Thiel Owners

@jchussey , The Thiel's are a complete speaker "system". The baffle angle was determined by the depth of the specific drivers chosen at the time. The crossovers were designed to compensate for the specific driver anomalies used at the time. And, time is the operative word here, in that is the very thing that most separates Thiel's from the very vast majority of other loudspeakers; their almost unique ability to preserve accurate time. Any changes without further very specific modifications will undermine that quality.

While I was aware of the scarcity of genuine Thiel CS 3.5 midrange drivers, I was unaware that that extended to early CS 3 tweeters. Does Rob at Coherent Source have any? Perhaps @tomthiel can comment if something like this might be a suitable replacement:

Thiel CS2 Tweeter in original Thiel box Dynaudio D-28AF - Speakers (highperformancestereo.com)

and there's this:

THIEL CS3 ELECTRIC BASS EQUALIZER | eBay

These are listed as pickup only, but as this pair's woofers need replacement, I'd hate to see two pairs of these classic speakers ruined if one pair could be sacrificed to save the other pair. Perhaps the seller would consider parting out:

Thiel CS3 80s Real Wood | Dan's Shop | Reverb

Though there's something to be said about modifying these old classics to be improved, but I find it an anathema to see these old beauties bastardized Willy Nilly to undermine them to be compromised versions of what they once were.

Good luck with the project, I do hope you can save them.

Tom D - the whole matter of bass is deep and sticky. I have spent considerable attention on the problem, and am still in the dark. But here are some thoughts. Put your boots on because we're going in over our heads.

At Thiel we did tons of (non-marketable) work on subs. There are so many issues  that we avoided the area until Home Theater required something. I think Jim's subwoofers deal with the issues extremely well. I have some now and agree with you that stereo subs are the way to go. There high enough frequencies for specifically directional cues. But of course its far more complicated than that.

Part of the problem is that there is no practical way (within budgets) to solve the bass phase issues which are so fundamental to Thiel designs above the deep bass. The (pre CS3.6) Model 3 family (plus the 01) had sealed enclosures which produce a perfectly damped low roll-off at 12dB/octave (second order) which mimics real acoustic instruments in real unrecorded spaces. Very satisfying bass. That bass roll-off introduces progressive phase shift that sounds natural because it is. Jim added active equalization to apply more amplifier power where needed in the bass without exceeding the power already required for clean midrange peaks.

However, when  a subwoofer (any subwoofer) is added there is an electronic crossover between the sub and the normal woofer, plus some unknown physical space offset injecting unknown time smear. Best case is a discontinuity that launches sub sound a full cycle behind woofer sound - albeit in-phase (I prefer "polarity") for augmentation. The ear-brain sorts out the temporal discontinuity, but not without consequences. What you do get is a pressurized environment (especially with stereo subs positioned closely) where the woofer "sees" a better physical impedance match between its force-motion and the air-space it is working into. Woofer distortion drops and articulation increases. 

Notice that on Jim's passive sub crossovers, you rarely actually hear the sub-bass. But it measures properly; it's there in the room. My fairly extensive auditioning and measurements corroborate his solution. Run the woofer full range for best placement and decay cues while using the sub just below audibility to create a better working environment for the woofer. The question arises as to how 'correct' the woofer signal is. In the retrospective work I am presently doing, and knowing intimately Jim's keen interest in 'the meaning of bass', I see Thiel's migration to reflex bass as a sell-out and know that it wasn't easy for him to accept. The transition from woofer to "port" (or passive radiator-same result) occurs at 4th order - 24dB/octave which puts the reflex fundamentals a full cycle behind the body of the bass coming from the woofer. Many folks (I'm one of them) hear that discontinuity as "slow bass" and less than natural, because it isn't. Reflex bass gets you an added octave (more of less) at extremely little cost, permitting overall system sensitivity to remain (twice) more efficient. Hard to side-step when nearly every peer does it even with speakers costing $6 figures.

Back to the question of how / why the laminar flow enhancements add so much sonic value - beyond understanding. Having spent the better part of the last couple years working on this problem I can offer some hints that may be gradually moving toward understandings. Let's examine some issues through the psycho-acoustic portal. We under-appreciate how much of what / how we hear is synthesized, including neural sub-circuits to enhance sonic recognition . . . we build models based on sonic inputs, not just from our ears, but also the mastoid process, and (lower than that) the solar plexus and skull and abdominal flexion and resonances. Saying that we don't hear below 20Hz is like saying we don't see outside some old-fashioned video frequency limits, or we don't smell unless we consciously identify what we're smelling. I know of weaponized sound at 3-6Hz, and euphoria-inducing sound at 7-12. Everything matters. Let's call it 'infrasonic' and pay attention.

Greg Lukens, the legendary inventor-audio engineer, evaluated an early version of DP's laminar flow technologies and hypothesized a sonic building-block explanation. The ear-brain builds its sonic conjectures (synthetic hearing) from the bass up. All harmonic structures are built on the foundation of the fundamentals. When those aren't there, the ear-brain "creates" them as phantom fundamentals. We don't know we're not hearing them. I posit that process of imagination to be quite benign. However, when we add that missing fundamental (as through a subwoofer) and that fundamental arrives at best a full cycle behind the upper harmonics and the upper harmonics of the subwoofer (or port, etc.) are a cycle behind the natural sonic structure coming from the woofer, we don't like it. We come to terms with it - we're quite excellent at accommodation - but nonetheless a lot of processing power is burned to get a less-than-satisfying hearing outcome.

We've gone pretty far here, mostly to address some of the difficulties of explaining a very complex subject. What I propose is that part of the reason we have such difficulty finding relevant measurements for the problems of "fake bass" or the bass-upper integration, or the unexplained imaging improvement, etc. is that these phenomena don't exist quite in the realm of the measurable. They exist in the realm of ideas and understanding, of epistemology - the study of how we know what we know, in this case what we hear. Let's posit (which I believe) that the surface-flow rectification contributes to fundamentally better organized leading-edge wavelaunch transients which produce substantially more lucid and interpretable fundamentals on which to build a harmonic stack that can be deconstructed into its component parts to sound like a musical event. Imagine that by removing significant chaos from those interacting onset transients, the auditory cortex can grasp the sonic event clearly whereas previously the event was dubious. A profoundly interesting aspect of my measurements is that "treated" laminar flow systems possess significantly better information from 0 to 20Hz than their untreated control system.  A very interesting aspect of all this is that we the listeners (controlled settings) do not necessarily attribute the sonic improvements to the bass, but more so call on qualitative observations such as "clarity, realness - accessible, involving, sweet, delicate, etc..  

So to your initial question regarding soundstage enhancement, I believe more brain power becomes available to deconstruct more spatial subtlety than when it was preoccupied with figuring out the basic harmonic structure of the sounds. Rather than technical measurements of the sound, we might come to greater understanding through brain activity monitoring.

That's all for now - For the Love of Music.

tomthiel

I did in fact remove the 5.5" lag bolt and used a (perfect fit) 3" hole saw to remove the back of the cavity. The piece that came out was 1/2" deep. The Morel MDT32S are both in and I have the speakers playing for the first time in 20 years as I type. The fact that you were very likely the one who shaped the baffles on these in the early eighties is pretty special. 

Thank you for the recommendation on the bass equalizer repair. My uncle kept all of the documentation with these speakers and I see that the equalizer was repaired (at no charge!) on 3-8-93 for "popping and static noise on output". The RCA jacks were replaced with the CS3.5 gold type.

Jim

unsound

Rob Gillum recommended the Morels. I looked high and low for the Dynaudios and placed an order with High Performance Stereo, but ended up having to get PayPal involved because they took my money and never shipped and would not respond to my inquiries. I advise everyone here to be cautious dealing with them. 

unsound - thank you as always for chiming in. Indeed a replacement driver is really the beginning of a redesign project and not the end. To elaborate: as you say, Jim considered all aspects as a system. Each driver has its Thiele/Small parameters carefully chosen for its tasks, plus its particular resonances and anomalies which he carefully modeled and corrected in the crossover such that the net acoustic roll-off was 6dB/octave with very flat impedance characteristics to realize very flat frequency, phase, time and impedance characteristics. Note that the only way he could get by with necessarily low impedances was by engineering an extremely resistive load. Changing a driver will change all or nearly all those parameters to defeat the subtleties of the design.

All that said, Madisound is a very knowledgeable distributor and they know a lot about their drivers. Rob has worked with them to choose the most appropriate available replacements.

Note also that Morel's inception was to pirate the Dynaudio designs, complete with international lawsuits, etc. So, the Morel replacements are quite likely the closest available replacement for the D28-AF.

Worth mentioning is the D28-AF was used in the CS2, CS3 and CS3.5. That 28mm driver has 20% more radiating area than 25mm domes.

As a generalization, I had hoped to endorse replacements by this time, but haven't gotten there yet. A real product update requires not only new drivers, but redesigned crossover networks. As a DIY, you can get pretty close with what's available, but a professional undertaking requires more diligence.