Thiel Owners

bcupari - and all - keep your radar out for spare 3.5 drivers. They are no longer made and in short supply. If enough interest develops, I hope to address the 3.5 upgrade in a subsequent round.

Regarding 2.4: The XO part of the SE upgrade is simply replacing the 2 coax feed caps with ClairtyCapSA (best at the time). Rob @ CSS has them.

Our upgrades are trumping that in spades. We are replacing all components with carefully chosen, cost effective, high performance parts, and then backing off for a few less expensive cost/performance plateaus. beetlemania and I are addressing the 2.4.
Highlights: later 2.4 XOs were made in China and have less than best coils. We are replacing with old Thiel spec high-test coils in shunts and high-test foil coils in feeds. Resistors were sand-cast and we are using Mills MRA-12s, best of form, great improvement for the buck and near best at any price. beetle has tested and I have confirmed. It's a go.
Caps are ClarityCap for lots of reasons, will explain more later. Two generations past the SA. The CSA is state-of-the-art. All CSAs including replacing all electrolytics with a custom CSA. Bypasses are combination of Thiel custom styrene/tin 1uF with ultra-bypasses: MultiCap RTX or CC-CMR and/or various silver mica and other ultra caps being auditioned. That's the highlights. These parts are CS7.2 and better plus a return to hardwire point to point boards, with added thermal sinking and vibrational damping. Serious stuff toward considerably higher refinement than Thiel's traditional max bang for the buck orientation. We're looking very closely at cost, but making the assumption that you, the owner, has and loves the speaker and will consider this upgrade compared with buying a different speaker. I am confident we will create an exquisite sonic outcome for reasonable cost. Stay tuned. Sorry for delays. Life intervenes.

unsound - Thank you for your point of view. It is entirely valid and I do not intend to recommend the Benchmark for the reasons you cite. But my task is to find source and process equipment that operate at the edge of serviceability rather than big, bulletproof, proven amps which are known to be great. Dealers that succeeded with Thiel generally used Krell FPB-600 or Big Levinson / Ayre etc. etc. You guys know what works. I know that those will all work better with my designs. I must find solutions to the larger problem of most amps not working well enough with Thiel designs. 

I am hoping to mitigate what I judge as somewhat misplaced design focus on Jim's part. He considered amp performance as amp-maker's problems and if an amp can be found to drive his speakers, then there is no problem. A disconnect that I see includes that the internal hoops we jumped to keep prices affordable were overshadowed by the costs of suitable amps. (Indeed those amps can now be bought used for less. Good.) I am working on the assumption that those amps will perform well without my intervention. My mission includes implementing design strategies that produce great performance with less than ultimate amps.

One such strategy is the particular bi-amp configuration I mentioned. I hope to prove a cost-effective way to drive Thiel's brutal loads with less cost than big iron amps. I want a very different amp than my classic Classe configuration. I will final test with Rob's FPB-600s among others. But I also want to see what I can get with a pair of affordable new world amps costing less than $3K new.

unsound - you are right. There may be some backlash. But we here are an insider beta group sharing a development process. Indeed we will test at full power. BUT everything we are doing will improve performance at high power. That end of the use spectrum is not my concern. My concern is getting maximum bang for the maximum head count from the upgrades.

Someone asked for a list of amps which I have dismissed. I don't have such a list. I read widely and systematically. I know that amps color the signal and how that generally works. I look for clues and attitudes in reviews and user comments, and add that to my personal experience. I know that if I use tonally and temporally accurate equipment with well made source files that my results will be valid. I am looking for the needle in the haystack and have not itemized the haystack.

Beyond amps, I am using Klippel and SpectraFoo Phase Torch and signal analysis gear. I know symptoms from square wave behavior and am familiar with measurement vs listening protocols; I helped develop those protocols at Thiel. Jim went on to amass tens of thousands of hours of experience correlating data with listening. I can't claim that. But I do know the territory, including considerable live and recording experience beyond what Jim brought to the table. I am taking an alternative perspective.

I'll be straight with you guys. I do not have the financial means to buy a dozen amps for comparing and proving. I must choose some tools within my means which tell me as much as possible about what I am doing and where I am going. As the project might gain momentum, I might get access to amps on loan, etc. (For reference, at Thiel we always had more than a dozen amps on reciprocal trade from top manufacturers.) I hope that JAFant or others of you might create a recommendation / dis-rec list from your considerable combined experience - beyond the conversation we have been sharing. That list would provide a huge service / reference to the 4000+ members of this forum.

Off for now. I'm running sound at the Open Mic at our Village Arts Center.
Next week the Piano Guild is testing my alternative bridge in their project piano. The piano, the guitars and the speakers all co-inform my understanding. Interdisciplinary Boogie.

Indeed - mixing amps is extremely treacherous, which was the main reason, along with mixing cables, that Jim axed the dual binding posts. Good points about cables and special considerations.

I suspect the bi-amp advantages are not special to the 3.5. Amps constrict in various ways when high current draw surpasses the reserves. Vertical bi-amping supplies double the power and also sequesters the problem area to itself. In other words it makes more sense if bass overdrive results in distorted bass (sequestered channel), than if bass overdrive results in distorted treble (full range channel).

By the way, please push me. I am feeling this out as I go.

jacob - we don't yet know what form the upgrades will take. One complicating factor is that Thiel crossovers since 1990 / CS2.2 use silver/tin solder which requires technical know-how. We will probably have a first-level upgrade of critical caps and all resistors that the adept user can implement. The higher levels are completely new layouts and will probably come as new boards with probable instructions for added cabinet braces and other tweaks. Too early to say. I know I shouldn't say. But, hey, you guys are family.

Pops - nice to see you. I picked up a pair of 3.6s last year and listened, and said wow. Then I measured it and said wow. Then I looked at the crossover and said wow. The 3.6 in 1992 followed the 1990 2.2. For the 3.6, Jim had delved into finite element analysis to make serious refinements to the driver motors to radically lower distortion - and Vifa loved the collaboration. Lots of that stuff is now in broad use.

The 3.6 was the last Thiel product in which I directly participated in the development. After that my work was almost entirely process and capability development for the coming larger products.

That tweeter was developed for the CS5 and trickled down to the 2.2 and 3.6.

prof - I believe the 2.4 has an MDF baffle, as the model 2 has had from the beginning. Whereas earlier 2s had 2" thick, I think the 2.4 is 3" thick like the 3.6. beetlemania knows, he has been in there.

The baffle is one thing. Another is that the 3.7 XO is all high-quality film caps in all feeds and the only electrolytics are in a resonance / shaping circuit of the midrange and tweeter, which is the most benign place for them and they're bypassed with the custom 1uF styrene / tin foil cap. The 2.7 adds a 400uF electrolytic feed cap in the midrange feed. But it is bypassed with a 15uF PP and the 1uF styrene / tin. That electrolytic feed probably does a little damage.

Then there is geometry. Looks aside, the 3.7 aluminum nacelle rocks from a functional perspective both interior and exterior diffraction. And as you say, the aluminum baffle. More budget for the 3 than the 2. Our 2.7 hot-rod will replace all electrolytics with custom ClarityCap CSAs. We'll compare sonics for cost efficacy.

Among the materials I have developed / tested are: fired ceramic panels, fiber reinforced hydrostone panels and/or corner braces, aluminum bars or channels, solid wood struts and, of course, the pierced MDF shelves in all Thiel cabinets. One thing that might not be readily apparent is that driving resonances higher in frequency is of great benefit. Heavy materials may be very stiff, but their mass pulls the cabinet modes lower where there is far more energy to activate them and the results are more harmfully audible. Also, damping materials spread resonances over broader ranges and make them last longer in time, becoming more audible. Each method carries its baggage.

Note that extremely expensive speakers spend lots of money on vibration control.

There are multiple brace shelves stacked in the speaker. I will find resonant areas such as between woofer and passive on the 2.2 and devise a brace. I am also getting promising results soaking the driver mount areas with a wood hardener. My super charged idea is to add a hard spine up the back of the cabinet and connect all magnet assemblies to the spine with rods for combined cooling and anti-recoil effects. An aluminum plate seems unfeasible, or at least I haven't gotten any ideas.

The side walls of the 2.7 are a smaller clip of the actual 3.7 panels, composed of multiple woods in assymetric layup with constrained layers - a technical tour-de-force for some high-buck brands to drool over. Thiel vacuum bagged the finish veneers, which they laid up in-house,  onto those panels in-house. Past tense seems a little weird. Curved walls geometrically resist standing wave resonance. Great idea, very difficult to make work. Imagine aligning everything for assembly!

The 3" baffle is MDF - which is nicely internally damped and pretty ideal except for the less than rigid driver recoil-launch characteristics, which I am addressing with penetrating hardener in and around the driver-mount recesses. No end to the fun.

Thanks for the 2.7 review leads. It is peculiar that Stereophile never reviewed it. The Thiel Audio company was sold within months of the 2.7 release, which may have broken the long-running history of mutual appreciation between Thiel and Stereophile. From the very beginning, Stereophile appreciated our efforts and consistently validated our results. I consider the review journals as among the best allies one could dream of.

prof - congratulations on your turntable isolation. Speakers add to the difficulty of managing isolation, the control of driver recoil and reflection-vibration, but without the luxury of isolating them from the cabinet which also serves as their spatial reference. Driver bounce reduces sonic incision.

You picked good materials for your isolation bases. Among woods, the maples have high internal damping which increases as frequencies descend, which is unusual and helpful in your situation. If your annual rings were running vertically (turntable-floor) then you minimized the sonically transmissive structures in the wood. As I mentioned, MDF is nicely damped. Two different thickness is a very good idea. By adding the SS bottom, you set up a wildly different resonance scheme, causing an impedance mis-match which serves to damp transmission. And because of the high tensile modulus of the steel, you turned the whole sandwich into an assymetric beam in bending (assuming the SS  is fixed to the MDF.) Materials with very different characteristics and dimensions works wonders.

Cheers

Yes. Most of the property relates to "New Thiel" inventory and their "Aurora" streaming club paraphernalia. The intellectual property will be sold and friendly buyers are welcome to raise their hands along with thielrules. I am in direct contact with the court and will post relevant developments.

prof - When Rob bought the Thiel service department earlier this year, he saw the New Thiel inventory and they said that 3 pair had been sold. Perhaps a few more since then.

New Thiel missed the mark, failed to identify the narrow Thiel niche, and that extant conventional speakers covered the other bases completely. Three was not a typo, it may be incorrect, but it is in the ballpark. Lots of stock remains. The 3rd Avenue towers got 5 stars from a Stereophile reviewer. New Thiel had design and engineering talent and resources to do the job well. Their crossovers are assymetric/variable pitch as required. They may stand up well against other non-coherent offerings on the market.

That stock will be liquidated by the court at probable pennies on the dollar.

In my opinion, there was room for Classic Thiel to improve and grow dramatically within the parameters of the original design thesis. Imagine coherent offerings designed around the last-generation coincident drivers with additional coincident lower midrange, active throughout or an amplified woofer section - higher impedances through improved drivers requiring less compensation - carbon diaphragms with integral voice coils, and so forth and so on. Expand the niche to include high-end recordists and ultra performance luxury goods for the elite and perhaps . . . who knows what. All could be done on the shoulders of Jim's work and in harmony with tens of thousands of extant customer base.

"There's a thousand ways to lose a horse-race". New Thiel knew lots of them. Yesterday I spoke with Bob Brown, an industry veteran who was brought in as COO at the beginning to help navigate the new waters. He remains frustrated and somewhat angry today. His tenure was 2013.

Regarding class D amps: In the mid 80s Jim began working with a very talented design engineer at Vifa who made all our custom-designed-from-the-ground-up drivers at the time. They co-developed the 1000 watt amp for Thiel's first subwoofer. It was class D before B&Os and other patents. At that time switching distortion was significant enough to relegate its use only to deep bass. But, I confidently speculate that Jim would have continued to improve that form of amplification for powered speakers, if Thiel had successfully cultivated internal design talent. Remember that Thiel's first Model A speaker in 1975 was an active 3-way with 3 amps and active xos. We never marketed it due to our internal limitations, and bucking the established passive topology paradigm. Rob says Jim's sub amps sound better than the subsequent BASH-Canada amps which are dubbed x.2, as far as I know. Anyhow I have collected an SS1 and three SS2s. A pair, broken and working, is being evaluated by my new tech partner for possible reverse-engineering to become a repair station for those amps. I'll report more when I learn more.

Similarly, this forum led me to a reverse-engineered schematic for the CS3.5 equalizer. Thank you unsound. The repair shop will also evaluate a borrowed EQ for potential improvements. For starters, I see plenty of caps and resistors plus the phono jacks and switch points which could be upgraded for cleaner signal, short of the fully balanced option. That project is in incubation. I suggest that merely separating the woofer from the upper drivers (as in the CS3) would keep the boosted bass amp out of the detail range. Cleaning up the EQ components would bring life to the woofer. Remember that at 6dB/octave the upper end of the woofer makes contributions through the entire midrange. I suspect that Rob can help with the dual binding posts. Cardas and others supply good jacks. The cut switch may be good or may be upgradable - DIY project for someone here. Also, someone here might report on resistor and/or cap brand or type. Investigation required. I will explore these areas with my tech shop, but input is welcome.

While we're at it, can anyone supply a schematic and/or photos or other information re the 3.5 XO schematic? But, but I'm not working on that model. But, I do love it. Oh the conflicts of the head and heart.

This answer may be a little lame - but here it goes anyhow. Note that I have been away for decades and my personal experience listening to music through a robust system ended with the 3.5 and the following 2.2. Now I have PowerPoints and 2.2s and have heard a few systems when visiting folks. So my experience is rather dated. But it was full-emersion / intense.
So my reason for loving the 3.5 is personal. I was intimately involved in its development, parts sourcing, voicing, manufacture and tweaking. Those 3D baffles were all hand-carved by . . . moi and an assistant, in an amazing hand of man effective method . . . and then tooled on a vintage inverted router (souped up with an 8" diameter forming tool that I designed. You get the picture, full frontal engagement. The fiber (silk, paper) upper drivers do not have the resolution of metal, but they are more forgiving. The woofer was our first polypropylene and we custom designed the cone profile for significant breakup improvement. The caps in the 3.5 were Solen when Solen was using their own best-in-world French film. The micro bypasses were 1% styrene x tin foil. Lots of tweaking and voicing and audiophile sensibilities. Later products lost that last level of detail due to cost / value engineering. There just wasn't enough improvement for the considerable extra cost. Also the newer metal drivers changed the cost / performance equation again and caps suffered - inserting a little vague jangle, in my opinion. I have always locked on bass authenticity from my musician and recording days of youth. The 3.5 sealed bass has a rightness from the very bottom that just doesn't quite bloom with reflex bass. OK, the reflex bass hits all the adjectives better - tight, punchy, etc. and plays more than twice as loud. But the sealed bass of the 3.5 and 5, digging lower than the recording, somehow lights my fire more naturally.

To your question. New Thiel models and generations always mitigated the problems of the previous generation and improved everything they addressed, within the limits of balanced performance. I think a newer, bigger model would almost always please more people more often. Try to audition some.

The gist of my quest is to resurrect some of the old models which can be had at bargain prices and apply lessons and technologies to make them shine brighter than they ever did, and brighter than so many other contenders that don't deliver the full, broad, complete parameters that Thiel brought to light. I hope to conjure those last ineffable nuances of music. Love is a personal thing.

Removing portability opens many options. There's a whole world to explore. Study before you play. Don't be fooled by heavy and hard. Concrete can ring like a bell. Etc. Etc. Etc.

In the metals I would look at magnesium. Similar mass to stiffness as aluminum but higher damping - and much more expensive, of course. I really like the hydrostone / fiber thing I developed and was used on some Hales models. Tension skins on a different & multi) core has lots to recommend it.  Remember that the air (acoustic) resonances are only part of the cabinet vibration problem. The drivers must mechanically couple to the shell for rigid launch. They bounce and propagate energy into the shell. Excellent idea is to oppose each driver with an identical one across the cabinet (as in front and back). Bummer is that you then have a bi-pole radiator instead of a point source, which rules out time & phase alignment. So I am opposing the drivers against a rigid vertical spine up the cabinet back. In a new design, that might be quite large. In extant Thiels it will be 1"x 1" and connected to every brace shelf and the cabinet back to damp it. A rod-strut connects each driver magnet to the rod for cooling and mechanical rigidity.

I am exploring an outer shell for extant cabinets. Take a Thiel cabinet, add a few strategic braces if necessary, set the spine, screed non-hardening Permatex type 2 goo, add wood exterior to taste (3/8 x t&g strips). Viola!  Inner structure keeps high integrity driver mount x acoustically stiff enclosure volume, and outer shell decouples surface resonances for less sound transmission into the room.

No magic bullet. Plenty of room to play. Let us know your success!

As I mentioned, I am searching for a repair solution. A schematic would help greatly if anyone has one.

rule - I would need some prompts to address the lack of braces. The 3.5 development dates from around 1987 - I am unclear exactly when. That's after our first CNC which is when the pierced shelf braces became feasible. I remember front-to-back lumber braces as well as the structural midrange tube enclosure which handle the heavy lifting. The baffle was 3" MDF laminated cross-direction. I remember a lumber transverse cross-brace in the lower area of the sides, but don't remember whether that made it to production. Memories fade. As a side note, the O3 and O3a & b had birch ply cabinets, but I think the CS3 &+ went to MDF. If I saw one I could reconstruct the picture.

The 3.5 pre-dated Stereophile's review analysis (or Thiel's ability to do so) so we don't really know the nature of its resonance behavior.

Note that any replacement driver would necessitate crossover changes since all parameters affect the various resonances which are compensated and corrected in the XO. Also, that driver had some level of Thiel proprietary magnet geometry and so forth; that motor came after we were developing sophistication beyond normal industry offerings. So, let's try to keep the motor and find or create a replacement coil system.

Motors are forever. Cone systems with their surrounds and spiders are for pretty long term precluding damage. It's the voice coil and lead braids which burn out or fatigue. It is possible that we could source that voice coil, on its former or not, and devise a replacement protocol, essentially rebuilding the driver using its own motor and cone system.

What d'ya think?

Great ideas, guys.

jon - I highly recommend checking out USG Hydrostone instead of concrete, for reasons I've previously enumerated.

rule - someone has to pick up the ball, I'll do my part.

andy - good ideas. I suggest that Thiel's present driver technologies are world-class and viable. The likely buyers for those patents are FST, Thiel's Chinese supplier and Meiloon, New Thiel's Taiwanese supplier. Whoever buys those patents would have a big jump on the kinds of products you envision. I would add pro-audio to the mix. Create the music with Thiel monitors.

By the way 24bit x 192kHz digital is now feasible and wow does it work. Also, active crossovers are readily executable in analog for the purist products. A visionary partnering company capable of producing the controlled amps is key. New Thiel also developed some seminal digital technologies. For the right buyer, the Thiel intellectual property is likely to be a huge bargain.

andy - I am way behind schedule due to competing priorities. No one has received anything. But I have all the parts and am making the new boards for beetle to populate. Not yet, pretty soon you'll be getting some early reports.

Your confusion may come from the SE availability. Years ago, Jim developed a Signature Edition for the CS2.4. Most of the upgrades were cabinet related, but there was and is a crossover upgrade which replaces the 2 feed caps (bypassed with styrenes) with 2 ClarityCap SAs, unbypassed. SAs were best of form at the time. The sonic improvement is noticeable. Some folks buy those caps from Rob.

Our upgrades are considerably more extensive and expensive and not yet finalized.

Regarding theater:
I know a universe where the audio dimension of theater adds a profoundly deeper and more complete involvement to the video. As coherence enters that total immersion world, serious magic happens.

(Little known factoid: Thiel worked with IMAX and Skywalker Ranch with hugely promising potential. But, Thiel was not a collaborator in that league. My daughter Dawn grew up inside the company. After Jim's death she said "Thiel Audio wasn't a real company, it existed so that Jim could design speakers and have a life." (Same girl who defined the 'leaf effect'.) Bottom line: Coherence is huge in theater playback and Thiel has technologies and know-how to take that niche over the top. George Lucas et al spend untold resources keeping every ounce of phase/time relationship intact through triple digit track counts and processing paths. I vote that we shortchange ourselves by saying it's OK to scramble it at the end. 

Now, imagine the CS4 (ever notice it's absence?)

I recommend you all find a way to audition a pair of PowerPoints on the ceiling at "normal" listening geometry with subwoofers configured on the floor beneath them driven by your hi-fi amp. That's my studio setup. I expect you to fall in love.

Good work. I should comment that lowering the RF in itself hurts things. However, it is likely that you also decreased the resonances for net gain in quieting the cabinet.

silva - I believe that you made improvements. I am merely addressing the physics of the cause. I suggest googling "speaker enclosure resonances" or somesuch and see what you can learn.

It's quite complex - resonances couple when harmonically linked (like octaves, etc.) But a very big deal is that the lower a resonance occurs the greater the available energy to stimulate it and the longer it will last in time (all else equal) . . . both being detrimental. The dance is to shorten their duration and push them higher where there is less available energy to stimulate.

Lead is mechanically very absorptive and self-dampening. It is the right idea for the right result, but the lowered resonance frequency is an undesired by-product rather than the cause of improvement. 

unsound - What you envision could be done with the SCS, but not the PowerPoint. The SCS has the same driver and XO, but its cabinet is made for 3D space.

The PowerPoint's genius (patented -?-) is its 45° wave launch from a known, unobstructed plane - ceiling is best, walls work OK. The geometry obviates the problems of floor bounce and unknown reflection environment.

Because the ear-brain doesn't differentiate well in the vertical plane, and because the wave-front propagates evenly from the ceiling, the image presents as centered floor to ceiling. Add a woofer beneath it and the image locks into 3D space very believably.

I am accumulating some original PowerPoints for parts. The PP1.2 has an all aluminum enclosure which makes a surprising improvement.

holco - you are welcome to share which capacitors you changed and why and your specific listening experience.

harry - regarding 2 vs 3. The target frequency response is identical, including the upper bass x midrange. I suspect the difference you hear is based on power response in the room. The 3s go deeper and move more air. The 3s have a larger diameter midrange with a lower crosspoint, so room fill can sometimes benefit. Also, the top end of the 10" woofer breaks up more than the 8" in the 2 - that breakup adding more "meat" in the lower midrange. Some folks like that even though Jim considered it a flaw - less than accurate. Both the 2 and 3 have a first-order hand-off from woofer to midrange.

The bass rightness I was referencing is the deep bass. The sealed bottom end of the 3.5 and 5 allow a 12dB/octave bottom end which makes a gradual phase shift as it rolls off, supplying a natural sounding foundation.

The ported / reflex models (1, 2s and news) add a pole for 18dB, and much steeper below the port tuning for significant phase shift at the bottom. (Remember that the industry judges this phase shift as of no consequence. Most modern products and subwoofer crossovers produce 4th order / 24dB/octave slopes.) Nonetheless, real acoustic instruments played in live spaces and recorded carefully do benefit, in my experience and opinion, from more shallow rolloff with greater phase integrity. There's something less hi-fi and more real.

Yes, I would say the coax feed cap upgrade removes a slight coarse veil over the mid- upper frequencies. The MRA-12 Mills resistors add some ease. The drivers / cabinet have enough inherent quality to support considerably better XO parts. Keep up the good work.

Regarding wire. I suggest leaving it as is. Thiel pioneered great wire before wire was a thing. I believe I alluded some details in a previous post earlier this year. Summary is that Thiel hookup and coil wire is world-class and most attempted "upgrades" would in fact downgrade the wire. Our first wire in the late 70 resulted from cousin Teddy's recommendation for ITT aerospace 6-9s pure long crystal in teflon twisted 2/inch right-hand. ITT abandoned the wire business after aerospace contracted and similar wire is now made by others. Thiel wire is all certified at least 4-9s+ twisted in teflon. The coils are of the same wire and the gauges are optimized per use. Don't think that larger diameter, lower gauge coil wire is superior. It is not. The exception is late-stage 3.7 XOs from China with CYC caps which have Chinese wire that claims the same specs, but which can't be verified and industry insiders doubt its purity.

Our upcoming upgrades will add gauge to some input and driver runs, and replace series-feed coils with foil for extant wire. All certified 4-9s+ best-of-form. 

harry - big southern mansions have their charms. After my time at Thiel Audio I did some installation consulting work which included Bluegrass Horse Farm Manor Houses. The sonic potential of those rooms in those dwellings is a wonder to behold. 1982 would have been the development of the CS3, which was our game-changing breakthrough product. I suspect that Larry Staples (hello Larry) arranged that listening session. Sounds like a focusing experience for you.

holco - beetlemania and I have combined our resistor inquiry and included some things I know about brands from history. We are convinced that the MRA-12s are our first choice. Part of that choice is their patented technologies and the Aryton-Perry non-inductive design, since I am paying particular attention to electromagnetic eddy currents. There are higher-tech resistors out there at multiples of the cost. My interest is in honest engineering-based solutions without cross-talk into sonic editorialism. Jeff at Sonic Craft and other knowledgeable operatives have been quite helpful in wading through the brands and claims and factors involved. You and all here are welcome to construct a comparative test for our collective learning.

Hed - the EQ itself is not the problem; the problem is that producing that much bass down to 40 or 20 Hz takes lots of muscle. No accident that subwoofers sport kilowatt amps! The eq presents a benign, non-reactive amp load which is easier to drive than the more reactive passive radiator loading of newer Thiel products. In Frank's small room, his amp might fill the bill.

My approach is to identify and model exactly what each coil is doing and optimize toward the ideal behavior. That is different than choosing which sound I might prefer. Coils are quite technical in how their electromagnetic fields and therefore transient behavior develops. A primary source of distortion comes from wire squirm which smears the signal. Thiel used a dipped and baked coil round wire. Foil further reduces mechanical squirm as well as develops tighter field effects. They are better inductors with fewer side-effects. They also cost more. They also remove an (artificial) roundness and excitement caused by the euphonic effects noted above.

Due to practical conciderations, I am investigating replacing series feed inductors at 18 gauge and larger with foil. Smaller gauge foil equivalents are not readily available and Jim used 22 gauge coils in shunts specifically for the fine-tunable inductive resistance they provided. (Bigger is not always better.) Those resonance circuits do not carry current and are explicitly tuned to driver and overall circuit parameters. Besides, they are world-class high purity copper in custom dimensions and oven-baked. Hard to improve.

J.A. - such considerations were an integral part of our development projects and accumulated knowledge. Wire effects are subtle and complex, but far from trivial. Yes, I agree that part of Jim's genius was to pay close attention to everything and pay no attention to conventional wisdom such as "bigger coils are better".

Of probable interest here is the possibly little-known fact that we never performed competitive evaluation; there was never another-brand speaker in our lab or listening room. Jim considered such an idea as at best a use of time which we couldn't afford, and at worst a distraction from the work at hand, which was understanding and solving as much as possible from a first-principle approach. The same went for materials and cabinet concepts and construction schemes and production methods. We did original work.

Considerable feedback came from the press, dealers and customers. But such feedback was on the conversational level. The real work included Finite Element Analysis, fine-tuning the lab and its equipment, building a world-class listening room, implementing an anechoic environment on the roof-top, experimenting with leading-edge materials and the excruciating iterative process of developing complex electromagnetic / acoustic / electronic systems to get the job done. And, of course, pruning each model to fit our market niche of affordable high-performance musical tools.

gas - when the dust settles and we have accounted for the differences in effective series resistance of the new caps, coils and so forth, the frequency balance and phase response should be nearly identical to the original design. The improvements will be in the areas of ease, definition, clarity, perceived dynamic range from quieting background noise and so forth. We may lift a veil and sweeten things a little. But the tonal balance, and so forth will not substantially change.

It bears saying that we may well be improving the Thiel strong suits.
We cannot reduce the amplifier demands because the low impedance load is necessitated by the various correction circuits to cover the wide bandwidth of each driver. But, we may (if the customer wants) add a bi-amp option. When done right with the same amp for highs and lows (4 identical channels), each amplifier has a much easier time driving its load, plus any distress is sequestered to its own range. Audible improvements are certain. But additional amplification and cable requirements add expense.

Nothing comes without cost. Our intent is to jump performance league with a modest enough cost burden to justify the expenditure.

FYI: Bryston was a long-term collaborator with Thiel. The cubed (3) series is said to be strong with Thiel. The CES introduction of the 3.7 was with Bryston to high acclaim. I don't know their line, but your reported disappointment may be because of too little power / current.

Jimi - I recommend restoring the CS1.2s as a very fine little speaker. I use the CS1.2 drivers in an 02 cabinet to compare components,wire and cabinet treatments. Rob at Coherent Source Service has parts. I think you'll be glad you did.

CoherentSource Service will probably also have carpet spikes, pointed and long enough to penetrate carpet. 

jon - indeed a digital device and/or passive preamp is often not robust enough to drive the particular cables and/or input section of the chosen power amp. Active preamps are designed for that chore to ascertain that the signal voltage swings will indeed arrive at the power amp's inputs. Some cables are harder to drive than others and long runs especially can deplete signal vigor.

holco - good judgements, upgrading the signal paths.
Coupla comments. Depending on your serial number and if it is Lexington-made and whether or not it is on masonite or a PCB, the .72mH woofer coil may not be the correct value. Details would help evaluate those particulars. Likewise, if it is Lex-made, the coil will look very tight and well-wound. If so, that is an ERSE coil which, in my opinion, will not be improved by the Jantzen, unless the J is foil.

On the coax board, you are ignoring the parallel feed in the first position. I suggest investigating the 43uF cap and its 16ohm series resistor. At minimum replace that 16ohm with MRA-12s.

Beetlemania might have comments on the other resistors.

Andy - couple of quick thoughts. 

1) Correct. A primary difference between the model 2 and model 3 is bass quantity and extension. The model 3 has a 10" woofer with longer excursion and larger passive to produce more than 1.75X the bass of the 2.
2) Jim does not add the customary underdamped bass hump, so Thiel bass is honest.

oblgny - I am collecting parts for my ongoing upgrade project. The 3.5 is now on the list due to interest. I would like your drivers, etc. Where are you located?

Tom

You guys have been busy while I've been away . . .

holco - further investigation has shown that the OX seems to have 3 types. Lex 1 is PtP on Masonite with Solen (etc.) and first-rate ERSE / Jantzen coils. Lex 2 has same high quality components on PCBs. That seems to be what you have. Those coils are good and they match your boards. Don't swap them for the schematic you posted because those values were for masonite.

You can physically measure your coils with an impedance meter if changing coils. But if you change gauge, then the DC resistance also changes which must be compensated in the resistor values. Since you have good coils, I suggest the easiest way out for now is to stick with them.

Go with the Mills resistors. Sonic Craft has New Old Stock Mills, which is best. As beetle said, we're working on custom values, etc. and hope to have news to report by end of 2018. Keep us posted.

beetle - good advice.
Regarding your 2.4SEs, the evidence suggests that the original boards made in Lexington were Masonite PtPoint. Perhaps after Jim's death (2009) Lexington changed to PCBs with the same traditional parts. (I don't know when that change happened, but ERSE had those PCBs for sourcing parts before China. Many folks say the PCB per se does not reduce sound quality. Your SEs came from after XOs were sourced from FST in China with CYC parts, including some downgrades from Propylene to Polyester (T). It seems that those parts were supposed to be clones of the extant parts, but sources say the verification is weak or missing. We are assuming the unverified Chinese parts are of lesser quality. Despite the unknown parts quality, including wire alloy, the coils are not tightly wound, which results in squirm losses.

Regarding sandcast resistors, you have heard and I support upgrading to Mills. However, I know that the Lex sandcasts were ERSE which we know to be best of form. But resistors are great bang for buck, so why not upgrade while you're in there. 

31 - regarding your CS5 vs 7.2 query: my considerations would be more strategic than any one aspect of performance. That performance aspect is the type of bass alignment where the CS5 is sealed and therefore produces more coherent bass as it rolls off the bottom end. However, both products go deeper than most program material, plus the ear is least sensitive to time/phase in the bass, so I would attribute minor significance to that aspect. (Albeit, I personally like the CS5 bass, which isn't the point.) On to amplifiers: the deep bass of the CS5 drops well below 2 ohms and therefore requires significant and specialized amplification to drive it well. If I had a pair of 5s, I would split the bass from the upper frequencies for easier amplification.
Repairability: the CS5 lower drivers were Kevlar by ? (I may remember some day.) I don't know their availability or whether Rob at Coherent Source Service can rebuild them. Check on that. But, those were 1988 specialty drivers from a model that sold less than a thousand pair. So caution is advised. The tweeter is Thiel and also used in the 2.2 and 3.6, so it's available. Note that the 5 is a 5-way, so the integration of the drivers yields a very smooth frequency response and reduces excursions required of other Thiel products with fewer drivers.

The CS7.2 uses Thiel drivers and I believe Rob can repair those. He says there are virtually no problems with the 7.2.

As a general observation, Thiel approached product design as a process of incremental improvement. Each design stood on the shoulders of everything that proceeded it. I think of the 7.2 as a fully modern, realized design whereas the 5 as an evolutionary step. As I've mentioned before, I find significant flaws in the 5's use of bucket brigade delay for the 2 midrange drivers. That's a lot of circuitry for the signal to navigate.

FYI: I have heard the CS5s more than the 7.2 and have heard them both perform gloriously with the Krell FPB-600. Reviews suggest that any amp short of that or similar full-out muscle is likely to whimp-out under the CS5 demands.

All that aside, if you are inclined toward collecting icons, the CS5 is unique in Thiel's history. Its development was spurred by demand, especially from Japan, for Jim's ultimate statement. In fact I judge that he pulled some significant punches; it could have been more ultimate, even at that time, had he allowed a budget of $15 to $20K rather than the under $10K at the top of his psychic tolerance.

Now, a flight of fancy. In my dreams I would remake the CS5 to a higher standard. An easy fix is a cabinet resonance or two. Another fix is to break out the bass from the upper drivers for separate amplification. Now we're in no-problem performance land. Furthermore, let's upgrade passive parts for further significant improvement. Then the serious business - reshape the cast marble baffle for physically proper driver placement rather than electronic compensation. I remember the first 100 pairs were Brazilian Rosewood from my private stash - considered the finest cabinet wood in the world. I am unaware of any other fully 5-way coherent source transducer out there.

Such an undertaking would be a collector's edition re-imagined re-issue of the iconic and obscure CS5 from 30 years past. It would fly.

oblgny - I reside in Lake Grove,  N. Y.   Very close to Stony Brook University.  Again, free for pickup only.
My travels don't take me to Long Island. Thanks anyhow.

Pops - most of my Lexington time was on Georgetown Road, where we started Thiel Audio,  around the corner from the Kentucky Horse Park, where I rode my bike nearly every decent day. Secretariat was exemplary of his Bold Ruler line in being interactive and playful. His grandson that I bred added a dash of mischief. He would do things like untie the girls' shoe laces or pick them up from their belt, and could unlatch a safety gate latch. Lovely and capable family of horses.

mallikh67 - Have you worked through standard troubleshooting practice? Does the bad sound persist with another speaker, or headphones? If only your 2.2s are at fault, is it in both speakers? Try to narrow the problem down to an individual problem and then we can help you mitigate it.

unsound - there is always more than is obvious behind product development decisions. As I have alluded previously, the company supported Jim's design aspirations. Kathy, the marketing director, supported Jim's design aspirations. Drivers and electronics were Jim's primary interest. Cabinets played a supporting role as a necessity.

Developing that baffle was a major undertaking. I made the mahogany patterns with late night CNC time stolen from day time production requirements. The convoluted baffle would have been a far larger undertaking, in design, pattern-making, casting and production finishing. The company did not have enough confidence in the product's projected sales to invest in that larger baffle cost. C'est la vie.

Such dynamics are the stuff of small, self funded entrepreneurial companies. Given the luxury of capital to invest, including more staff, and the confidence that a market would emerge, I believe that flagship product(s) could have been developed which would have leap-frogged Thiel's market niche.

The cabinets are indeed serious functional elegance. Thanks for that.

No apologies required or sought. I am merely shedding light on the back-story because some of you guys are interested. No scab, really, just some story that usually gets politely ignored.

Prof - yes, the CS6 and 1.5 were my last products before leaving in 1995. The machinery and manufacture processes all continued until the sale in 2012 when the factory was dismantled.

That Macassar Ebony had no stain, that was the natural color of that South Seas species. 

Also, No. It's not that simple.