My take on this debate is that it’s not just low bass we’re talking about, but the critical mid-bass to lower midrange frequencies of 80 to 500hz. Frankly, this is a difficult frequency range for any dipole speaker, not just ML’s.
The issue is two-fold, the first is back wave cancellations. As frequency goes down, the wave length grows, making it easier for the back wave of the panel to ’see’ the out of phase front panel wave and cancel out at ranges from around 250Hz on down.
Secondly, this is also right in the core range of the ML lines panel to dynamic woofer crossover. (Monoliths at 150 up to Clarity at 450hz) As most of you know, a passive crossover has ‘interesting’ phase and frequency response compromises that will affect the sound at and around this point.
An ETF measurement of an ML in-room almost always has some serious dips in the 80 - 400 range.
Now, how does the above affect the ‘sound’ when compared to dynamic speakers? First, my experience has been that the upper bass, lower midrange suck out at the lower end of the Panel FR is the primary culprit. Remember the ASC-BackBox (
http://www.asc-hifi.com/whats_new.htm )? It would damp the rear wave. I recall someone commenting that it made a big difference in the FR we’re talking about as well.
I personally put an absorber (Auralex Corner bass trap) on the rear shelf of my monoliths to help dampen this range. It is measurable in ETF, and I’ll go dig out the graphs to post.
Therefore, if the speaker system is not putting out enough energy at the listening position in this FR, it will not sound as powerful as one that is flatter and more uniform in dispersion (at these frequencies)
The second way this issue affects the sound is due to what I believe is the power compression curve differences between electrostats and dynamic speakers. My take is that an electrostat is very clean with very low compression artifacts above 500hz. Whereas most dynamic speakers have crossover FR/Phase issues above 500hz along with the natural tendency of a 1” tweeter to start to either distort or power compress as the SPL’s go up. An electrostat has such consistent mid’s and highs that they tend to mask the actual power of the bass and mid-bass in a comparison with a dynamic speaker, whose mid-bass is usually steady, but the highs are getting compressed (in relative SPL), thus tilting the perception towards ‘punchy‘.
Therefore, adding a sub is never the full fix. It’s definitely part of the equation, but as stated, I believe the perceptual difference is in the upper bass and lower mids. Fixing this through room acoustics, placement and judicial use of EQ are all valid ways to deal with it.
For instance, I was never happy with the original Monolith passive xover. After I switched to active, it was a whole new world across the board, but especially in the 80 to 300 range, it was much smoother and better integrated (fewer phase issues for one). After switching to the DriveRack speaker processor, even more benefits accrued and FR is the smoothest yet as now I can adjust delays, adjust phase and pick the right slopes and xover models. Oh, and I can EQ away as well. But I still have significant dip around 250 (-6 to -8 DB) that is primarily back-wave cancellation.
To my mind, the fix is to do what ML has already done in the Statement e2 and use a ‘midrange dipolar array’ (or a monopole array might work as well). To get the kind of ‘punch’ one hears from dynamics, I guess you just have to find a way to blend them in.
As always, audio is the art of using science to arrive at a compromise you can enjoy.