• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Exploration of data provided by vendors

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
There are a lot of available data on PA speakers. Companies are used to share data mostly because buyers need them to compute how many speakers and which ones they need to light up a stadium. I discovered recently that a lot of companies published data in CFL and GLL format. GLL has a free viewer and you can export data. So that's what I did: I used windows automation to automate the extraction of data from the GLL files provides by JBL, QSC, Neumann, Genelec, Meyer Sound you name it.

EDIT Spinorama covered
this post
- Genelec 8361A
- Neumann KH 420G
- JBL CBT 70J-1
- Meyer Sound Ultra X20
- Meyer Sound Lina
- QSC K8.2
- JBL 9310
- JBL C211
- JBL C222
post #6
- L Acoustics coax 108P
- L Acoustics X8
- Meyer Acheron 80
post #8
- Fulcrum FH15
- Fulcrum DX15
- Fulcrum CCX15
- Fulcrum CCX12
- Fulcrum GX12
- Fulcrum GC15
post #16
- Alcons Audio VR5
post #17
- Kling Freitag PIA M
post #19
- Alcons Audio QRP20
post #21
- Yamaha DXR8mkII
- Yamaha DZR315


I am transforming a lots of GLL files to something we are used to see. Thanks to @test1223 for this idea. That's not a fully automated process and results needs to be manually checked.

Concordance with speakers measured by @amirm is usually great but I also see a lot of very smoothed data.
Some examples below:

ASR v.s. Genelec gll data
1663673787721.png



ASR v.s. Neumann gll data
1663673800174.png



Things can diverge a lot more
1663673815781.png


Here are a few PA speakers examples:

Meyer Sound Ultra X20
CEA2034


Estimated In-Room Response
SPL Horizontal Contour
SPL Vertical Contour

Meyer Sound Lina
Interesting speaker, please not the DI is almost constant and that freq response can be EQed. They are for line arrays.

CEA2034

SPL Horizontal Contour
SPL Vertical Contour

QSC K8.2 (data is very smoothed)
CEA2034


JBL 9310 (data is very smoothed)
CEA2034


Building a ranking system for PA speakers will be difficult since they don't necessary have the same domain of application.
Classical dipole or tri-ways, yes they match the directivity of home speakers.
Line arrays are different beasts and you cannot really compare them to something else.
Surround are also different.
Most of the PA speakers are also not covering the whole frequency range and subwoofers are
always added to the mix.

Some fun discoveries:

JBL C211
Interesting spike and very flat DI above 2k.

CEA2034.jpg


JBL C222
Freq response looks crap but look at the DI and the PIR:

CEA2034.jpg

Estimated%20In-Room%20Response.jpg

Not bad at all.

Conclusion so far
-
Lots of data out there some of it of high quality (Meyer Sound, Neumann, some of the Genelec), medium quality (JBL), low quality (QSC). I have found hundreds of gll files in a few Google searches.
- PA speakers have different DI (no surprise)
- Some PA speakers are excellent (Meyer Sound Ultra X20, X22, X23, X40) for example.
- You need to read the footprint in each gll file (smoothing, precision, low freq validity, which angles are available...)
 
Last edited:
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
Thanks a lot @pierre, really amazing work!
Will you be adding these new datasets to your web spinorama tool?

Yes, you can see some of them on the dev site (with plenty of small issues that I need to fix before i push them on the main site).
You can have a peek at https://dev.spinorama.org/ which may or may not work.
 
Last edited:
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
I did convert a few more files.

L Acoustics PA speakers 108P, 112P, X8, X121, X15 HiQ. I give some example here and you can find all the details here. The are PA speakers designed to deliver
120dB+ and sometimes a lot more. They are expensive but you can find them used for a lot less.

L Acoustics coax 108P
Good results even if the data is smoothed. Could do well as a LCR for a (very) large home cinema.

CEA2034
SPL Horizontal Contour
SPL Vertical Contour


L Acoustics X8 (passive )
CEA2034
Estimated In-Room Response


All graphs are on github.io.

Meyer Acheron 80


Frequency



Since you have a lot of headroom add some EQ:
Frequency with EQ
Frequency with EQ
Frequency with EQ
[/URL]

I have guessing the blue horn from Meyer are excellent. They also have a nice Acheron LF to put below the 80 and enhance the bass.

Next post will be about different kind of speakers. Line arrays, or columns with different directivity. If you want a GLL file to be converted, PM me.
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
BTW, Fulcrum Acoustic is another high quality PA speaker manufacturer that provides such data for most of their lineup.


Some interesting ones to look at:



*pattern coverage can be selected

Unfortunately, there's no way to practically (and aesthetically) fit a huge horn such as the one above in people's homes!
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
@ernestcarl

Some fulcrum in different sizes and different horizontal and vertical patterns.
FH15 for exemple is very large

Fulcrum Acoustics CCX1500.jpg
Fulcrum Acoustics DX1526.jpg
Fulcrum Acoustics FH1500.jpg
Fulcrum Acoustics GX1200.jpg


ErKvrW5XUAgPKLd.jpg:large

CEA2034.jpg


CEA2034.jpg
CEA2034.jpg


CEA2034.jpg

CEA2034.jpg

CEA2034.jpg
 

Attachments

  • Fulcrum Acoustics FH1500.jpg
    Fulcrum Acoustics FH1500.jpg
    75.9 KB · Views: 49
  • Fulcrum Acoustics CX1226.jpg
    Fulcrum Acoustics CX1226.jpg
    48.6 KB · Views: 63
  • filters_eq.png
    filters_eq.png
    933.7 KB · Views: 55

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Can you also state which coverage patterns were selected for the spin plot models posted, thank you!

Turns out the last two digits indicate the pattern model.

26 is 120x60 (wide)
65 is 60x45 (narrow)

Though, I would rather check the website & specsheet to make sure. Some beamwidth patterns are more linear than others, but the trend seems to be sided more on the narrower ones as somewhat being more even. I also expect the ‘reference’ series to be flatter due to the lack of grill and its edges, and likely EQ’d as well to within tighter tolerances.

Two models @pierre posted only have ‘00’ at the end so don’t know which pattern they are exactly.

As to the suggested PEQ graph, I would generally avoid super steep and narrow filters in the HF — those mostly innocuous sharp diffraction dips that’s maybe left over deliberately, while some of the peaks seem may be left in place to offset off-axis dips at the wider edges of the pattern coverage. We also don’t know the unit-to-unit tolerances in these drivers. Much better to EQ yourself if desired if you can take a spatial average and/or multi-point (perhaps also windowed) measurements. Dave Gunness himself noted in an interview that he is rather quite conservative in the EQ (uses as little as he deems necessary) despite use of FIRs.
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
Turns out the last two digits indicate the pattern model.

26 is 120x60 (wide)
65 is 60x45 (narrow)

Though, I would rather check the website & specsheet to make sure. Some beamwidth patterns are more linear than others, but the trend seems to be sided more on the narrower ones as somewhat being more even. I also expect the ‘reference’ series to be flatter due to the lack of grill and its edges, and likely EQ’d as well to within tighter tolerances.

Two models @pierre posted only have ‘00’ at the end so don’t know which pattern they are exactly.

As to the suggested PEQ graph, I would generally avoid super steep and narrow filters in the HF — those mostly innocuous sharp diffraction dips that’s maybe left over deliberately, while some of the peaks seem may be left in place to offset off-axis dips at the wider edges of the pattern coverage. We also don’t know the unit-to-unit tolerances in these drivers. Much better to EQ yourself if desired if you can take a spatial average and/or multi-point (perhaps also windowed) measurements. Dave Gunness himself noted in an interview that he is rather quite conservative in the EQ (uses as little as he deems necessary) despite use of FIRs.
Re EQ I agree, I don’t know why it went crazy like this. I will tame it later.
Re 00: that’s the 100x100 pattern
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Re EQ I agree, I don’t know why it went crazy like this. I will tame it later.
Re 00: that’s the 100x100 pattern

Ah… didn’t think there’d be something that wide. Still, that’s good to know — thanks for the clarification!
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
For the CCX15 here are all the patterns. Same speaker but with a different horn each time and you see the impact on DI and Horizontal contours.

CCX1500.jpg
CCX1526.jpg
CCX1565.jpg
CCX1577.jpg
CCX1595.jpg
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
For the CCX15 here are all the patterns. Same speaker but with a different horn each time and you see the impact on DI and Horizontal contours.

View attachment 233182View attachment 233183View attachment 233184View attachment 233185View attachment 233186

Nice! Looks easier to quickly compare all them at once. From this perspective I would have to say that I like 65 and 95 the most. For wider dispersion needs such as in surround applications 26 (120x60) may be an acceptable compromise as well.

It seems like as one tries to increase vertical coverage here, evenness or flatness in the horizontal plane gets compromised so the more symmetrical 00 and 77 patterns may be better reserved for more specific installation applications where wider vertical dispersion is equally as important as the horizontal.
 

test1223

Addicted to Fun and Learning
Joined
Jan 10, 2020
Messages
503
Likes
511
Hello

@pierre very nice work. You add substantial value to our small audio nerd community!

It would be really good if there is a warning or something similar on your web page that some specific measurements provide a very low resolution (e.g. the QSC speakers). The preference score of such speaker is high because of low quality measurements, which is unfair.


I am interested in the spinoramas of the following speakers. It would be really nice if you can generate them :)

Kling & Freitag PIA M
https://www.kling-freitag.de/content/uploads/kf-pia-m_gll.zip

KS-audio c-04

Alcons audio VR5

Best
Thomas
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
Alcons audio VR5

Best
Thomas

Alcons VR5 is a small monitor but well designed.
Nice spin with very smooth DI so EQ will work well.
CEA2034.jpg


Early Reflections.jpg


Estimated in-room response is excellent out of the box:
Estimated In-Room Response.jpg


Horizontal dispersion is well controlled with +40/-40 pattern:
SPL Horizontal Normalized.jpg


SPL Horizontal Contour.jpg


Vertical pattern is narrow: better have your ears at tweeter height:

SPL Vertical Contour.jpg


Let's add an EQ to the mix

Code:
Score: 4.7 with EQ 5.5
 + Sub 7.7              8.5

filters_eq.png


Good stuff and excellent data provided by Alcons.
 

Attachments

  • Estimated In-Room Response.jpg
    Estimated In-Room Response.jpg
    75.5 KB · Views: 46
  • CEA2034.jpg
    CEA2034.jpg
    92.2 KB · Views: 41
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
Kling & Freitag PIA M is a column speaker. It is a passive one with 2 sections: the bass part can be set to bass reflex or cardiod mode, the high part can be set to 3 different patterns (narrow, medium, wide). So you have 6 modes to play with.

Let's start with the bassreflex wide: that's chaos or it looks like it.

CEA2034.jpg


DI curves are different from the classical dipole speaker. You see the impact on the Estimated in room response which is not bad at all.
Estimated In-Room Response.jpg


The interesting parts are the radiation patterns: horizontal first normalized (pretty good)
SPL Horizontal Contour Normalized.jpg


and then horizontal:
SPL Horizontal Contour.jpg


The vertical pattern is typical of column with a very very narrow angle:
SPL Vertical Contour.jpg



Switching to cardiod wide mode, you clearly see a nice impact on the horizontal view:
SPL Horizontal Contour.jpg



The 6 variants will be available soonish on the website with all the gory details.
 
OP
P

pierre

Addicted to Fun and Learning
Forum Donor
Joined
Jul 1, 2017
Messages
962
Likes
3,046
Location
Switzerland
Now is the turn of the passive small column from Alcons Audio the QRP20. It exists in 2 versions with a 90º and a 120º pattern. Let's look at the 90º first.

As often with columns you expect a different spinorama than for a dipole:
Alcons Audio QRP20

CEA2034.jpg

DI is very linear which bodes well for post EQ.
Harmann score is 3.5 (4.7 with EQ). With a perfect subwoofer, score is 6.7 and with EQ 8.0 which is excellent.

filters_eq.png


Horizontal contours of the 2 profiles where you clearly see the difference of 120 vs 90 degrees.
newplot (2).png

newplot (1).png

The vertical pattern is very narrow. If you listen at 5m which is a good use case for this then it is not bad.

newplot (3).png



Excellent little speaker. You can put one on each side of your screen and be happy. SPL will not be an issue.
 
Last edited:
Top Bottom