Anechoic data set for Focal Speakers

[highfigh]

Yes, knowing the anechoic response can be helpful but knowing your in room response is what matters most. Speakers like op's are from a reputable manufacturer known for making very nice, accurate speakers so I think in this case precise anechoic data isn't really necessary. I think it's safe to assume the speakers in question are going to be pretty linear and measure well.

I wouldn't assume anything. How are drivers chosen? Largely, by looking at the response graph and the specs, right? Are they not measured in some kind of space where they're not affected by the acoustics and if so, why not measure the speaker system in the same way, to ensure that the drivers are properly phase-aligned and the crossover is doing its job as designed? Designing a speaker system on paper or in a program isn't difficult, but knowing the drivers' response is necessary.

How many times have people responded with "ooh, ahh" when they see the frequency response from Dennis Murphy's crossovers and "What crap!" when they see the data from Wilson speakers?

 

[Pogre]

I wouldn't assume anything. How are drivers chosen? Largely, by looking at the response graph and the specs, right? Are they not measured in some kind of space where they're not affected by the acoustics and if so, why not measure the speaker system in the same way, to ensure that the drivers are properly phase-aligned and the crossover is doing its job as designed? Designing a speaker system on paper or in a program isn't difficult, but knowing the drivers' response is necessary.

For building a speaker yes, absolutely. To figure out response correction placement for a linear in room response with the speakers as a whole tho?

Am I missing a piece of the puzzle here or misunderstanding the goal of this post?

 

[highfigh]

For building a speaker yes, absolutely. To figure out response correction placement for a linear in room response with the speakers as a whole tho?

Am I missing a piece of the puzzle here or misunderstanding the goal of this post?

I think the OP is just trying to get the info provided by many other manufacturers, in order to make the right decision. I have seen speakers with bad response sell like crazy because they have the 'happy face' EQ built in, from the factory- boom/crash was popular in the late-'70s and early-'80s.

 

[Pogre]

I think the OP is just trying to get the info provided by many other manufacturers, in order to make the right decision. I have seen speakers with bad response sell like crazy because they have the 'happy face' EQ built in, from the factory- boom/crash was popular in the late-'70s and early-'80s.

Sure. It's nice to have the data. I have it for my speakers, but wouldn't find it necessary for dialing in my room response. The fact is, not many manufacturers even provide that kind of info and most folks manage without it.

 

[Pogre]

WRT 'anarchic', google 'Portland, Oregon'.

Definition of ANARCHIC

of, relating to, or advocating anarchy; likely to bring about anarchy; lacking order, regularity, or definiteness… See the full definition

www.merriam-webster.com

Yeah, I looked it up after making that post. Looks like auto correct betrayed @rjharle, lol

 

[Pogre]

I find it odd that they would write "the readings are done during Transducer Design in order to make sure they attain the design spec"- how can speakers be designed to work well in random rooms if they don't know the acoustical characteristics? The whole reason for using an anechoic room (or approximating anechoic) is to find the response from the speaker without any interaction with the room. Obviously, the in-room response needs to be found, but using anechoic measurement is still helpful for determining if the driver has any issues that might make them difficult to work with in a system.

Also, it's funny that Dr Toole would insist on knowing that data when Harman/JBL don't provide it with a lot of their own products!

 

[highfigh]

Also, it's funny that Dr Toole would insist on knowing that data when Harman/JBL don't provide it with a lot of their own products!

But they do provide it for their pro speakers because the people who usually use them need to know and the success of their projects demand correct choices. They don't have the luxury of returning things to Best Buy or Amazon if they aren't right and in commercial/industrial jobs, time IS money- someone's gonna pay for the time needed for removal, return & re-installation and nobody is going to volunteer for that. Consumers often try to make decisions based on real data but when they see all of it, their eyes glaze over and they just say "Let's get it done", listen to it and say "Well that didn't work" and then, the frustration and negative reviews start.

Having the data is good, knowing what to do with it is better.

 

[highfigh]

Yeah, I looked it up after making that post. Looks like auto correct betrayed @rjharle, lol

Bobcat Goldthwaite had a bit about product endorsements by people who wouldn't be expected to be paired with those products. One line had a picture of the Sex Pistols and was captioned "Anarchy! Anarchy in the UK! And Seagrams" with Johnny Rotten holding up a bottle of something similar to Zima.

 

[Pogre]

Having the data is good, knowing what to do with it is better.

Absolutely. However since Focal does not provide that data I would still assume a reasonably linear fr from such a reputable manufacturer. I can't imagine them being much more than +/- 3 dB anechoically through the midrange and higher frequencies. Much below that and the room is in charge anyway. It doesn't matter how they measure (within reason) anechoically below the Schroeder frequency of the room.

 

[Pogre]

FWIW Stereophile does have some detailed measurements of the Focal Aria 936.

Focal Aria 936 loudspeaker Measurements

Sidebar 4: Measurements

www.stereophile.com

 

[Pogre]

Some more detailed measurements of the Focal Aria 906 at ASR...

Focal Aria 906 Speaker Review

This is a review and detailed measurements of the Focal Aria 906 Bookshelf/stand-mount speaker. It was kindly purchased by a member and drop shipped to me. I think it costs US $2000 for a pair but I see it on Amazon including Prime shipping for half that at US $999. Better yet, they have a...

www.audiosciencereview.com

 

[Swerd]

I think there might be some confusion here between measurements for individual drivers and measurements for a finished loudspeaker where several drivers are mounted in a cabinet. I think you do understand this, but others reading this thread might not.

I wouldn't assume anything. How are drivers chosen? Largely, by looking at the response graph and the specs, right? Are they not measured in some kind of space where they're not affected by the acoustics and if so, why not measure the speaker system in the same way, to ensure that the drivers are properly phase-aligned and the crossover is doing its job as designed? Designing a speaker system on paper or in a program isn't difficult, but knowing the drivers' response is necessary.

When driver manufacturers publish a frequency response for a driver, they don't intend it to be mounted in only one size cabinet. But they must mount it somehow in order to measure it. They usually tell you enough so a loudspeaker designer can know that info before attempting a design.

For example SEAS, Scan-Speak, and others do provide measurements for raw drivers, but they aim to show what the driver is capable of without limiting it to a particular cabinet design (also known as cabinet alignment).

When SEAS shows data for an individual driver, they suggest a few enclosures that they know can work, but they don’t try to test all possible alignments. An example, for the SEAS Prestige CA18RNX (H1215) 7" Coated Paper Cone Woofer they suggest one sealed box and three different vented boxes.

• Sealed box of 0.2 ft³ for an F3 of 100Hz.
• Vented box of 0.4 ft³ with a 1.5" diameter vent by 5.5" long for an F3 of 65Hz.
• Vented box of 0.55 ft³ with a 2" diameter vent by 5" long for an F3 of 51Hz.
• Vented box of 0.75 ft³ with a 2" diameter vent by 3.75" long for an F3 of 46Hz.

And when they show a frequency response graph, they include this fine print:

"The frequency responses above show measured free field sound pressure in 0, 30, and 60 degrees angle using a 12L (0.42 ft³) closed box. Input 2.83 VRMS, microphone distance 0.5m, normalized to SPL 1m. The dotted line is a calculated response in infinite baffle based on the parameters given for this specific driver. The impedance is measured in free air without baffle using a 2V sine signal."​

Scan-Speak actually provides Tech Notes where they describe some of their methods in more detail:

• Measurement of Thiele Small parameters at Scan-Speak
• Suggestions for cabinet size and tuning for your Scan-Speak drivers

Focal makes its own drivers, but doesn’t offer them for sale as raw drivers – only in loudspeakers they design. They don’t publish details like SEAS or Scan-Speak.

 

[Swerd]

While designing a loudspeaker, you must test all the individual drivers while they are mounted in a cabinet whose dimensions you choose, as determined by (among other things) the bass response of the woofer(s).

Most people here understand that the cabinet inner dimensions (volume) and port dimensions (if any) determine the low end of the woofer’s response. This has to be determined first, before any further measurements can be done. You have to build a test cabinet in which the remaining drivers can be mounted and tested. All this has to be done before designing a crossover. Sometimes, you find that a tweeter or mid-range don't work as well as you had initially thought. So, you have to choose again.

Once you have a cabinet design and suitable drivers mounted in it, you can measure their raw frequency responses and measure the speaker's impedance. It's best to use this data as you design the crossover. Those measurements will change somewhat after a crossover is designed, but their raw responses and impedance values are necessary to design a good crossover.

The baffle step response (an elevated response in the lower mid-range) will be determined by the cabinet's external width. The speaker designer has a number of ways to compensate for this elevated baffle step. One method is to include a baffle step compensation circuit on the crossover board. But there are other ways too, depending on the speaker’s design. Is the speaker a 2-way or 3-way, and will there more than one woofer?

The tweeter’s low end response will also be determined by the external cabinet width and the tweeter’s placement on the front baffle. Is it the same distance from the two sides (centered or off-centered) and top edges, or are there different distances to all the edges?

Here’s an example of a typical 1" dome tweeter mounted in a bookshelf size box (19" H × 8" W) suitable for combining with a 6½" mid-woofer in a 2-way design. The tweeter was measured without any high-pass filter. Notice the peak below 2 kHz and the dip below 3 kHz. That’s what you get from center mounting the tweeter in an 8" wide cabinet. No data from the tweeter’s manufacturer can show that because they don’t know your cabinet plans. The speaker designer must know this info before selecting a crossover frequency. In this example, a steep sloped crossover at roughly 3 kHz looks like a good choice.

 

[highfigh]

Absolutely. However since Focal does not provide that data I would still assume a reasonably linear fr from such a reputable manufacturer. I can't imagine them being much more than +/- 3 dB anechoically through the midrange and higher frequencies. Much below that and the room is in charge anyway. It doesn't matter how they measure (within reason) anechoically below the Schroeder frequency of the room.

I know that a lot of companies stop measuring at 200Hz because anechoic rooms that are effective into the <100Hz range are extremely expensive and to make it truly anechoic below 50Hz is insanely expensive. Much easier and cheaper to go into a field and hope there's no wind.

Still, low frequency problems can be handled without being absurdly expensive- I watched in REW as I placed the panels and it eliminated the trough which made me really happy because it as making me crazy. I do absolutely nothing as far as EQ.

 

[highfigh]

I think there might be some confusion here between measurements for individual drivers and measurements for a finished loudspeaker where several drivers are mounted in a cabinet. I think you do understand this, but others reading this thread might not.

When driver manufacturers publish a frequency response for a driver, they don't intend it to be mounted in only one size cabinet. But they must mount it somehow in order to measure it. They usually tell you enough so a loudspeaker designer can know that info before attempting a design.

When I mentioned measurements for drivers AND speaker systems, it was because I have pored over the many parts in catalogs and if the driver response doesn't work for a design, it's easy to see why in the response chart. I see a lot of tweeters that have great 30° off-axis response but the on-axis response is very 'peaky' at frequencies where this is unwanted. Knitting the drivers together via a good crossover can only be modeled and tested if accuracy is to be had, though. Guesstimating comes easier with much experience and/or luck. I have seen some explanations of testing methodology that showed drivers mounted in a specific size baffle that is then mounted into a wall that's large enough to be considered 'infinite baffle' with additional info about the surface (smooth, or rough).

I wouldn't discount luck in this, either- I have been and I suspect many others, too.

 

[Pogre]

Still, low frequency problems can be handled without being absurdly expensive- I watched in REW as I placed the panels and it eliminated the trough which made me really happy because it as making me crazy. I do absolutely nothing as far as EQ.

I eq my bass. It's pretty bad without it, and I've done a lot of experimenting with the sub crawl and different placement. I've got it dialed in really well with eq now, but I keep telling myself the next time I get the bug to spend money I'm gonna look into some acoustic treatments.

 

[highfigh]

I eq my bass. It's pretty bad without it, and I've done a lot of experimenting with the sub crawl and different placement. I've got it dialed in really well with eq now, but I keep telling myself the next time I get the bug to spend money I'm gonna look into some acoustic treatments.

Mine were leftover panels for a home theater demo I did but ended up not using. I made most 2" thick, a couple are 4" thick, most are 2'x4'- Owens Corning rigid fiberglass on pegboard. I wrapped them with Muslin because they were for a demo but I don't mine seeing them that way. Look for an insulation contractor or distributor or an acoustic supply house and ask if they have any of these with damaged edges or corners. I got almost a whole bundle for less than half price because they just wanted to get rid of it. I hated the sound- the suck-out was very steep and when I moved my head to one side ot the other, the sensation was very unpleasant- it's like higher pressure on one side and lower pressure on the other. If you have a spare woofer, connect it to an amplifier to play music and turn it so you're looking directly at the edge of the frame. Rotate it and pay attention to what your ears feel. I hate being sensitive to it, but it makes diagnosing audio problems much easier.

 

[rjharle]

Correct. And you get those results by taking careful measurements with a mic and REW. Trying to achieve a linear response just by ear is nearly impossible. You can hear when something is off, but you can't necessarily pinpoint which frequencies need addressed without measurements. That's where REW comes in. I see from your following post tho we seem to be on the same page now.

As far as ARC goes, that is also room correction software isn't it? I'm not familiar with ARC. By most accounts it is one of the best programs out there, but isnt it what you're trying to avoid? There are other ways to to tweak your response by leveraging the room's acoustics in your favor using careful placement techniques, phase correcting, etc. If you read Floyd's book I know I don't have to explain that to you. He's a very strong proponent for positional eq and careful placement. I was assuming you were going to use that approach.

Happens all the time, lol. Plus it looks like I could do a little better job reading replies and letting them sink in before responding. I might have misunderstood you at first.

Anthem ARC Genesis is a good for room correction with many features. Modal cutoff can be set from 0hz - 20khz. Since there are 4 profiles each can be set anyway you want.

Since I have to use Anthem ARC for room correction my plan is to apply ARC to Modal 0 - 300 (room) and acoustical panels for the Stochastic 300 - where needed. Will be using REW and Omni.