Well, with the pandemic of 2020/2021, one thing that has changed markedly in my family is that we're spending way more time watching movies and miniseries at home this past year. In fact the last movie I watched at the local "cineplex" was 1917 right before the Oscars on February 9, 2020. As such, 2020 has been a year of microwave popcorn, Netflix, Amazon Prime Video, Disney+ and the occasional UHD Blu-Ray played on my Oppo UDP-205 with the family on the sofa.
Wanting to use the Energy C100 bookshelf speakers elsewhere (see what I was doing here), I figured it was time to grab a pair of actual "Atmos elevation speakers". Not unexpectedly, the ELAC Debut 2.0 A4.2 Dolby Atmos Add-On Speakers were on sale before Christmas which you see in the image above.
This is the first time I've had any ELAC speakers in the home. The speakers are actually not large, but they shipped in a larger-than-expected box. Well nestled in foam protection.
You can see the driver behind the grille. It's a coaxial design with what is described as a 4" aramid fiber (a synthetic textile used in stuff like body armor) woofer with a small 1/2" soft "polymer" tweeter.
Okay then, let's have a quick look at how these little speakers measure... Given that these are passive speakers, let's have a peek at the impedance graph first:
Typical for my speaker measurements, I'll use the Emotiva XPA-1L monoblock as the amplifier. Here's an estimate of the speaker sensitivity with a 2.83V signal, measured at 1m:
From the graph above, we can already appreciate that we're not looking at a flat response speaker! Here's the quasi-anechoic graph based on both nearfield and 1m measurements with baffle compensation applied averaging readings +/-30° (captured at 10° increments) around the coaxial driver:
Note that it wouldn't make sense to do a CTA-2034A-style graph since the speaker is not meant to be listened on-axis. Also for technical reasons, due to where the speaker connectors are located, the only stable way for me to measure off-axis is along what would typically be the "depth" axis of the speaker. Here's how it was arranged if you imagine the microphone pointing straight at it and rotated around the horizontal plane:
In this orientation, check out the directivity polar map:
I have to admit that this is a strange looking polar map :-). There's wide dispersion although it narrows quickly from 6-8kHz then widens again. There's some funky stuff happening above 10kHz, especially at 11kHz where on the frequency response, we see a bit of a dip. If you're wondering, ELAC identifies the crossover frequency at 5kHz.
So what's going on here?
Well, remember that these Atmos speakers were not designed to be "high fidelity" transducers like your main speakers with the aim of achieving relatively flat frequency response on-axis or within a typical "listening window" like +/-30°. Check out the Dolby patent for their "Virtual Height Filter for Reflected Sound Rendering Using Upward Firing Drivers". In particular, we see this diagram for a theoretical Atmos speaker:
"... derived from a directional hearing model based on a database of HRTF responses averaged across a large set of subjects. The black lines 303 represent the filter PT computed over a range of azimuth angles and a range of elevation angles corresponding to reasonable speaker distances and ceiling heights. Looking at these various instances of PT, one first notes that the majority of each filter's variation occurs at higher frequencies, above 4Hz. In addition, each filter exhibits a peak located at roughly 7kHz and a notch at roughly 12kHz..."
In the time domain, here is the step response:
Looks like the drivers in the coaxial arrangement are both in positive acoustic polarity (that first dip down is ringing from the miniDSP UMIK-1 USB microphone rather than the tweeter).
Not great, but fine for what it's suppose to do I think. There's a bit of extra energy around 5.9kHz (cursor) and 11-12kHz.
We can compare a few parameters:
Before with Energy bookshelf monitors acting as rear Atmos channels: