ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 ... 4 5 [6] 7 8   Go Down

Author Topic: d&b B22  (Read 19960 times)

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 639
  • Oslo, Norway.
Re: d&b B22
« Reply #50 on: November 17, 2016, 04:28:34 pm »

Mircea, that is the exact system i am talking about. We were discussing a lot about it with the EAW engineer that was with us on site. I even had the opportunity to peek inside the OTTO. So i am not at all surprised with the Helge comment :-)

Six of those Ottos ended up in Norway, I've been fortunate enough to play with them and the assosiated Annas, great stuff.

The guys who owns it just did a ballroom tour covering 600 people with one Otto and two Annas a side, that's a lot of people compared to the required truck space.
Logged

radulescu_paul_mircea

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 75
Re: d&b B22
« Reply #51 on: November 17, 2016, 05:17:40 pm »

I played and listened to stacks with  up to 12 subs a side or more than 40 subs in front of the stage and I can tell you the Otto subs are powerful and clean but not more than a dual 18 like the WS218 X from Martin or a J sub from D&B . I even heard them combined ;)

Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8696
  • Atlanta GA
Re: d&b B22
« Reply #52 on: November 18, 2016, 07:07:47 am »

Ivan,

The 9dB issue is from comparing a 1m measurement to a 2 meter one. (6dB there)
The other 3dB is the difference from reporting a "peak" SPL from the waveform and an "RMS" SPL. Note that the speaker output or waveform has not changed. Only the method of reporting the output. This is explained in the CEA-2010 section below. Also note that a lot of places or manufacturers are reporting a 1m half-space "peak" output. That's not what I do at Data-Bass I use a much more conservative and IMO more realistic 2 meter half-space (1m full-space) SPL reported from the "rms" method of analysis of the waveform.


Yes, the standard is a 1M SPEC-but not measurement distance.  Back calculations are often used.

HOWEVER, in many cases (especially with subs) actually measuring at 1 or 2M will NOT give a number that is useful in designing sound systems.

When designing we need to know how loud it will be at certain distances (NOT at 1 or 2 M)

So Danley (and others) measure at a further distance-and then "back calculate" to get the 1M equivalent.

Danley uses a 10M distance and 28.3V.  So a 20 addition in voltage offsets the 20 dB loss in distance.

If you measure at 1M you will not get our spec numbers in many cases (on our tops or subs)

But if you use the 1M number and figure out the level at say 40M, you will get a number that will be right at 40M-or any other distance that is further away than a couple of meters.

In some cases, the actual measurement at 1M is 6dB or lower than the spec.  But at a distance-the numbers are correct.

This all has to do with the size of the device, the acoustic origin of the sound and the measurement distance.

You need to get far enough away so that the numbers will be repeatable at different distances.

To me, the numbers in a spec sheet need to be useful to a designer for figuring out if a particular loudspeaker will cover the intended area-at the needed SPL and have a freq response that is proper for the intended application.

Having specs that include "3dB for the crest factor of a sine wave" are misleading-at least in my opinion.

That is not something that can be measured with normal test gear on site and will not be realized in any way by the end user.  Yet can lead to designing a system that not quite loud enough for the intended application.

To me- it falls into the "Peak SPL" that is based on a single freq that sticks out in the freq response.  People "think" the speaker can produce that over the whole bandwidth, but in reality it is only a very narrow freq that must be notched out in order to sound good.

In some cases, this results in a number that 10dB greater than what you can actually achieve out of the cabinet.  But some people like to believe this "simple number" and will argue that XYZ speaker is louder than ABC speaker because it had a 10dB greater peak SPL.  Truly sad in my opinion.

But that is just my opinion.  I know others feel different.
« Last Edit: November 18, 2016, 07:14:53 am by Ivan Beaver »
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Marjan Milosevic

  • Full Member
  • ***
  • Offline Offline
  • Posts: 229
    • MM-acoustics
Re: d&b B22
« Reply #53 on: November 18, 2016, 10:50:28 am »

I feel exactly the same.
It is really sad looking at the spec by most of the top dogs and realizing they are extremely misleading.
It then force others to jump in the same wagon or they will have very hard time presenting a comparable product because on paper it will look inferior.
People love looking at big numbers. And that fact is used by marketing departments.

Josh Ricci

  • Newbie
  • *
  • Offline Offline
  • Posts: 27
    • Data-Bass
Re: d&b B22
« Reply #54 on: November 18, 2016, 03:04:16 pm »

Hello Ivan and Marjan,

I fully agree with both of your last posts. The creative use of numbers & specs is alive and well. It's hard to blame the manufacturers either. Consumers love big numbers. The bigger the better. More SPL YEAH! More watts! If you start losing sales to other companies who are rating their products in a completely different manner that isn't comparable in the least, but less educated buyers do not know any better, that's not a good position to be in. You could attempt to start an education initiative to trickle down knowledge to the wider market base and trust that the products eventually gain a following based on use in the field (Difficult and long term) or you can simply change some spec sheets, potentially gain some sales immediately and go on with the rest of your probably busy schedule (cheaper and quicker). It's not hard to see how this happens.

BTW Ivan,

Sorry about the huge book of a post but someone had brought up Data-Bass so it was easier to copy/paste than retyping all of that.

My measurements are usually taken at 1,2,4 and 10m for each system so it is known how the proximity to the cab under test is affecting the response. In general the large cabinets lose a bit of apparent bass response at 10m. Not a huge amount but a couple of dB isn't uncommon for a larger cab. Also the response tends to smooth out a little at distance. The only time that I have seen that the 10m 10x voltage measurement is higher than the 1m measurement is when the cabinet radiates from multiple surfaces. The 10m sensitivity measurement is made with 10X voltage versus the 1m, etc. I cannot measure much further than 10m because background noise becomes a serious issue as does the proximity of large buildings a couple hundred feet back on the very lowest frequencies tested. It would be a rare place indeed that has the space and noise floor needed for bass measurements at even greater than 10m distances.

The SPL numbers are reported as what is functionally 1m full space (2m half space equivalency.) They are not measured at 1m other than the 1m sensitivity measurement in most cases. The big, powerful cabs end up needing measured at 4 meters in most cases to ensure headroom in the mic/preamp anyway. A few have been at even greater distances. It all gets normalized back to the 1m full space SPL for exactly the reasons you pointed out in your post. For comparison and ease of calculations sake. Also the largest systems I have measured to date are single cabs with not much more than 2000sq in of frontal area. Big but not out of the ordinary for large subwoofers and small relatively compared to a block of multiple subs. A 24x48 frontal area such as a common 2x18" BR is  1152sq in. There is a trend seen as the microphone distance is increased, when comparing the measurements of single cabs in that size range , but generally there is little difference seen comparing a 4m or in some cases a 2m measurement and a 10m one. Most of the "boost" seen at 1m is gone once beyond 2 or 3m distance outdoors. This is with moderate sized, single cabs though. A block of 8 cabs would be a different story and need greater measurement distances. 

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8696
  • Atlanta GA
Re: d&b B22
« Reply #55 on: November 18, 2016, 07:20:08 pm »


My measurements are usually taken at 1,2,4 and 10m for each system so it is known how the proximity to the cab under test is affecting the response. In general the large cabinets lose a bit of apparent bass response at 10m. Not a huge amount but a couple of dB isn't uncommon for a larger cab. Also the response tends to smooth out a little at distance.
I would argue that it is NOT that they lose output at 10M, but RATHER that the measurements made up close include false bass INCREASES. 

This is particularly present in large cabinets.

That is EXACTLY why we don't measure up close.  It gives a false sense of what the cabinet can produce at far distances.

So while the numbers would actually be what is measured- it CANNOT be translated into what would be expected at further distances.

So it comes down to whether you are selling "numbers" or something that will actually perform as the specs say.

People have different opinions on this-depending on whether you are selling or buying------
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Art Welter

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1473
Re: d&b B22
« Reply #56 on: November 22, 2016, 09:57:18 pm »

1)measurements made up close include false bass INCREASES. 
This is particularly present in large cabinets.
2)We measure using a swept sine wave via TEF.  For subs in particular we measure at a distance of 10M with 28.3V input.  So that comes back to a 2.83V 1M measurement.
3)The 6dB between continuous and peak is simply the increase in power to a maximum of the driver-with a 6dB peak based on the power capacity of the driver.
This is assuming no power compression-since that takes time.
The 137 will hold for a certain period of time-based on heating.  Of course the actual levels without power compression will depend on the dynamic range of the signal.
4)We do no measure distortion at full output.  But a horn cabinet will typically be less distortion than a front loaded/ported cabinet simply because the driver is not moving as far.
Dear Ivan,

1) Your statement is inconsistent with your measurement guru's dictum.
I'll cite it if you ask politely, but don't want to embarrass you with it, we all have our "drunk posts" that we have to walk down the "hall of shame" while holding the trophy with some appendage of our shamed existence held prominently for all to see.
Having far too many "shame walk" atrophies evident in my copious "drunk post" historical evidence, which progressively disappear from the inter-webs at the approximate rate of  their initial propagation, I regret any "collateral damage" inflicted in my reply to your dissertation.
2) The "power capacity" of the drivers DSL uses are simply reprints of the AES rating of the driver taken in free air.
When the driver is loaded in a "tapped horn", the impedance of the driver is far lower than in free air.
Since the AES rating is based on the nominal impedance, not your DSL TH impedance, your power specification is inflated by a rather large factor.
As an example, let's use the DSL TH-118 loaded with a "4 ohm" B&C 18SW115 rated for 1700 Watts.
To drive a nominal 4 ohm load to 1700 W requires 82.5 volts, so we assume that is the voltage used for the AES test, even though the driver's average free air impedance is far higher- the driver is actually dissipating a fraction of the "watts" specified.
Put the same "4 ohm" B&C 18SW115-4 in your DSL TH-118 and apply the same "continuous" 82.5 volts at Fb, where impedance is as low as the 3.3 Re DCR of the driver and in about twice the time it takes Porky Pig to say " ba ba be  hey hey hey, presto",  the smell of burnt adhesive will start to be evident near the cone.
Leave that sine wave tone fixed at Fb another 30 seconds or so and the smell will be enough to drive you out of the room unless you have an air evacuation fan going full tilt for a cold December half hour (true story). Wish I had more than a 1500 Watt heater in the shop..
Now take that "warmed up" driver, and apply halfway between the "continuous" 1700 Watts (those Watts that used to be redundantly called Watts RMS) and the peak rating of 6800 Watts, "3400 Watts" requiring 106 volts.
I don't presently own an amp that can put out  3400 Watts continuously, but I guarantee the "poor little" 115mm double sided double wound 18SW115 voice coil won't last long even if you use 106 volts (RMS, not peak) pink noise rather than a sine wave at Fb.
Yes, cooling is better when the cone is violently flopping around, but we can't determine the musical proclivities of end users, EDM "drones" may have less crest factor than the "terrible" 3dB of a sine wave.

In weeks of abusive testing, and years of shows ranging from jazz to hip hop to EDM, I never had any problem with the B&C 18SW115-4 loaded in my Welter Systems Keystone Sub design, also "sporting" the same 3.3 ohm impedance minima at Fb the DSL TH-118 has.
The fellow I sold the original pair of Keystone Subs to managed to burn one up in the first week.
My guess is he accidentally panned kick and/or bass hard left, leaving one sub to do +6 dB more "heavy lifting" to achieve the level he became used to on the first several gigs.
Bass is addictive, I see +5, +10, and +20 addicts ruining our collective lives on an ever more frequent "bass"is.
A single mono bridged Crest CA 9 driving one Keystone sub can't put out more than 80 volts while drawing 37.8 amperes from a 120 volt line.
37.8 amperes line voltage to a shade under 100 volts, when directly connected to a mains power transformer with about 40 feet of 4AWG to the mains panel, and 100 feet of 10AWG to the amplifier.
If the Crest CA 9 were run anywhere near that 80 volt output level the 20 amp circuit breaker the fellow's amp rack with two other amps of equal power draw on the same circuit would have popped, but the breaker didn't, the coil lost it's immortal soul.

The "take away" from all this is your specifications are not "real world", so you constantly have to inform your DSL clients to limit the speaker to some small  fraction of the AES rating your DSL specs are based on, or they will let out the "magic smoke" and stink up the area surrounding the DSL cabinet.

This is the choice made when DSL wrote specifications based on fictional "power" ratings, like so many companies play the grid iron playing field here.
Yet you have  always liked to play from the top of the hill, showering us with  "trickle down" theories like a "golden shower" of truth on the 49% of the 99% inhabiting the USA;^).

2) I like to firmly bolt my drivers down, how far is "too far" for a driver to move?
Thanks for finally updating your TH-118 specs to the 18SW115 AES ratings rather than the 18NLW9600-4 that is still pictured and your tips six years ago about why the B&C is the better choice :^).
Sometimes I wonder if M.H's influence has shifted you, but I know the "force" is strong deep within you, Ivan- do not succumb to the "Dark Side".

Jokes aside, the reason the 15.2 gross cubic foot Keystone Sub "tapped horn" also loaded with the B&C 18SW115-4 delivers 6 dB more average SPL in the 40-100 Hz pass-band than the same driver in a 9.24 gross cubic foot "bass reflex" enclosure with the same low corner frequency is because the driver is forced into "doing more work" for a given excursion.
Work is hard, some "force" always will pay for any work done in the omni-spherical dimension we inhabit.
Yes, the Keystone Sub "horn cabinet" delivers more SPL for a given input voltage, but the TH will also have more distortion at that voltage than a BR.
Recently, while reviewing TH distortion and SPL compared to BR, while also comparing impedance curves, found the Keystone had "compression" at the upper end of the pass band at "only" 77.5 volts of sine wave input, while the BR was the opposite, due to "port compression".
The impedance curves made obvious "what went on".
Impedance is kind of like a variable frequency chastity belt, the upper reduction in the TH SPL was due to it's upper pass band low impedance dips "sucking, or demanding" power, while the BR has a steadily rising impedance "spurning, or rejecting" the Scottish Watt's "power".
The more power the voice coil "sucks" surrounding the pole piece, the hotter it becomes, which makes it's impedance rise, protecting it from the very "power" that heated it up, a cosmic entropic conversion of energy, - sex, drugs and rock and roll are not required to witness the conversion, but two of three are usually involved.
Since most of us audioholics test from "bottom" up, my sine wave tests were also started at the bottom of the pass band, that way you can only go "up".
The average increased excursion down low provides ample "heat pumping" (forced cooling), then proceeds to upper frequencies, which progressively pump less heat from the voice coil.
Ouch, too hot to touch- and that smell..
The upper pass band compression noted in the TH compared to the BR was due to the lower upper impedance causing voice coil heating, raising impedance, causing power compression- at 77.5 volts, after considerably less than 50% duty cycle over a few minutes time.
I know the duty cycle was less than 50% because I am "digitally challenged" and it took me more time to save each record than it did to let the SMAART screen settle down.

 Data in post # 12 here:

http://www.diyaudio.com/forums/subwoofers/185588-keystone-sub-using-18-15-12-inch-speakers.html

"Full power" should have been 82.5 volts, not 77.5 volts.
77.5 volts is 1500 Watts into a nominal 4 ohm load, the B&C 18SW115-4 is rated for 1700 Watts AES. I accidentally substituted another B&C driver's AES rating, and conducted the test at a fraction of a dB "too little" power.
As the Immortal Carroll Shelby said: "Too much power is Just Enough".
I bet Carroll burned up a lot more 427 Cobra engines than I have drivers, but his drivers got paid more than mine ;^)..

Cheers,

Art
Welter Systems, Inc.
« Last Edit: November 22, 2016, 10:10:44 pm by Art Welter »
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8696
  • Atlanta GA
Re: d&b B22
« Reply #57 on: November 23, 2016, 10:19:26 am »

Art, I understand what you are saying, but I wanted to add some comments for those "playing along at home".

The power handling tests are NOT flat (they rolloff on the high AND low freq), and they are for specific time periods.

They are also not sine waves at specific freq.

So as you go lower, there is less energy applied to the woofer.

Personally, I feel that these test curves do not represent modern music.  They are from years ago when we did not have as much bass present in music.

So if a flat response curve is used, the power capacity would go down.  And marketing depts would simply not allow that.  So we are stuck using the same test signals.

Yes, at certain freq, the impedance does get a little bit low, but at other freq it is quite a bit higher than the 4 ohm spec.

Yes we use the standard impedance numbers.  BELIEVE ME, if you put down anything other than 2-4-8-16 ohms the phones will start to ring off of the hook with people who cannot figure out what amp to use with a 6 ohm speaker.  YES it did happen back when the Sh50 was rated for 6 ohms.  We changed the "paper number" to 4 ohms and the phones stopped ringing with that question and sales went up-because it was a rating people could understand. :(

The reason I (and many others) often tell people to run long term limiters at 1/4 to 1/3rd continuous power has nothing to do with the rating of the driver or how it is rated.

And EVERYTHING to do with the typical usage for EDM or dance shows.  Very often pure sine waves are used.  And NO MANUFACTURER (that I am aware of) rates their speakers using sine waves anymore.  Years ago-yes, but not now.

Agreed sine wave will heat up a speaker VERY quickly-IF the freq is in a impedance low spot.

Also these shows are not your typical "rock show".  They go on for many hours and sometimes DAYS, with NO break in the sound.  So the driver just keep getting hotter and hotter. 

The specs are not for 24 hour operation.

This applies to ANY loudspeaker model.

So-as usual-there are a lot of things to consider when looking at the "simple number specs".
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Nathan Riddle

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1498
  • Niceville, FL
    • Nailed Productions
Re: d&b B22
« Reply #58 on: November 23, 2016, 10:50:46 am »

When I get rich I want to buy and test every speaker/sub/amp/dsp/mixer/light and use what y'all say/wish and remove the "I have/cant do this 'cuz marketingz" and just put everyone/everything on the same ballpark and test and test and test and then random people on the interwebs can have their cake/pie and eat it too.

In the meantime, I enjoy the conversations and eat my popcorn  ;)
Logged
I'm just a guy trying to do the next right thing.

This business is for people with too much energy for desk jobs and too much brain for labor jobs. - Scott Helmke

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 8696
  • Atlanta GA
Re: d&b B22
« Reply #59 on: November 23, 2016, 01:29:11 pm »

When I get rich I want to buy and test every speaker/sub/amp/dsp/mixer/light and use what y'all say/wish and remove the "I have/cant do this 'cuz marketingz" and just put everyone/everything on the same ballpark and test and test and test and then random people on the interwebs can have their cake/pie and eat it too.

In the meantime, I enjoy the conversations and eat my popcorn  ;)
Back before the turn of the century, I thought that if I won the lottery, I would start a REAL review magazine.

I would buy the products and then test and use them as I saw fit.

Then publish the results.  I would not accept any advertising, so would be free to publish what I wanted.

People would read it to get an unbiased opinion.

Of course, I need to win the lottery FIRST--------------
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!
Pages: 1 ... 4 5 [6] 7 8   Go Up
 


Page created in 0.058 seconds with 21 queries.