Your thoughts on 2.4 upgrade vs all-FST might be of interest to the group.
I suppose my impressions are scattered across the many pages of this thread, so maybe it is worthwhile to summarize in a single post.
I bought my CS2.4SEs in January, 2018. Of course, these cannot match the bass extension and definition of the better mega-buck speakers. And I was also aware that the 2.4s also fall a bit short of the very best in terms of qualities such as resolution and image density. But at this price point (and even 2-3 times more expensive) I don’t think you can do better for an overall fine speaker. I was really happy and could probably have lived with the SEs forever . . . except for a “glassy” quality in the midrange that I became increasingly aware of after listening to the SEs for multiple weeks. Meanwhile, the dialog with Tom Thiel on this thread had me dreaming of substantially improved sonics possible with improved crossover parts.
My 2012 built SEs had FST-sourced crossovers with MKT (polyester) caps. The first thing I did was replace all sandcast resistors with Mills MRA-12s. With one channel upgraded, I listened in mono using Roon’s DSP to compare with and without the Mills (I used this procedure for all later comparisons). The Mills channel sounded richer and fuller, with just a bit more texture/microdynamics (possibly because of a lower noise floor). Listening in stereo with Mills in both channels, bass seemed to have more impact than I remembered and music was presented with more ease. Most importantly, the glassy quality in the midrange was mitigated.
Tom enrolled me as his beta tester for the 2.4. After several months of correspondence and buying a few parts on my own, Tom sent me a kit with caps, coils, pre-drilled Masonite boards, and miscellaneous supplies. Most of the caps are Clarity CSAs, custom made to Thiel value capacitance, and I independently purchased Multicap RTXs to use as bypasses on the coax feed caps. Coils are ERSE or Jantzen air cores, matching Lex parts, except for one each on the coax and woofer boards in the feed position. These two are ERSE Foil Q, one of which was custom built to match the Thiel inductance value. The new build retains nothing from the FST boards.
There were a few bumps in the road and a few things needed to be tested and measured before I began the build in earnest. And I had to back track at one point as an aspect of the build proved somewhat deleterious. I’ll spare the details other than to share an unrelated caution from Tom Thiel to the DIY community: be careful to match the DC resistance of the original crossovers!
In mono listening with FST+Mills in one channel and the new build in the other (retaining only the Mills MRAs from the previous iteration), the CSA channel had better resolution, more textures, and apparent microdynamics for *every* voice and instrument. The CSA channel was more open, clear, and transparent. The FST channel was relatively veiled and “woolly” sounding. On some tracks these differences were subtle and on others more obvious, with textures on vocals and guitars having previously escaped my awareness.
Later, I compared with and without the RTX bypasses. These produced more subtle differences (in fact, some tracks sounded indistinguishable to me), with improved transients (most notable with percussions) and slightly improved resolution/textures. A few tracks, especially concerto selections, sounded richer with the RTX. I also compared two voltages of CSA caps on the woofer board. Again, differences were relatively subtle and varied from track to track. That said, the higher voltage version seemed to have more heft on bass, kick drums and the like. Many, but not all, songs seemed to have improved ease of presentation with the higher voltage version. These are both shunt caps and it is controversial whether these have sonic consequences. To my ears, they do! As an aside, I did my comparisons after 100-200 hours of burn in but I suspect the new parts were not fully settled until closer to 300 hours.
That said, it wasn’t until I installed the new boards in the other channel, listening in stereo, that I really appreciated the upgrade. I literally had a smile on my face for the next couple of hours. The clarity, intelligibility, openness, ease, resolution, and transparency are exceptional. I think the upgraded 2.4 nearly matches the very best speakers I’ve heard in those regards (my short list includes Vandersteen 7, TAD Ref One, and Vivid Giya – all north of $50K). Bass depth, of course, has not been improved by this upgrade although I perceived improved bass impact/heft. That “error of omission” aside, I can only muster two criticisms: 1) image density does not match the very best I’ve heard; and 2) the high resolution reproduction reveals flaws in recording and mastering. This latter point seems to be the only downside of the upgrade. But, hey, I’ve always said I want to hear what’s on the recording to the utmost fidelity. And well recorded music is sublime. Without question this is the best my system has sounded. It’s easier than ever for me to turn the lights off and bliss out listening to my favorite music.
I was very surprised at Tom’s Lex v FST results. But I told him I was *not* putting the FST back in. In fact, the old boards are on their way to him as I write this.
;^)
