Virtual system - I posted some photos of my unique setup.
Showing 50 responses by tomthiel
Hello folks; just letting you know that no grass is growing under my feet. I am making good strides toward first test samples. The CS2 2 development has dropped back till autumn and the PowerPoint and CS2.4 are now the first-tier projects. The big bundle of ClarityCaps is shipping in a week with lots of trial samples including the custom CSA-160 volt cap. This cap utilizes CSA technology with a 160 volt film to be small enough to fit inboard layouts. The outboard XO option seems to be dropping out via better inboard solutions. On another front, I learned that Rob got the Klippel Analysis rig from ThielNashville, so it will likely find a place in this development project. World-class tool. Also, I have ordered the new, long-awaited update hardware for my Metric Halo SpectraFoo analysis rig which now sports 32bit x 192kHz ultra clean conversion and 64bit real-time multi-chip processing. This rig is used for recording and concert hall analysis, but applies well to speaker development. I am learning it. Also, my Classé amps return this weekend from the rehab shop. While re-capping, I also upgraded the caps and added some bypasses. DR6 (late) preamp and 2x DR9 power amps. (100 w @ 8 ohms, 200 @ 4, 400 @ 2 ohms, one amp for each channel. They're old 1990 gear, but I know them well, and love what they do. Additionally, some very knowledgeable and experienced folks are contributing to the project. By year's end I hope to have a version or two of upgrade crossovers for the 2.4 and PowerPoint. Stay tuned. Tom |
Dn, sounds like a wonderful setup. I don't know those amps, but can offer some general thoughts. This group might offer opinions of the Lyngdorf. Read between the lines, it might be class D and/or digital amplification, which you would have to sonically evaluate yourself: Get return privileges. It doubles from 200wpc@8 to 400wpc@4. FIND OUT how it behaves into 2 ohms. If at least 600wpc into 2 ohms, it will not distort under a Thiel load. The Stellar M700 reads well too. Same suggestion regarding 2 ohm performance of at least triple the 8 ohm performance is required. Regarding the PowerPoints: your dealer might help you through the power requirement equation. Your room and listening levels matter a lot. However, don't underestimate the PP and MCS. I am using PPs for my near-field and mastering monitors, and they are my first XO upgrade project. They and the MCS are in the same hi-end league as the floorstanders. If in doubt, and if you can, lavish amplification on them. This is a long-shot including some hearsay. Jim began working with a very talented circuit designer at Vifa/Denmark in the mid 80s on what became the SmartSub amps. I have heard that he is or has worked with Lyngdorf. If true, that's a very good indication of excellence. They look excellent anyhow! Sorry that I've never heard them. I love PS audio, but have not heard those amps either. I notice that you use the plural for the SS2s. Congratulations. I know from experience that there is enough location information in their signal to warrant careful positioning. Try putting them near the primary speakers and pay attention to arrival times. Avoid putting them out of sight or in a corner as is often done. Even though bass waves are long and therefore near omnidirectional, their leading edge transient component is intact and important. Everything matters. Keep us posted. |
Beetle and Rob - I got an answer from RG that all pre-sale (2013) CS2.7s were made in Lexington, with PtP, parts like Rob's, etc. but with standard 18 solid twist hookup wire. We think this pair was a trial at New Thiel and we're investigating whether any parts values were changed to accommodate the different wire configuration. |
For the record, the CS1.6 is on my upgrade list, as it has sophisticated, all-Thiel drivers and represents the pinnacle of the 1-series (pre-coincident driver) and there were thousands of pairs produced. There has been no expressed interest in the 1.6 upgrade, nor do I have a pair. So it is on the back burner, but simmering nonetheless. |
Prof, as a technical note, that wavy driver behaves better than any of the previous cone drivers as well as supports the tweeter wave-form launch very nicely. Therefore the direct series crossover path is simpler than any previous Thiel driver. Jim had always corrected for slope anomalies in all the drivers, which puts additional components in the signal path. In the x.7 models, fewer components are required because the fundamental driver performance is so good. Part of what you hear is the thrill of nothingness. I am glad to hear your enthusiasm. My present work on prior models is to refine the required passive components to minimize their contribution to the signal. Most of the parts are on-hand for my PowerPoint experiments which will also inform beetlemania's CS2.4SE upgrade. We can't brag about vaporware, but I am pretty excited about the potential improvements. |
Prof, you're right about the "hollowing effect". Phase cancellation is a central problem making first-order implementation very treacherous. The output of 2 combined drivers produces a lobe which cancels or augments output when the "listener" is not on the proper vertical axis to integrate the two signals correctly. The ideal way to implement phase coherence requires a true point source, as in the coincident upper drivers of more recent Thiel designs. Short of that coincident solution, the closer the driver spacing the better; notice the touching/clipped perimeters of Thiel mid-tweeters. Plus, the crossover frequency must be as low as possible because the beaming of the upper end of the larger (lower) driver interferes with and is discontinuous with the radiation pattern of the lower end of the upper (higher) driver, making for a room power response different from the on-axis driver response. To wit: Thiel upper crossovers are very low, lower than considered feasible by most; Thiel tweeters cross in around 3KHz. Such a bold XO frequency requires a tweeter that behaves all the way down, below 100 Hz. At these low frequencies there are bound to be resonances which would destroy the tweeter if not mitigated with notch filters, which are themselves costly. Broadband drivers are a design feat as well as a management challenge. Most experts, including those at the "New Thiel" deem the task "impossible". You are correct: the coincident drivers do a better job than the older multi-driver solutions. But those multi-drivers are themselves extremely sophisticated and allow very low crossover points and both physical and electronic resonance control. Part of Thiel's low impedance (which we love to hate) is that each resonance correction lowers the system impedance, and the global system is really not serviceable until all resonances are effectively eliminated. As I perform my XO upgrade investigations, I am continually surprised how good all these Thiel drivers are. As an example, the non-resonant bandwidth of the 3.6 midrange spans 7 octaves, with similarly stellar performance from every driver in the stable. Part of my decision to reach back no farther than the CS2 2, is that previous drivers were merely modifications of off-the-shelf units from Dynaudio, Seas, and so forth. Newer drivers incorporate new sophisticated technologies toward success in first-order systems. Back to the lab. |
Ob, the Mac you cite is current-limited by definition. If it doesn't double from 8 to 4, that means there is insufficient power to do so, whatever combination that power comes from storage capacitors (instantaneous) or transformers (sustained). Beyond that doubling, the power into 2 ohms is most germane. Many Thiel speakers dip near 2 ohms over extended ranges. You need at least 3x the 8-ohm power into 2 ohms and preferably a second doubling (4x). Now, it's possible that you listen at low volume in a small, lively playback room . . . and less power is adequate . . . but, you're playing with fire if you ever wanted to crank it up. If I were looking, I would not put that Mac amp on my list due to self-admitted current limitation. |
Jon, I accept your perspective. The doubling itself is less important than is having the required power at 2 ohms. So, the most important factor is how well the amp performs at 2 ohms, which is the critical impedance of Thiel designs. That said, there are some matters of subtlety. When an amp is depleting its powersupply and therefore recharging capacitors in real musical time, the signal waveforms lose their integrity. Much of that "wow" factor that thrills Thiel owners comes from that very wavefront integrity. Ron, that PS amp is highly likely to do what you need. TM, I don't have such a list, being out of the loop for a long time. The amps that Jim used all did it. Big Krell, Levinson, I don't remember what else. The amp I use is a pair of Classé DR9s at Stereo: 100, 200, 400 and Mono: 400, 800 and 1200 watts per channel. I notice real improvement from one amp to two, and the one amp has 400 wpc into 2 ohms. My room is not huge and I listen at moderate levels. My main point is that many of the criticisms leveled at Thiel speakers are attributable to inadequate power and/or cable interactions. In manufacturing circles it is impolite to call out such perspectives. But, source signal integrity is a huge part of the picture. I'll also note that global negative feedback messes with phase integrity. One reason Ayre products sound so good on Thiel is that they both maintain phase integrity to a high degree. I would enjoy seeing the amp list that you guys might put together. |
Ob, Used gear offers great opportunity, with some cautions. Most gear, including Thiel speakers use electrolytic capacitors. Even very high quality E caps have a projected life of 25 to 40 years. Rob says he has never seen a Thiel outright cap failure, but nonetheless, their performance deteriorates with time, causing migration of crossover points and slopes, or a vast array of parameters in amps. Some old gear becomes non-inspiring due to this performance deterioration. So, you might pick up used, under-performing gear on the cheap and have it re-capped for like-new performance, or upgraded for better than new performance. In your room, you might consider Conrad-Johnson or other tube gear. Pay attention to its damping factor and bass management. Tubes can be very sweet. Best of luck with your system rebuilding. |
Thos, Replacement models are generated to push the art, grow the company and generate new interest. The old products aren't really old or defective, in fact in many ways Thiel designs out-do many contemporary products. The 2.3 and 2.4 are conceptually the same product, the product that Jim wanted to design in the late 1970s when we figured out that phase coherence was our passion. We knew from the beginning that a coincident driver was the end-game, but we were an internally capitalized skunk-works and it took all that elapsed time to develop the 2.3 coax. The 2.4 was the next generation, with a smaller diameter midrange. But the woofer, crossovers and cabinets are very similar, as is the sound quality. Many folks think the 2.3 presented a more cohesive projection than the 3.6, due to the coax driver. FYI, the 2-series always followed the next generation of the 3-series and benefited from upper-model knowledge and technologies. The 2.3 followed the 3.6.1 (internal nomenclature) which included some driver upgrades. I concur with Ron that you might best keep loving your 2.3s. One of these days when you are ready, a higher-order or newer model might be in order. And when the time is right, I hope to provide significant sonic upgrades to several of the classic models, applying newer component technologies. I don't know when or why Thiel abandoned its first tag-line . . . there was some legal bullying . . . and Linn added a word to our motto as an upstage tactic. From the 03 in 1978 until the mid 80s our motto was "For the Love of Music". Carry on. |
thielrules, I can't read your attached files nor do I have a 3.5 schematic. Although the role of caps is to block low frequencies and the woofer circuit doesn't have caps in its series feed, there are some smaller values in the shunt circuits which could lessen woofer output via passing more bass to ground. But, my recollection (fairly dim) is that all those 3.5 shunts were mylar film caps, which have an indefinite life. If an electrolytic cap has actually failed, it will usually show leaking goo. E caps are labeled "NP" for non-polar. This gig is something that Rob Gillum may help you solve. dsper, cap failure is a gradual thing until they might blow up, which should cause considerable sonic change. Gradual failing can be heard as frequency shifts in the xo regions, especially at high power and possibly noises like gurgling or sloshing water. As I mentioned, Rob has not seen any cap failures, Thiel used very high quality caps. The rule of thumb is 20 to 40 years of heavy use plus or minus 20. I'm not being facetious; if a cap is going to fail, it often fails in the short term. And if it's going to last, it generally lasts a long time. Yours didn't fail early, so they're probably good for a few decades. Rob at Coherent Source Service could replace your E caps or talk you through doing it yourself. Also, we are developing crossover upgrades using 21st century passive parts. Our upgrades-in-development will not use electrolytic caps to address the fact that those interested enough to resurrect and upgrade classic Thiel speakers would be well served to never have to worry about caps again. The polypropylene replacements cost an order of magnitude more than stock NPs. We are exploring cost-effective solutions at two levels; it's a little early to announce brands, but we are making progress. |
thielrules, troubleshooting is a hands-on undertaking. I'll speculate a little, but the problem needs actual triage, both speakers and the eq. The deep bass looks like the green channel has the eq and the red channel does not: textbook 12dB/octave rolloff. The red channel from 80 to 300 might be a woofer problem or the bottom end of the midrange phase-lagging causing that periodic (cancellation?) pattern. Of special interest is the combined blue output from 250 to 1K is lower than either speaker. This graph implies phase / polarity issues in the red speaker or the eq. I suggest you take out the eq, change cables, measure each speaker in the identical room location, swap amp channels, etc. to gather meaningful troubleshooting data. Perhaps you can arrange an appointment with Rob to send him your preliminary data beforehand and take your speakers when he's ready to put them on the bench. |
I don't know that gear, but I have some relevant experience. First, with regenerated power with good specs and enough storage, it doesn't need to be reconditioned. Proper regeneration is proper power. When mixing and mastering music and burning first-item gold master CDs, I have always waited until the electricity cleaned up around 2:AM to noticeable sonic improvement over daytime electricity, even with industrial line filters. Clean power is a big deal. I don't know the current products or PS' offerings, but it is worth paying attention to, in my opinion. |
andy2 - perhaps cable conductive stabilization is beyond the scope of this forum. But I would nonetheless like to post how such considerations entered the awareness in the development of Thiel Audio's peculiar relationship to the role of the loudspeaker as a precision playback device. Discomfort is a requirement for growth. I may have mentioned cousin Teddy awhile back, but to recap: he helped us appreciate the much larger world of science beyond armchair comfort with known basic concepts. Our cousin Ted Lyon was a senior theoretical physicist at General Electric Jet Engines in Cincinnati. Teddy was smart. Teddy also took an early interest in what we were doing with speakers. He first came to visit in the throes of prototyping the O3 to be phase coherent or not. (. . . to be, that is the question - sorry, I'm involved with a mentoring Shakespeare Company and couldn't help myself.) Our problem was that myriad sonic problems beyond our understanding invaded the sound-scape when phase coherent (first order) and vanished when not phase coherent (3rd order). We knew we had a tiger by the tail because musical nuance would also be higher as a result of such functioning when phase coherent. Teddy listened intently and nudged our conversation ever more far-fetched and speculative, far beyond what we as 20 somethings could comprehend. But Teddy was a patient teacher. After many hours and near exasperation he simply began to talk - for perhaps a half-hour. We learned that NASA had faced such exasperation in its long-field (millions of miles) aerospace avionics - instrumentation / transceiver communication. The communication and especially navigation-positional protocols depended on transient waveform integrity. He said he was hearing artifacts in our phase coherent prototype similar to what he heard in his avionics prototyping work. He offered 3 avenues for our exploration. Replace steel driver baskets with non-magnetic alloys to reduce eddy currents, upgrade the copper wire, and never evaluate "young" components. It bears noting that Jon Dahlquist of DQ-10 fame came from aerospace and that he used 18ga twisted pair solid copper in teflon when the rest of the industry used ordinary stranded wire. Hmmm. My summer of 1977 revolved around identifying and sourcing wire and other components that ended up being 99.9999% pure, long crystal, low oxygen, etc. in teflon or varnish from ITT, developed for NASA. As far as I know, we introduced "wire" to the audio industry, or at least we didn't hear about "wire" from anyone else beyond noting Dahlquist's unusual choice. Let's proceed to "never evaluate young components", which included wire, caps, resistors - everything. Teddy said that the grain boundaries in metals conducted non-linearly with frequency, current, voltage, temperature and so forth. Boundary effect phase discrepancies were certainly measurable and in fact the subject of considerable engineering effort - thus the ITT wire. Furthermore, those grain boundaries could (speculatively at that time) exhibit migratory properties, since ionic flow tends to push and form metallic micro-structures, which are somewhat mobile. Our breakthrough involved learning how many things that we did not understand could have effects on what we were hearing. So I suspend my own tyranny of intellect when relating to sonic phenomena. The gig goes: hear it, label some aspects in non-value terms (never use better or worse), isolate some aspect, address that aspect from different angles, note the outcomes. Choose a couple to measure, find or develop some meaningful measurement criteria, decide which path to pursue. Nowhere in this decision tree must we prove to anyone the validity of our reasoning or selection. The decisions must lead toward improvement by our own criteria and that of our customers, including reviewers and peers. It would take lifetimes to explicate and/or prove such matters, which would not promote designing speakers or building a company. So, I claim no proof for much of anything. I am merely stating my bias toward creative intuitive action. Here is a link to an abstract chosen off the web. It represents the kind of stuff I read that summer and informed my conversations with Teddy. I doubt that we would have jumped into the deep end without Teddy's input. Jim was thoroughly committed to the scientific method with a deep dose of skepticism. Teddy was a serious PhD who encouraged and insisted on exploring beyond the limits of understanding. He gave me significant encouragement and Jim the basic permission he needed to trust his ears. https://www-pub.iaea.org/mtcd/meetings/PDFplus/fus Enjoy. Tom |
I don't know models, but I know that Jim used Goertz flat cables in the development lab as his reference. Jim's perspective was that cable performance is cable business. He felt that Goertz performs cable business the best he found. That may seem obvious, but it is not commonly practiced. Many designers choose to ameliorate or mitigate various upstream problems . . . pull punches, soften, generalize, etc. to keep things polite, avoid offense. Jim, however, was committed to high resolution translation of the signal at the input terminals. Some folks appreciate that. Thiel assessed that to be the purest statement of the art. http://www.bridgeportmagnetics.com/contents/en-us/d62_MI_AG_Speaker_Cables.html |
The most common oddity from a woofer is the bottoming of the suspension: the spider hits the basket plate. Thiel's short coil is out of harm's way from bottoming on the back plate. The voice coil adhesive is high-temperature. So, unless you overheat to the point of burning out the voice coil, most noises are transient and do no harm. However, one can hope that your experience might help you set future limits. Drivers do burn out. |
Ron, shorter cables are always better than long. But parallel (not twisted) conductors like Goertz are inherently capacitive. If there is a problem depends on the amp. But Thiel's low-impedance results in more current being drawn through the cable, so any capacitance problem is likely to be exacerbated. I suggest using the Zobel filter. |
Ron, all of your failures sound like overheated voice coils, either delaminating or in the case of humming along, being loose enough to squirm on the former. The cause is most probably being fed a distorted / semi-clipped signal. The first driver was probably good out of the box and also failed via underpowering. |
Ron, your listening levels and amplification are very appropriate / safe. So something else is going on, perhaps bad drivers, but sometimes cables can induce amp misbehavior that adds stealth distortion. I can't troubleshoot, but I will try to find out the service history of those drivers. Prof, that cliché is true. Tube amps clip softly, without much hash, and therefore are far more gentle on speakers when driven hard than most solid state designs. As an aside, I have played hundreds of Thiel speakers for thousands of hours, often in very demanding and loud situations, and have never blown a driver . . . I have also seen hundreds of "blown" Thiel drivers and in the vast majority of cases the voice coils are burned, which is (practically speaking) only caused by distortion (the driver will produce ear-piercing volume without damage given a clean signal.) Those burned coils are user-damage, but Thiel generally covered them ONCE under warranty as good will. Legitimate manufacturing defects include subtly mis-routed lead wires, magnet-position and/or suspension mis-alignment or glue failure, etc. Those are generally batch problems and result in instant failure (broken lead) or immediate voice coil rubbing. If it's defective it will fail quickly. Part of my upgrade project is thermal management, which is generally ignored in hi-fi, but is very important in sound reinforcement and pro-audio. Heat is a real enemy. It can shift crosspoints radically, which adds substantial stress to the system. A 3KHz tweeter crosspoint can migrate to 2500, admitting damaging low frequency signal to a driver which is vulnerable due to the concurrent high-power situation. So, I am mounting all resistors in heat-sinks mounted to buss bars to dissipate waste crossover heat. My CS2 2 workhorse is getting an aluminum tube spine up the back of the cabinet to sink those buss bars as well as mounting an aluminum tube from the back of each driver which conducts heat as well as provides more mechanical stiffness to the driver mount. I am puzzled that Jim abandoned aluminum voice coil formers for nomex; probably for lower mass and smoother high frequency extension. But thermal conduction was traded off. This note is just to let you know where my investigations are leading. BTW, I have a pretty big pile of foil coils, resistors, wire and caps, plus the Clarity Caps are finally on their way. I'm especially excited about a CC custom polypropylene cap we have developed to replace all the electrolytic caps. I'll be comparing the new CC to the ERSE PulseX Polypropylenes in those applications. Either way is a big step up from the present electrolytics in the shunts. Progress is being made. |
I'll try to find out something from Rob about the 2.7 vs 3.7 failures. Same driver. Thieliste, good investigation. In the day Thiel, especially the CS5, was very successful in Japan. Luxman amps were the clear choice in ultra high-end systems. They are seriously good. The amp you reference is still doubling into 1 ohm, and their engineering choices are musical (most Japanese amps are not, in my opinion.) |
Ron, I don't have enough history to comment on your situation. I am far away and out of touch and coalescing probabilities from sketchy
evidence. The detailed lab records are MIA and most of what we know is in
Rob's memory. I am gradually back-engineering XOs and timelines from photos, previous suppliers input and whatever Rob sends me. Rob says that Thiel built all 2.7s in Lexington under his supervision, so he knows far more than I do. The schematic that I created from an engineering layout shows the tweeter with a parallel pair (20+3) netting 1.3 ohms in the tweeter series feed. It is possible that value was adjusted over time. Rob would know his rationale and I do not. |
For the record regarding product development. Jim was a sole developer with support from Gary, Rob, Kathy, etc. The 3.7 was his last product. The 2.7 was a spin-off, as the 2 series had always been. It took $6 figures to develop that product with outside engineering resources, primarily a Canadian consulting design engineering firm working with the Canadian National Research Institute. I understand that Phil was one of the designer-candidates scouted for possible contribution. |
Jeff, the point is that such a system (CSnext) required far more horsepower than available from an individual who may have played some minor role. Creating a design team after Jim's departure was a huge undertaking. The 2.7, 1.7 and MCS (#next) were developed at costs so great that the new owners abandoned that path. |
Prof, I had seen that quote from Phil here before. The story from Thiel doesn't really match all that well. Regarding the 2dB/decade, I don't really see that in my perusal of comparative test results. I wish that Stereophile had reviewed the 2.7 so that we might compare John
Atkinson's rigorous measurements between the two speakers. In the
absence of that direct comparison, I am postulating reasons to explain
what I have heard. Please note that I have not done any serious forensic engineering on these two models, since I am addressing older models first. As I have mentioned, I was at the Thiel factory when the final 2.7 arrived for confirmation, and we heard it compared repeatedly to the 3.7 with a couple different amps in the room that I knew extremely well, having built it in the late 80s. My opinion is that the two speakers share all their textbook and test stuff, but the 3.7 is a higher resolution device due to many particulars. That higher resolution comes at a price of revealing everything: all that stuff of amp and cable and RF and room mode interaction and on and on as audiophiles are wont to do. Among the reasons the 2.7 might be easier to take is that it has a high count of large electrolytic caps in its signal path. That statement might sound backwards, but please hear me out. E caps serve to extend the time signature, a distortion that provides a more forgiving presentation. Similarly, the 2.7 baffle is made of MDFiberboard, which is softer and absorbs a little of the leading-edge transient. Also, the more "normal" 8" woofer of the 2.7 will flex a little more, providing a slight sonic cushion. Here comes another controversial statement, one that I have noticed consistently over many aspects of music-making and playback over nearly 5 decades. We humans are more comfortable with the known. We use our history as our benchmark. Those tendencies translate into comfort with distortion, as long as it is low order and musically plausible. Note that most of the record-making craft centers on the introduction of various distortion components. Note also that most of those liberties are in the interest of second-guessing the particulars of the imagined playback milieu. I think the 2.7 comes closer to those assumptions of normalcy and is therefore what Natasha called "friendlier". I can tell you that in the Thiel music room, the 3.7 provided goosebumps and giggles and OMGs. The 2.7 provoked smiles of admiration and relief regarding a long, hard haul to create a respectable product worthy of the Model 2 heritage of translating Model 3 breakthroughs into a more affordable package. The 2.7 is gentler. The 3.7 is more vivid. But I would not attribute the differences to tonal balance, or at least not primarily so; the two speakers are pretty similar in frequency response and polarity patterns. The 3.7 comes closer to Jim's goal of authentic translation of the input signal. What a fine pair of products, no? |
Prof and Eagle, I really appreciate your thoughts; you have far more experience with the products than I do (one listening session in 2012I) I would need to get intimate the products before forming a hypothesis. I will say that the Model 2 has always been more responsive and delicate due to coupling with less air. The 3 fills larger spaces and gives up some finesse to do it. The crosspoints seem to be higher on the 2.7 (no published specs that I can find.) Jay, I don't know the sales figures, but they were small. The high-end had gone $tratospheric and the dealer network had fragmented and things were falling apart, necessitating the sale of the company . . . and the rest is history. Even in its healthy-hayday, Thiel made only a few thousand pairs of any model before upgrade-replacement. |
I believe that if home theater and other spectacular bass recordings (ie: 1812 Overture Canons) had not been on the rise, Thiel may have stuck with sealed enclosure bass. The CS5 was developed as a guidepost to the future; its tweeter went into the 2 2 and the 3.6 and its Kevlar drivers were precursors to the stiff aluminum diaphragms to follow. CS5 bass approached 20 Hz with 3: 10" woofers (two deep and one upper bass.) They didn't bottom on full symphonic crescendo or heavy rock. But they bottomed on some "modern" mixes. Jim wanted deep bass as the foundation of the music, so he settled on the passive radiator due in large part to its ruggedness. Without a voice coil to bottom, bend and burn out, and with proper tuning, the passive radiator can do a pretty good job. I like what Vandersteen has done with a powered subwoofer in the enclosure. I suspect that Jim may have gravitated in that direction, having spent considerable effort developing the SmartSub. |
Ron and Prof, Ron's 3.7s are new old stock as I interpret the particulars. Rob has various cabinets and parts and I believe he assembled that pair from such service parts. So, all the parts would be unused. Break-in is something that many engineering-oriented observers dismiss as voodoo or make-believe or user acclimation. From the very beginning, we perceived its reality beyond question, but have never developed any definitive causal narrative. Jim's official answer was "I have no idea why." Of course he had ideas, but didn't want to enter the controversy. Among the causes are physical elastomeric settling of driver suspensions: surrounds and spiders. The cabinet itself settles in via the extended vibrational patterns. All the passive parts have their micro-structures altered by electricity, magnetism and vibration. A huge deal is solder joints . . . heat distresses the molecular structure of the long-crystal copper, which the "heals" with use. I experimented with crimped (cold-welded) joints, to audible improvement, which isn't practical in crossovers. Although subject to derision in some circles, I will be comparing cryogenically treatment to non. Prior experience with guitar strings and knowledge of what is happening biases me toward expecting improvement there. The ear-brain is capable of immense discernment. Our job is to find synergy among the myriad variables to produce cost-effective outcomes. |
Beetle - that distinction may be the watershed between hi fi and high end. Hi fi, dominated by academic engineers, mathemeticians and physicists definitely listened with their fore-brains and negated the existence that couldn't at least be verified by double-blind ABX tests. High end, on the other hand, broke through myriad barriers because they/ we believed what we heard. When the cats payed attention, we really took something seriously . . . the cats didn't even read the specs. |
Peter Aczel attempted with the Audio Critic to reconcile the objectivist with the subjectivist models of evaluation. A watershed event was his endorsement of Andy Rappaport's AMP-1 which was a zero feedback, highly coherent power amp that carried significant noise, which Aczel allowed as a successful amplifier design. Peter Moncrieff went even farther with his International Audio Review by devising new measurement techniques to support his (generally brilliant) subjective interpretations. Julian Hirsch preceded these guys and set the stage for evaluative techniques to educate the masses. However, his reliance on measurements denied the possibility of sonic differences if they couldn't be measured. He wrote more broadly than for Stereo Review, but his approach was consistent, often summarized as: "Of all the products I've heard, this is certainly one of them." By the early 80s Thiel had established strong retail presence in the NYC area. One of our very supportive and influential NYC dealers convinced us, over our considerable resistance, to have Julian review one of our products. (I'm sorry I don't remember which, but probably the 01, 02 or 03.) A review appeared in Stereo Review and a related article in the New York Times. (Something besides Bourbon comes to the real world from Kentucky!) The response was overwhelming; Thiel had somehow become legitimate in the minds of many thousands of readers by getting JH's stamp of approval. Note that none of that dealer interest was appropriate for an emerging high-end speaker manufacturer, and we did not sell direct to consumers. Those articles produced a flurry of activity that disrupted the ephemeral path of growth we were on. However it did serve our education regarding what game we were in, which was not Stereo Review's game. The emerging high end at that time, considered JH and Stereo Review to be solidly Mid-Fi, which was dominated by Bose and Bang & Olufsen. JH's drumbeat of endorsement of Bose served as a cornerstone of the establishment of the high end. "If Bose is as good as it gets, then I (aspiring dealer, manufacturer, etc.) have a real shot at survival" was a consistent mantra of the emerging high end. |
Prof - From the beginning in '74-'75 the goal was to find the best platform on which to build a line. We investigated (translate designed and built) spherical arrays, line sources, panels, folded horns, powered multi-driver speakers, and I may be forgetting a few. Phase coherence was in the list of goals, but not found to be practical. The first real product to take shape, the Model O1, incorporated what we could achieve in practical terms within our constraints. Its strengths were very high efficiency (94dB?) with its equalized sealed box bass, built on a custom driver by Eminence, who built stage drivers for Peavey and others. That driver had a huge magnet, good thermal management, long excursion and good linearity, but with a normal, overhung voice coil; the best that we could find for our needs. BTW, better than SEAS, Dynaudio, etc. The tweeter to match was a Long Engineering 1.5" mylar dome with good performance. Jim messed with first and second order filters for that product and landed on 3rd - 18dB slopes as the most practical solution. It was fairly linear, bass below 30Hz and a not too refined treble in a medium bookshelf package in all the wood finishes Thiel became known for. We gathered a following, especially due to some serendipitous European export opportunities. Dealers wanted a more refined, audiophile product and Jim developed the O2 as a response to demand more than his own ambition. It was a ported 6.5" Seas treated paper woofer under a 1" Peerless silk dome with second order slopes. It delivered a more refined presentation, trading off bass response and some efficiency. I think it came in around 90dB and served as a stepping stone into the emerging audiophile market, which really hadn't gelled yet. By 1977 we had attended our first CES and had enough distribution to figure out that we had to do something unique, memorable, extraordinary to carve out a meaningful niche. The next year and a half of extreme difficulty went into developing the O3 as a minimum phase transducer. We went to our second show with a second order O3 as backup because EVERYTHING mattered so much more when phase coherence was added to the formula. There were deficiencies that were later solved. We mustered our courage and presented the Minimum Phase version, having the rectangular normal tower in the closet, just in case. The response was overwhelmingly positive, and we never looked back to normalcy. That sounds smoother than it was. We faced another year of tracking down weird stuff such as magnetic eddies, wire anomalies, diffraction and so forth, all of which were sonically invisible with high order filters, but glaringly obvious with first order. The ear-brain interpreted the sound as "real" and held it to a higher standard than regular canned sound. That's a big subject, but I must sign off for this evening. |
Robin - Yes, the SmartSubs are Class D. Jim began working on those amps with a very talented designer at/from Vifa in the late 80s. They put quite a bit of sophistication into those designs, which were done in the early dark ages of Class D. Impedance requirements are less of an issue when the entire amp-speaker is modeled in toto. I use a single SS1 and a pair of SS2s in my studio and love them. I am searching for a good repair solution for them. I have heard that Jim considered Class D suitable only for bass due to inherent topology constraints. But, a lot has advanced more recently. I have heard recent pro-audio class D amps, and I would not consider such for my music use, just because, no real experience, but an inclination toward a topology with more inherent wow. I am intrigued by what PS is doing (thank you ronkent) . . . a tube input before a mosfet output stage seems potentially great. Regarding Pass, Nelson is a brilliant designer in my opinion. Pass and Thiel shared equipment and insights beginning in the late 70s. I would expect Kent's opinion about synergy to be extremely well-informed. |
Effort actually goes into minimizing the weight of the speaker. Each element: panels, braces, driver structures and diaphragms, etc. tries to get as stiff as possible per unit mass, in order to drive resonances higher in frequency where they are harder to stimulate, do less damage and are damped more quickly and effectively. Adding up all the mass-saving elements of the design process, the speaker is considerably lighter than it might have been without such attention. |
Dsper - I suspect something other than your speakers or amps. Your room setup may invite improvement (they always do), but rarely cause high frequency problems, except from side-wall bounce. Is there a hard surface where you would see the speaker if a mirror were there? (Take down that ceiling mirror!) The CS5 tweeter is really sweet and well behaved. There is a possibility of XO capacitor leaking that could overload the lower end of the tweeter, or upper midrange. On that front, I suggest you contact Rob Gillum at rob@coherentsourceservice.com for advice. What comes to my mind is cabling - RF leakage into cables-as-antennae. Check your RF situation-try killing all non hi-fi devices. Try borrowing Goertz or other cable that is very different from your present stuff. Try changing lengths (interconnects vs speaker cable) and so forth. Try shielded AC-power cables while you're at it. Sources are also suspect in these matters. You might consider trying an analog source with some impeccable recordings. My intention is to broaden your troubleshooting mind. Your post suggests you may be focusing on the system elements that are highly likely to be exonerated at the end of the process. Keep us posted. |
A real problem with Thiel speakers is how ruthlessly they reveal upstream problems. That's not to say they themselves might be the problem, but . . . A tell-tale is that if you can extract excellent or trouble-free performance from an excellent source, then your problem might lie elsewhere. Why I mentioned cable is that Thiel's current-draw exacerbates cable problems. Dealers and/or direct sale manufacturers such as Morrow tend to offer generous cable comparison schemes. I suggest long pre to power amp interconnects with short source and speaker runs, especially if you now have something different than that. When you don't have a clue, mix it up and see what floats. |