sensitivity or frequency response for cabs?

[rpsands]

For what it's worth I would call the Neox112T 94db LF, 99db MF/HF, or something along those lines, if I were to use a generic measurement, and specify an approximate displacement limitation of 150-200w (realistically 75w @ 42hz, bearing in mind the frequendy distribution of bass guitar harmonics).

I don't think comparing subwoofers to full range bass cabs is a good call here; a more realistic comparison would be the Jack 112, I think.

The way most Subwoofers I've seen are categorized is they tell you what the average sensitivity is over the band (from the high pass to the low pass of the system). So the average changes depending on where you high pass and where you low pass.

If you look at the numbers for high end subs this is what you get; a sensitivity and maxspl depending on various common high passes. Say you move the HPF from 30 to 45hz, there's a big gain in sensitivity and power handling.

 

[JimmyM]

btw, this discussion has cured me completely of ever wanting to be a spec nerd.

 

[billfitzmaurice]

So Bill, it looks like you do measurements using 1/2-space as well. How would you spec. these speakers in the commercial world? How would you characterize the LF roll-off and sensitivity given the great differences between your mid rise and LF roll-off on the T-36? I see a 15dB difference, what would be the most fair characterization for assigning realistic (or valuable to the customer) numerical values to such a response curve?

http://billfitzmaurice.info/forum/viewtopic.php?f=12&t=11124&sid=24e381707c1c0e47f69d5c47d6e25b6c

The first chart shows that a 20" wide T60 has a sensitivity of 104dB, -3dB at 65Hz, -10dB at 28Hz. That's according to the standards set forth by the IEC/AES, and as they are they ruling bodies they make the rules. And one rule with regard to direct radiating alignments is that SPL above the range of pure pistonic function do not count and may not be averaged in to the result. In the case of the 2512 example as noted by RP the sensitivity is 94dB, and anything higher than that is covered by the +/-dB notation. Since the 2512 will run as high as 106dB at 2.5kHz it could properly be rated as having an SPL of 94dB +/-12dB from 28 to 5.5kHz. Or 94dB +12dB/-3dB from 40 to 4.5kHz. And then of course there will be those who will rate the sensitivity as 106dB with a range of 28 to 5.5kHz, and leave out all of the defining details. :scowl:

As for the notion that a speaker for bass should be subject to a different standard than one for PA or HiFi or whatever, that's the same as saying you should rate the gas mileage for a car based on a 5280 foot mile, for a truck for a 4000 foot mile, and a motorcycle for an 8000 foot mile. It's just plain wrong and only serves to confuse and deceive the consumer, be it intentional or not.

 

[agedhorse]

To my knowledge Winisd models in half space; in that case, the calculated and modeled SPL @ 50hz is about 94db.

I think your transfer function is accurate, but your assumed sensitivity is what's wrong.

I looked at Winisd and it does do a 1/2-space default and modeling with my data in Winsid 1/2-space shows 96.7dB/1W/1M for the low frequency component, and averaging the spectrum based on my data above 200Hz - 5kHz I get just about 99.1dB/1W/1M so again, it seems to me to be pretty darn close. I show an F3 of 46.8Hz and a -10dB point of 34Hz. Don't forget that we are using a cusom driver, taking advantage of optimizing the driver and box for the smaller size.

The transfer curve was very close between simulations too so that adds confidence to the results, or at least the math.

Note that simulations do not provide any meaningful data regarding information above say 200-300Hz, there are major variables that come into play that govern things above this reigon.

 

[rpsands]

Hmm, are you using the older Winisd or Winisd pro alpha?

Modeling closer to what the Neox112T actually is (2.3 internal cubic feet, tuning approximately 40hz) I get 95.1db from 55 to 200hz +/-3db, with an F3 of 55hz, F6 44.5hz F10 of 35.5hz.

If your tuning is higher, say 45hz, you peak out at 95.6db @ 90hz...however, that would be a pretty boomy cab, and I don't recall the Neox112T having that much of a midbass hump.

 

[agedhorse]

The first chart shows that a 20" wide T60 has a sensitivity of 104dB, -3dB at 65Hz, -10dB at 28Hz. That's according to the standards set forth by the IEC/AES, and as they are they ruling bodies they make the rules. And one rule with regard to direct radiating alignments is that SPL above the range of pure pistonic function do not count and may not be averaged in to the result. In the case of the 2512 example as noted by RP the sensitivity is 94dB, and anything higher than that is covered by the +/-dB notation. Since the 2512 will run as high as 106dB at 2.5kHz it could properly be rated as having an SPL of 94dB +/-12dB from 28 to 5.5kHz. Or 94dB +12dB/-3dB from 40 to 4.5kHz. And then of course there will be those who will rate the sensitivity as 106dB with a range of 28 to 5.5kHz, and leave out all of the defining details. :scowl:

Generally, the "frequency range" of a subwoofer IN THE PRO AUDIO INDUSTRY is considered to be from the -3dB LF point to between 100Hz and 125Hz, so the efficiency of a speaker that is marketed as a pro audio subwoofer (or compared with one) should be calculated within this bandwidth. So IF I were rating the sensitivity of that box as a subwoofer with an applied bandwidth of 100Hz, the average SPL would be somewhere around 100.5dB/1W/1M. If the applied bandwidth was 125Hz then the average sensitivity would be about 101.5dB. Averaging gives a good approximation of what a reasonable user could expect from the box over the bandwidth of it's intended use.

As for the notion that a speaker for bass should be subject to a different standard than one for PA or HiFi or whatever, that's the same as saying you should rate the gas mileage for a car based on a 5280 foot mile, for a truck for a 4000 foot mile, and a motorcycle for an 8000 foot mile. It's just plain wrong and only serves to confuse and deceive the consumer, be it intentional or not.

You are inaccurately twisting my words...

Using your vehicle example, If I am buying a motorcycle, I want to know what the performance is under typical motorcycle conditions. For example, the performance of a motorcycle is going to be specified carrying (or towing) a motorcycle load BUT the performance of a truck is going to be specified carrying (or towing) a truck type load. Each vehicle's weight loads should be different because they apply to completely different markets. I want to know the performance of a vehicle based on it's intended application, not some standard weight that can not apply to all vehilces regardless of market or class.

Using bass amplification as an example, the piston reigon does not matter because there is much desireable tone (tone that is sought after by bass players) in the non-linear reigons that create the characteristic sound, feel and playability of both speakers AND amplifiers. Some of this is achieved specifically by inserting (carefully) non-linear terms or components into the design of the speaker. This is absolutely standard operating procedure for guitar speakers too. This is NOT something that is sought after for a pro audio subwoofer, therefore I would agree that for the pro audio market, that kind of non-linearity is something that is undesireable and the measurements should reflect the differences in the intended markets.

 

[agedhorse]

Hmm, are you using the older Winisd or Winisd pro alpha?

Modeling closer to what the Neox112T actually is (2.3 internal cubic feet, tuning approximately 40hz) I get 95.1db from 55 to 200hz +/-3db, with an F3 of 55hz, F6 44.5hz F10 of 35.5hz.

If your tuning is higher, say 45hz, you peak out at 95.6db @ 90hz...however, that would be a pretty boomy cab, and I don't recall the Neox112T having that much of a midbass hump.

I used the newer version (alpha), but note that the parameters of our driver are different than the stock Eminence DL you are using. Our parameters are propriatary and I can not release them. There is considerable voicing modification done with the cone and dustcap specifications, which of course factor into the TS parameter target design.

Just food for thought, how do you think sensitivity might be altered if the geometry of the voice coil was altered? There are things that a designer can do (if they have access to designing their own drivers) that don't make sense in an off the shelf product.

 

[rpsands]

I am not an expert on any of this stuff, but I would think if you could squeak out more sensitivity <100hz without costly sacrifices up top or increasing Vas or people would be doing it. The best 12" drivers on the market I've seen have a LF sensitivity of around 94-95db.

Can you tell us what the calculated sensitivity of your driver is in Winisd (or similar)? By calculated I mean you plug in all the speaker parameters and let it calculate the SPL for you.

Here's what it looks like in Winisd Pro for me (with the 3012LF and 2512 II)

 

[agedhorse]

95.72dB

Some manufacturers are indeed doing this, especially in their propriatary boxes. Think EAW, JBL etc.

 

[billfitzmaurice]

Just food for thought, how do you think sensitivity might be altered if the geometry of the voice coil was altered?

I know how it would be altered. Qes and Qts will go down, sensitivity will go up. F3 also will go up. You can have higher sensitivity with higher F3, or you can have lower sensitivity with lower F3. What you can't have is higher sensitivity without a corresponding raising of F3.

There are things that a designer can do (if they have access to designing their own drivers) that don't make sense in an off the shelf product.

I am a designer, and I have access to designing my own drivers. I don't want to be rude, but you keep bringing up this OEM driver stuff and what you're claiming for them just isn't true.

 

[rpsands]

Nifty. So 98db is not 100% accurate from 50-200hz but it's pretty reasonable all things considered; seems you should be chilling around 97db average under 200hz. Assuming you factor in the big boost in the breakup regions certainly not unreasonable to claim 98db.

Are you sure about the F3 of 47hz? I get closer to 55hz with a Deltalite in a box that size. Not sure all of that could be accounted for by subtle driver differences.

It could be that my guesses at box volume and tuning are off of course, which would adjust the numbers slightly.

 

[agedhorse]

Nifty. So 98db is not 100% accurate from 50-200hz but it's pretty reasonable all things considered; seems you should be chilling around 97db average under 200hz. Assuming you factor in the big boost in the breakup regions certainly not unreasonable to claim 98db.

Are you sure about the F3 of 47hz? I get closer to 55hz with a Deltalite in a box that size. Not sure all of that could be accounted for by subtle driver differences.

It could be that my guesses at box volume and tuning are off of course, which would adjust the numbers slightly.

The average sensitivity from F3 to 5kHz is certainly better than 98dB. There is some intentional midrange and low midrange voicing that needs to be factored in as well, but nothing like the big peak in the stock DeltaLite (that some people really like I might add). Also, the midrange energy bandwidth is wider so that factors into the weighting average numerically.

Try increasing the box volume up a little bit, that's effectively what I did by altering the driver parameters. There's a limit to how much can be done with this method, of course.

 

[agedhorse]

I know how it would be altered. Qes and Qts will go down, sensitivity will go up. F3 also will go up. You can have higher sensitivity with higher F3, or you can have lower sensitivity with lower F3. What you can't have is higher sensitivity without a corresponding raising of F3.

I should have added... "and now with the increased sensitivity other parameters will have impact."

I am a designer, and I have access to designing my own drivers. I don't want to be rude, but you keep bringing up this OEM driver stuff and what you're claiming for them just isn't true.

Then you understand that calculations for the stock driver do not apply to calculations for our driver.

 

[rpsands]

The average sensitivity from F3 to 5kHz is certainly better than 98dB. There is some intentional midrange and low midrange voicing that needs to be factored in as well, but nothing like the big peak in the stock DeltaLite (that some people really like I might add). Also, the midrange energy bandwidth is wider so that factors into the weighting average numerically.

Try increasing the box volume up a little bit, that's effectively what I did by altering the driver parameters. There's a limit to how much can be done with this method, of course.

I appreciate you taking the time to chat this out with me!

 

[agedhorse]

I appreciate you taking the time to chat this out with me!

No problem. I hope some of the folks here who thought that a single number specification can tell the whole story or even the "Reader's Digest" version of the story now have a better appreciation for the thrust of my initial comments about how multi-dimensional it is to describe complete and meaningful performance of a speaker. There's a lot of inter-related information that depends on interpretation specific to different conditions. Each market is different and manufacturers need to describe their products in a way that makes practical sense to each market without presenting too much information (which Roger busts me for ALL the time, and maybe rightly so) that it becomes confusing rather than helpful.

Incidently, the same excercise applies to bass amplifiers, pickups, instruments etc. Everything in the bass (and guitar) world have individual voicing characteristics and to pretend otherwise is not realistic.

 

[agedhorse]

Here's a measured plot carrying on from my 20-200Hz decade simulation plot to the next decade (200Hz-2kHz) measured plot, which picks up within a fraction of a dB at the 200Hz transition, meaning that the LF simulation and the measured MF methods agree remarkably well.

Calculating the average and it looks to be about 97.7dB which also agrees with the averages that I came up with when we published the specs. There is a little rise between 2kHz and 5hHz where I stopped calculating, this easiliy pushes the average over 98dB so our spec's. indeed appear to be reasonable.

Hope this helps you guys understand where I am coming from and why, even amongst folks that are pretty familiar with this stuff, that there are plenty of areas of confusion that can lead to misunderstandings.

 

[agedhorse]

I can't edit my posts for some reason (TB hangs), but it should be noted that between 100 and 200Hz, the measurement/calculation method is based of different conditions so the data is accurate in the simulation for 20-200Hz and inaccurate in the measurement due to the way the driver is loaded. This is why I break the system up into two discrete components, the LF and MF conditions and if the two don't match within a dB or so then there is something wrong somewhere in the system of equations or the measurement. It's the engineer's check and balance for good data.

 

[billfitzmaurice]

Looking back on this thread I haven't seen so much baloney in one place since an Oscar Meyer truck rolled over on I-95.

Awhile back someone noted that Aguilar's sensitivity and response specs were out of line. Dave at Aguilar could have engaged in a prolonged campaign of stonewalling, with the intent of confusing the issue and our members with minutae. He didn't. He owned up to the errors present in their advertising and fixed them. This was his response to the situation in an email to a member here:


"Hi Mike,

The driver hasn't changed. When I realized that someone had "ballparked" our sensitivity specs, I changed them to reflect the speaker data. One could make the case for this or that spec; but we try not to roll that way. As you know the DB 112 will sound different; but the sensitivity is the same as the new and the older GS 112 and 112nt.

Please let me know if you have any other questions, or if you need assistance at any time.

Best Regards,

Dave Boonshoft
CEO"

 

[jnewmark]

Looking back on this thread I haven't seen so much baloney in one place since an Oscar Meyer truck rolled over on I-95.

Thank God someone said this; I thought I was the only one.

 

[rpsands]

Here's a measured plot carrying on from my 20-200Hz decade simulation plot to the next decade (200Hz-2kHz) measured plot, which picks up within a fraction of a dB at the 200Hz transition, meaning that the LF simulation and the measured MF methods agree remarkably well.

Calculating the average and it looks to be about 97.7dB which also agrees with the averages that I came up with when we published the specs. There is a little rise between 2kHz and 5hHz where I stopped calculating, this easiliy pushes the average over 98dB so our spec's. indeed appear to be reasonable.

Hope this helps you guys understand where I am coming from and why, even amongst folks that are pretty familiar with this stuff, that there are plenty of areas of confusion that can lead to misunderstandings.

Hmm, based on that chart I'd ball park you're looking at 96db up to almost 2khz. It's a little questionable in my opinion to average 2khz+ data with 50-2khz data, and say that 103b @ 2khz averaging with 95db @ 400hz makes a 98db average cabinet.

Not a huge deal I guess. I would like to see the measurement from 20 to 100hz, though, if you have it. I'm pretty sure the 48hz figure is closer to the F6.

My thinking here is based on some anomalies in the specs though; if you average it out to 98db and then say the -3db point is where it hits 95db, your F3 is going to be really really high. If you say the driver's "real" sensitivity (pistonic) is 95.72db then look for the -3db point (92.72db) it's going to be another thing.

Where it sounds like the confusing is happening is you're saying the -3db point is @ -92.72db, but 98db is your average sensitivity. Kinda makes it confusing (hence my suggestion of a separate LF and HF sensitivity number).