hz and it has to do with mouth area. A 12 inch circle has 4 times the surface area of a 6 inch circle but is only twice its diameter and because we measure SPL which is a measurement of air pressure which a unit is psi, pounds per square inch running 4 cabs v plated keeps the Xmax in check and so its safe to go to the 40hz at max recommended power on multiple cabs in open area.

Where is the post, ok educational links, I wont get shot then. I was over at another forum and using hornsrep they can predict xmax, xlim in the sim of a cab design. Im no traitor it came up on a search for direct radiators vs horn loaded designs.

Iv fiddled quit a bit with the crossover settings on the PA+ with the infinity loaded T39"s and they are a 4 ohm driver so at 2 for the cab its 6 and parallel wired its 3ohms (explains now why my QSC with shutting down on thermal protection when I tried to use one 2 channel amp to drive 2T39s and 2 OT12's) and I ran 40hz LR24 up to 46 volts for a long time.

Would I want to risk 40Hz with a quad setup yes at 46v.

And although there is no direct relatioship between V and Db using http://www.sengpielaudio.com/calculator-volt.htm

46V = 33.26db
50V = 33.98db
55V = 34.81db

so that 46V to 55 V jump only netted me 1.55db an unaudible change. But dont quote me on this when the SA 3055 gets here I will be using every cent of its cost to run tests.
Ive dialed in the subs really nice till this some came along

http://www.youtube.com/watch?v=xI65Shx6zqg

This a song I demo the system on, put it run walk away, the acoustical intro throws everyone off then clarity and boom but its a little muddy at high volumes and Im trying to figure out the base line. MC if you want to analyse this for me that would be nice especially what happens from 1.19 onwards

cheers

ps thats what happens for having a party and not inviting me :fruit:

MissileCrisis
Posts: 1128
Joined: Sat Dec 15, 2012 1:30 pm
Location: Walnut Creek, California

Re: Is there a way to quantify extension?

#19 Post by MissileCrisis »

I just love these conversations :). Pumps I'll look into it when I get home from work.
Built
2 x TLAH
1 Titan 39 (lab 12, 15 inch)
1 Tuba 60 (lab 12, 20 inch)
SLA Center
2 x Mini MLTP subwoofers
1 x 212 TH (custom)
2 x Mini-Karlsonators
Owned, but not built
2 x Jack 12 (2512, melded)

Grant Bunter
Posts: 6915
Joined: Mon Nov 28, 2011 11:12 am
Location: Ilfracombe Queensland Australia
Contact:

Re: Is there a way to quantify extension?

#20 Post by Grant Bunter »

MissileCrisis wrote:I just love these conversations :).
Me too!
However, I would still like to know if there is a way to quantify (as well as the method to work it out) the extension I am getting from my 4 x 20" BP102 loaded T39's stacked and V plated.
That being the topic and all...
Built:
DR 250: x 2 melded array, 2x CD horn, March 2012 plans.
T39's: 4 x 20" KL3010LF , 2 x 28" 3012LF.
WH8: x 6 with melded array wired series/parallel.
Bunter's Audio and Lighting "like"s would be most appreciated...

User avatar
Radian
Posts: 2028
Joined: Tue Jul 29, 2008 12:56 pm
Location: Tucson, AZ

Re: Is there a way to quantify extension?

#21 Post by Radian »

pumpsfast wrote: And although there is no direct relatioship between V and Db using http://www.sengpielaudio.com/calculator-volt.htm
Yes there is...that's idea behind the creation of the unit...to relate various quantities.

The ratio of two voltages can be stated in terms of the decibel.

See here: http://www.net-comber.com/decibel.html

There's even a relationship between dB V and dB SPL, but it is not linear.

That's why response curves referenced at 2.83V instead of unit power are ideal to work with. 8)
Good food, good people, good times.

4 - AT
1 - TT
1 - THT Slim
2 - SLA Pro 4x6 Alphalite

Ryan A
Posts: 867
Joined: Thu Mar 01, 2012 11:19 am
Location: Minneapolis, Minnesota

Re: Is there a way to quantify extension?

#22 Post by Ryan A »

Grant Bunter wrote:
MissileCrisis wrote:I just love these conversations :).
Me too!
However, I would still like to know if there is a way to quantify (as well as the method to work it out) the extension I am getting from my 4 x 20" BP102 loaded T39's stacked and V plated.
That being the topic and all...
It's 40hz, since that's what you high pass at. It may be a few hz lower but probably not more than 2-3. Its mostly related to the length of the horn.

Grant Bunter
Posts: 6915
Joined: Mon Nov 28, 2011 11:12 am
Location: Ilfracombe Queensland Australia
Contact:

Re: Is there a way to quantify extension?

#23 Post by Grant Bunter »

Cheers Ryan,

So do you think that's the case regardless of filter type and slope value?
Built:
DR 250: x 2 melded array, 2x CD horn, March 2012 plans.
T39's: 4 x 20" KL3010LF , 2 x 28" 3012LF.
WH8: x 6 with melded array wired series/parallel.
Bunter's Audio and Lighting "like"s would be most appreciated...

MissileCrisis
Posts: 1128
Joined: Sat Dec 15, 2012 1:30 pm
Location: Walnut Creek, California

Re: Is there a way to quantify extension?

#24 Post by MissileCrisis »

Grant Bunter wrote:Cheers Ryan,

So do you think that's the case regardless of filter type and slope value?
The slope/type simplify determines the transfer function of the input signal AKA it changes how much voltage at what frequency gets to the subwoofer, the SPL that you experience at that frequency is entirely dependent on the efficiency at that frequency. E.G. If you send a 40 V sinusoid at 40 Hz when your HP is set to 40 hz it will be attenuated to (in the case of a 48 dB L/R) to 20 V before it gets to the sub. This is how those frequencies are filtered out, there is reduced power made available to them (based on the filter). Now, that being said, if you have flat response down to 38 Hz in the SPL chart that means for a given voltage you'll get the same amount of volume, I know you know this but it needs reiterating. Since usually 38 Hz notes are not sine waves and go to about 60-70 Hz in a bass hit (see below), those frequencies that have been squelched in terms of voltage by the filter are effectively not heard, i.e. less extension.

In order to get the flattest response down to your cutoff you would need to add eq to boost these frequencies with as high a q as possible by about 3-4 dB to counteract the downward slope. Mind you since you're using 48 dB/octave you will not be getting much if anything below the cutoff frequency but at least you'll be flat down to said cutoff, say 40 hz.

BTW here are some pics of that song someone requested be analyzed

Here I sampled one of the low "sweeps". The low end in this note you would have missed with your T39's but notice how over half (if you visually integrate aka take area under the curve) the bass energy is still above your highpass. What this sounds like is a "cleaner" or less deep sounding note but still plently loud.
lowest note.jpg
This second image is a picture of a typical bass hit in that song (drum) Notice how 85-90 percent of the bass energy is centered in the 40-100 Hz region. This would have sounded loud as hell and you wouldn't be missing anything
most of the upper bass.jpg
Mind you I just downloaded a youtube quality version of the song so expect the highs roll off that you see but you get the big picture.
Hope this has helped someone out!
Built
2 x TLAH
1 Titan 39 (lab 12, 15 inch)
1 Tuba 60 (lab 12, 20 inch)
SLA Center
2 x Mini MLTP subwoofers
1 x 212 TH (custom)
2 x Mini-Karlsonators
Owned, but not built
2 x Jack 12 (2512, melded)

User avatar
Radian
Posts: 2028
Joined: Tue Jul 29, 2008 12:56 pm
Location: Tucson, AZ

Re: Is there a way to quantify extension?

#25 Post by Radian »

MissileCrisis wrote:The slope/type simplify determines the transfer function of the input signal AKA it changes how much voltage at what frequency gets to the subwoofer, the SPL that you experience at that frequency is entirely dependent on the efficiency at that frequency.
Careful of the context here. Your not using the definition of the "transfer function" correctly. :ugeek:
MissileCrisis wrote:In order to get the flattest response down to your cutoff you would need to add eq to boost these frequencies with as high a q as possible by about 3-4 dB to counteract the downward slope. Mind you since you're using 48 dB/octave you will not be getting much if anything below the cutoff frequency but at least you'll be flat down to said cutoff, say 40 hz.
I highly recommend you try that out sometime for yourself and listen to how that actually sounds. You can even use headphones and a VST plug-in with a media player to give it a shot.

With the filter impulse response AFU, it's bad sound, before it even makes it to the subs. :idea:

Hi Q filters look good on paper (in the freq domain), but don't particularly cut the mustard well where bass is concerned.
Good food, good people, good times.

4 - AT
1 - TT
1 - THT Slim
2 - SLA Pro 4x6 Alphalite

Grant Bunter
Posts: 6915
Joined: Mon Nov 28, 2011 11:12 am
Location: Ilfracombe Queensland Australia
Contact:

Re: Is there a way to quantify extension?

#26 Post by Grant Bunter »

This is where it's gets very confusing for moi!

Lets try to sort some things out...

I don't use PEQ, I use 31 band GEQ's, I'm not altering Q, unless the unit is changing Q itself .

This is what confuses me:
In a T60 v T48 chart it is stated that the T48 has the goods over the T60 above 38Hz.
Here's the rub. The charts are raw data charts, and don't account for HP in any way.
Since a single T48 should be HP at 40hz, with "as steep a slope as possible" ie in my case LR48, doesn't that mean that there isn't a hope in hell you get 38Hz at the same voltage as 40Hz even though the SPL chart is relatively flat there?
In fact, all the SPL charts are raw data charts AFAIK and as such don't accurately represent usage in everyday life.

That's fine for the T48, but T39's aren't the same. A single T39 has a HP of 45Hz, so presumably, with the same LR48 slope you won't even exhibit any response at all at 40Hz.

Then we move to herds. Where we are told that, as a unique attribute to horn subs, multiples go flatter and lower with extension In the case of the T39, in a group of 4, that's 40Hz for me, + extension. If I go to 8 cabs, then I would have thought that I would get (even) more extension. This has been settled though with Bill's response, which I take as a product of the law of diminishing returns lol.

Really, while it is interesting to explore, is there any theoretical way to get a simple number?

2 posts in the thread so far that have actually tried to quantify the value, and vary between 2-3Hz to 5Hz. The variation in those two figures, at the bottom end, are quite significant.
2-3Hz can basically be ignored, 5Hz is in the same order as changing HP when going from 1 cab to multiples.
And with no direspect meant or implied to either poster, there is no answer about how those figures were arrived at, as yet.

I don't know if there is a simple answer, or if I'm quite not putting this in a way that's well understood.

ps MC, thanks for posting up those charts, yes, it tells me I actually was producing a reasonable percentage of the LF content...
Built:
DR 250: x 2 melded array, 2x CD horn, March 2012 plans.
T39's: 4 x 20" KL3010LF , 2 x 28" 3012LF.
WH8: x 6 with melded array wired series/parallel.
Bunter's Audio and Lighting "like"s would be most appreciated...

Michael Murphy
Posts: 343
Joined: Sun Dec 30, 2012 2:17 pm
Location: Trinidad

Re: Is there a way to quantify extension?

#27 Post by Michael Murphy »

Grant I think Im getting your question. As we know the different cabs have different corner frequency's, with the T39 a single cab is 45hz. One of the parameters to input in the raw data would be the cabs corner freq, a method of finding this is to take the speed of sound in feet/sec multiply it by 12 to give you inches per/second (13548) then divide this by the desired freq say 45hz (301) then by 4 for 1/4 wave length (75). The horn path should be approx. 75". So that's what 1 T39 should be able to do, if we shoot for 40hz the path ends up being 85". I believe coupling the T39's simulates this.

Part 2 of this looks like it may hold some answers
http://www.fohonline.com/home/73-speaki ... horns.html

but I have not gotten there as yet

MissileCrisis
Posts: 1128
Joined: Sat Dec 15, 2012 1:30 pm
Location: Walnut Creek, California

Re: Is there a way to quantify extension?

#28 Post by MissileCrisis »

@grant
My pleasure.
Most of the time I surprise myself even with the frequency content of songs being higher than one would expect.
Built
2 x TLAH
1 Titan 39 (lab 12, 15 inch)
1 Tuba 60 (lab 12, 20 inch)
SLA Center
2 x Mini MLTP subwoofers
1 x 212 TH (custom)
2 x Mini-Karlsonators
Owned, but not built
2 x Jack 12 (2512, melded)

User avatar
Radian
Posts: 2028
Joined: Tue Jul 29, 2008 12:56 pm
Location: Tucson, AZ

Re: Is there a way to quantify extension?

#29 Post by Radian »

Grant Bunter wrote:Here's the rub. The charts are raw data charts, and don't account for HP in any way.
...Correct, but in the plans, the recommended HP filter set points are based upon running the cabs flat out, wide open.

Prudence dictates that the operator should be running inverse square law math to insure a conservative number of cabs are employed to begin with.
Grant Bunter wrote:Since a single T48 should be HP at 40hz, with "as steep a slope as possible"
:conf: Who ever said that? It should read "as steep as required." 12-18 dB / oct more often than not, will suffice. If in doubt, play around with Hornresp and take a look at the excursion plots using a generic front-loaded horn model.

A high-pass effectively neuters extension...so it simply doesn't pay to build a pile of cabs without allowing them to deliver their maximum low freq potential. That's the whole point of the design. Get too aggressive with the filtering and you'd have been better off building smaller subs to begin with. In short, don't use 48 dB /oct. :cop:
Grant Bunter wrote:Really, while it is interesting to explore, is there any theoretical way to get a simple number?
Yes, use a woofer tester to take an impedance sweep of one cab. Note the freq of the spike near Fc. Then assemble the entire stack, wire them all in parallel, and sweep again. Once more, take note of the freq of the spike near Fc. It'll have shuffled down by a few Hz. That'll be how much extension you've picked up by adding cabs. Results will vary depending on model, width, and number of cabs used.
Good food, good people, good times.

4 - AT
1 - TT
1 - THT Slim
2 - SLA Pro 4x6 Alphalite

User avatar
Bill Fitzmaurice
Site Admin
Posts: 28619
Joined: Tue May 02, 2006 5:59 pm

Re: Is there a way to quantify extension?

#30 Post by Bill Fitzmaurice »

Radian wrote:
Grant Bunter wrote:Since a single T48 should be HP at 40hz, with "as steep a slope as possible"
:conf: Who ever said that?
I will. Excursion jumps rapidly below the horn Fc, so to allow maximum voltage input without running the risk of over-excursion the higher the slope the better, though I wouldn't jump through hoops to get more than 24dB/octave.
A high-pass effectively neuters extension...
That's exactly what you want to do. If you actually need to go lower than the recommended high pass filter frequency for the subs that you have you need different subs. That doesn't mean that you can't go with a lower frequency high pass, but if you do you have to back off on the voltage as well. The T48, for instance, will see excursion doubled at 28Hz versus 35Hz, so if you decide to high pass a stack of T48 at 28Hz you need to cut the voltage limit by half to be safe. By the same token a 12dB/octave HP filter won't sufficiently reduce the voltage swing at 28Hz compared to 35Hz to protect the driver.

Post Reply