Understanding and interpreting CTA-2034 AKA Spinorama or Spin Curves

[Arthur Edwards]

I opened up this thread because so much general information is being posted in the reviews of speaker by the experts. It is going to get hidden so I thought it should be collected in one place.

I am not expert on this but I have a small opening post on the details of how each early reflection curve is calculated.

I hope this thread will have contributions on different aspects of the curves.

In an earlier post I mentioned this is the first time we are seeing the internals of the early reflection curve. The spin curves were created so that consumers could understand the performance of a speaker from a small curve set of 5 measurements. The two new curves to the set were the Early Reflection Curve and its Directivity Index. The listening window, power response and its directivity index are decades old concepts. The paper below was presented at the Audio Engineering Society conversion by the developer of the spin curves:

Devantier Allan, “Characterizing the amplitude response of loudspeaker systems”, presented at the AES 113th convention, October 2002, Convention Paper 5638

Unfortunately this is under the AES firewall. Google Scholar is not showing any free downloads for the paper but perhaps the can be found. From the abstract:

“A survey of 15 domestic multi-channel installations was used to determine the typical angle of the direct sound and the early arrivals. The reflected sound that arrives at the listener after encountering only one room boundary is used to approximate the early arrivals, and the total sound power is used to approximate the reverberant sound field. Two unique directivity indices are also defined and the in-room response of the loudspeaker is predicted from anechoic data”.

“The paper shows the 15 rooms in most cases 5 channels. L shaped rooms are included. In some case the speakers are placed across a corner (triangular ). “All of the rooms were located in Southern California, nine were in homes and six were in apartments.” Volume runs from 40 to 160 square meters.

“The “First Reflections” curve is simply defined as the average of the Floor, Ceiling, Front, Side, and Rear curves”. The radiation angles used to calculate each of the 5 reflection curves was found by looking at the average angle for the 15 rooms and the distribution.

The curve marked “Early Reflections” in the JBL 305P MKii and Control 1 Pro Monitors Review shows the 5 reflection curves and the average. As far as I know this has never been published. It is not in the Devantier paper. One can see that each individual curves are different and the Total Early Reflection average smooths out much of the information in each curve. I expect when a good speaker is identified in the traditional spin curve the shape of the individual reflections may give clues as to why the “good” speakers sound different.

Klippel supplies additional graphs. “Vertical Reflections” is just for the floor and ceiling reflection which are dominated by the vertical radiation pattern of the speaker. The new curve is the average of just the vertical curves. “Horizontal Reflections” are the walls which are dominated by the horizontal radiation pattern of a speaker. In this curve set the new curve is the average of just the horizontal reflections. Very different design aspects of the speaker create differences in the vertical and horizontal reflections. I believe the vertical and horizontal averages are also first publications.

I know some want to simplify the amount of data being presented but in these early days it is not clear what is important, especially for the case of these reflection curves which are a new way to present radiation pattern measurements.

You can also tell me what I got wrong in what I wrote above.

Thanks to everybody that decides to contribute to this thread if it looks useful. Perhaps it is not a good idea at all.

 

[pozz]

Thanks for starting this thread.

I think you should buy @Floyd Toole 's book Sound Reproduction, specifically the 3rd edition from 2017. A good part of it explains the reasons for the measurements in the standard, as well as the technique and methods of interpretation and application. If you summarize his findings it will be immensely useful given the confusion reigning in the first few speaker reviews Amir's posted.

There is also this talk, not about the standard specifically but useful nonetheless to understand the background: Floyd Toole - Sound Reproduction: Art & Science/Opinions & Facts.

 

[mitchco]

Another good resource with explanations is the actual ANSI/CTA 2034-A, “Standard Method of Measurement for In-Home Loudspeakers" itself.
It is a free download

The standard includes a number of informational annexes to help readers gain a more thorough understanding of techniques for acquiring loudspeaker data in both anechoic and non-anechoic environments, as well as methods for using this acquired data to predict loudspeaker performance.

 

[napilopez]

A few things:

• I'd mentioned it in the other thread, but buchardt publishes ceiling, floor, and wall reflections curves. Still very rare to see.
• The Early Reflections Curve does not include the full "horizontal reflections"curve. The latter includes the full 360-degree horizontal measures, while for former only goes out to 90 degrees and includes a single 180-degree measurement. I assume the "horizontal reflections" curve is meant for non-monopole speakers with unusual radiation patterns.
• The Ceiling reflections curve specifically is the average of 40,50, and 60 degrees above the reference axis
• The Floor reflections curve is 20,30,40 below the reference axis.

I very much appreciate seeing the ceiling and floor reflections curves, though the bulk of their effect can be seen in the ER and sound power curves.

 

[Arthur Edwards]

Thanks for starting this thread.

I think you should buy @Floyd Toole 's book Sound Reproduction, specifically the 3rd edition from 2017. A good part of it explains the reasons for the measurements in the standard, as well as the technique and methods of interpretation and application. If you summarize his findings it will be immensely useful given the confusion reigning in the first few speaker reviews Amir's posted.

There is also this talk, not about the standard specifically but useful nonetheless to understand the background: Floyd Toole - Sound Reproduction: Art & Science/Opinions & Facts.

I have Dr. Toole’s book both edition 1 and 3. Edition 2 is the same as edition 1 with a publisher change. Edition 3 has material re-ordered making it easier to read and I highly recommend it. It is impossible to summarize his work and it is best to get a copy of the book.

Harman used to have PPTs which had figures that would go into the book. Many figures on the location of the bonces in a room and large section on how to read what would become CTA 2034 graphs. Harman took it all that down. I tried the Wayback Machine and got to the old download site but none of the PDFs will download. You can see they had many of these on different topics.

https://web.archive.org/web/2015011...e/inf/us/about/content.jsp?id=ourTechnologies

I do not know if someone archived these PDFs on some other site.

 

[Arthur Edwards]

A few things:

• I'd mentioned it in the other thread, but buchardt publishes ceiling, floor, and wall reflections curves. Still very rare to see.
• The Early Reflections Curve does not include the full "horizontal reflections"curve. The latter includes the full 360-degree horizontal measures, while for former only goes out to 90 degrees and includes a single 180-degree measurement. I assume the "horizontal reflections" curve is meant for non-monopole speakers with unusual radiation patterns.
• The Ceiling reflections curve specifically is the average of 40,50, and 60 degrees above the reference axis
• The Floor reflections curve is 20,30,40 below the reference axis.

I very much appreciate seeing the ceiling and floor reflections curves, though the bulk of their effect can be seen in the ER and sound power curves.

I did not see what Burchardt presented. What is new on this site is the reflection from the three walls are shown. Note how much stronger the front wall is then the rear wall as a result of the longer distance the signal has to travel to the rear wall before the bounce. I do not know what implications this has on wall treatments.
.

 

[napilopez]

I did not see what Burchardt presented. What is new on this site is the reflection from the three walls are shown. Note how much stronger the front wall is then the rear wall as a result of the longer distance the signal has to travel to the rear wall before the bounce. I do not know what implications this has on wall treatments.
.View attachment 46017

This is the specific Buchardt Graph I'm talking about (who, by the way, also use the Klippel robot).

Same stuff, they just don't include the rear wall bounce. In any case, you're right these curves are very rare, so I'm glad amir is publishing them, and I'll be including at least the vertical measurements in my reviews too going forward(I started with my most recent one).

One correction: The rear wall curve has nothing to do with the delay or longer distance in rear-radiated sound.

Remember, these are anechoic measurements; there is no bounce in the data. Instead, the rear reflections curve is a measure of the actual SPL levels a speaker is radiating at different frequencies in its rear hemisphere . As expected for a monopole, that's almost nothing in the treble because the tweeter is directional, but virtually identical in the bass because woofers become omnidirectional once the wavelengths become larger than the baffle s the waves will "wrap around" the enclosure.

This is the same reason the DI curves for any monopole show speakers are at zero in the bass. THe DI curves are the difference between the listening Window (or sometimes on-axis) and the ER or SP curves. When speakers are omnidirectional in the bass, there basically is no difference.

It's pretty easy to infer the rear reflections curve for a monopole speaker. It will almost always look like the one above - with the slope varying a bit depending on the woofer placement and baffle size, and will make a negligible contribution to what we hear in a room. It is much more useful in omnidirectional, bipole, and dipole designs, because then the back-radiated sound has a more meaningful effect on our perception.

 

[Arthur Edwards]

This is the specific Buchardt Graph I'm talking about (who, by the way, also use the Klippel robot).

Same stuff, they just don't include the rear wall bounce. In any case, you're right these curves are very rare, so I'm glad amir is publishing them, and I'll be including at least the vertical measurements in my reviews too going forward(I started with my most recent one).

One correction: The rear wall curve has nothing to do with the delay or longer distance in rear-radiated sound.

Remember, these are anechoic measurements; there is no bounce in the data. Instead, the rear reflections curve is a measure of the actual SPL levels a speaker is radiating at different frequencies in its rear hemisphere . As expected for a monopole, that's almost nothing in the treble because the tweeter is directional, but virtually identical in the bass because woofers become omnidirectional once the wavelengths become larger than the baffle s the waves will "wrap around" the enclosure.

This is the same reason the DI curves for any monopole show speakers are at zero in the bass. THe DI curves are the difference between the listening Window (or sometimes on-axis) and the ER or SP curves. When speakers are omnidirectional in the bass, there basically is no difference.

It's pretty easy to infer the rear reflections curve from in a monopole speaker. It will almost always look like the one above - with the slope varying a bit depending on the woofer placement and baffle size, and will make a negligible contribution to what we hear in a room. It is much more useful in omnidirectional, bipole, and dipole designs, because then the back-radiated sound has a more meaningful effect on our perception.

 

[Arthur Edwards]

One correction: The rear wall curve has nothing to do with the delay or longer distance in rear-radiated sound.

Yes I was wrong and you are correct. Thank you for pointing this out.

 

[Dave Zan]

I opened up this thread because...it should be collected in one place.

Hi Arthur, I had the same idea and opened a similar thread in the Speaker Review, Test sub-forum a few days previously.
https://www.audiosciencereview.com/...a-2034-but-that-sounds-dull-apparently.10862/
So perhaps we should unify the two threads? our issues seem very similar.

Best wishes
David

 

[Arthur Edwards]

Hi Arthur, I had the same idea and opened a similar thread in the Speaker Review, Test sub-forum a few days previously.
https://www.audiosciencereview.com/...a-2034-but-that-sounds-dull-apparently.10862/
So perhaps we should unify the two threads? our issues seem very similar.

Best wishes
David

David

Please go ahead and use your thread and close this one. Sorry I did not see that you started one on the same subject.

Arthur