Thiel SCS4 Design Overview

Introduction

Anxious to see when these speakers would arrive, I logged into the UPS tracking website to find that I would be getting two packages, totaling 29 lbs. Imagine my surprise when only one of the two packages (marked 2 of 2) was delivered! Apparently someone at UPS thought I wanted one delivered and one left at the shipping hub! Given the weight of the first package, I expected both speakers to be packed in there and the second package to contain the speaker stands I requested for the review. When UPS brought the second package, I opened it revealing a single 25 lb speaker. (One speaker per box, no stands.) If you do the math, I am not sure how we ship two 25 lb speakers and get to a 29 lbs total weight, but then again, I would probably crash if asked to drive a big brown van, so who am I to complain?

The Thiel SCS4 is too small to be a floor standing speaker, and although on the large size for a bookshelf speaker, it will fit on all but the very smallest bookshelves. It will also fit easily on a typical speaker stand, but my first listening was done with the speaker on the floor. This is not the preferred location for this speaker, and Thiel clearly states this in their manual. When my son came into my listening room hearing the speaker for the first time, he immediately proclaimed, “Wow, that has way too much mid-bass!” (Adam is the Golden Ear in the family). Even elevated, this speaker has a bit too much mid-bass for my liking, and placing it on the floor exacerbates this issue. If we examine the impedance magnitude curve of this speaker, and place this information over the frequency response, it is clear that the output of the speaker is highest where the power demand is highest, and the impedance magnitude is lowest. In my opinion this speaker will sound more natural when driven with a tube amp than with a transistor amp because of the higher output impedance of the tube amplifier compared to a transistor amp. “Fattening” the low frequency output in this region is an easy way for dramatic affect to be added to movie soundtracks, and is sometimes done by manufacturers to create ”the wow effect” in the untrained ears of potential end users. I do not believe that was Mr. Thiel's intent. After speaking to him by phone, I believe he is motivated to make the best most neutral speaker he can, as he is by himself the engineering department, and wants most of all to make something he himself would buy.

Coaxial Driver Pros and Cons

What I did find disappointing about this speaker was that the on axis frequency response was not very flat for a speaker of this size. I do not recommend this speaker be toed in to put the listener on a direct axis with the system. The frequency response slightly off axis looked smoother than does the on axis response. This is one of the disadvantages of using a coaxial tweeter. The advantages of the coaxial tweeter mounting (used in the SCS4) are that you will get the same sound dispersion pattern whether the enclosure is oriented with the long dimension of the box horizontal or vertical. It also offers extremely smooth off axis response and phase linearity, especially in the crossover region. (Since much of the sound reaching us is reflected, some in the business argue the overall power radiated at all angles better represents what we hear in a live room than an anechoic axial frequency response does.) Phase linearity is very important to Thiel, and he makes it a priority in his designs. I personally have spent much time working in the “Pro-Audio” end of the business, where 97db for 1watt @ 1 meter is commonplace, and DJ's with 3000 watt amplifiers run into constant clipping is a daily occurrence. (The DJ market could care less about subtleties, they want as much raw output as they can get). In part because of my recent history with the Pro Audio business, this speaker is one I would describe as delicate. It was able to reveal musical subtleties that I would never hear when listening to the monstrously sized DJ gear that frequently occupies my lab. Despite its small size and coaxial design, I found the room location of this speaker surprisingly easy to localize, and believe that it is best used as the mains in a medium to small room, especially when assisted by at least two other channels for ambiance. If you want a high volume of deep bass, (read really loud music) you have no choice but to add a subwoofer. If on the other hand, you rarely play it above modest levels, the extension of the Thiels is more than enough for any but the biggest rooms or most ardent bass aficionados. A distinct advantage of putting a small driver in a box of modest size is that one can achieve a very low cutoff frequency. Thiels specifications for this model claim a cutoff frequency of 47 Hz, and my laboratory measurements bear this out. The disadvantage of the small driver approach is that the sensitivity of the system is low, and the absolute maximum output will also be very low at the lowest frequencies meaning the dynamic range of the system is modest. That said, considering its size, the low end is surprisingly good.

Design Overview

The build quality of the Thiels is excellent. As you remove them from the over sized shipping cartons, complete with very thick shock absorbing custom made packaging foam you will notice the protective black sock which completely covers the cabinet, and the plastic guard to protect the wood cabinet from being marred by unintentional movement of the metal grille. No one was counting pennies when the packaging was designed, and you would expect no less from a product of this cost and quality.

Over sized gold plated “5-way” terminals are used on the rear of the cabinet. Screwing the speaker wire ends or terminations in place is easy and the ridiculous standard ¾ inch center to center binding post distance (NOT used here) for banana plugs is thankfully discarded. This means men with normal size hands and fingers can actually screw the terminals down without having to scrape the skin of our fingers off on the terminals.

The cabinet is well built, and liberally braced. Knocking on the panels reveals a very high frequency of resonance of the cabinet panels, not one to be easily excited by the woofer, and not likely to radiate much in the way of its own sound. The front baffle is actually made from a single piece of cast aluminum. As Jim Thiel explained, this material is not only used for a beautiful appearance, but primarily because of the very high strength and lack of mechanical compression aluminum exhibits. The coaxial driver itself is rear mounted, and the front panel details allow Thiel to minimize the baffle diffraction by a careful and judicious attention to the driver box interface. By careful attention to the geometry, Thiel claims to minimize the diffraction effects of the box. (Off axis measurement at 90 degrees seems to bear him out).

The two ports, one each above and below the coax driver, are also cast into the aluminum panel. The air inlet inside of the box is constructed with a flared plastic tube which glues onto the straight end of the cast aluminum port hole.(flaring reduces port turbulence at high outputs and therefore lessens distortion)

The crossover was a surprise after being told the aim was a first order network in the transition band (first octave on either side of the crossover point). By my part count, this looks like a fourth order network with at least one Zobel network (Impedance compensation). The crossover network uses exclusively air core inductors. The capacitors were either Electrolytics or Film (polyester perhaps)?

Where Electrolytics were used, they were placed in parallel which results in less series inductance and resistance, so their negative effects are minimized. Unfortunately, the resistors are all glued right down on the PCB, minimizing their ability to dissipate heat. This is a minor issue with low power speakers like the Thiel SCS4. To minimize the weakening of the speaker box, the thru-hole and backplate panel are made as small as possible in order to keep the box strength at a maximum. The plastic plate appears to be just large enough to mount the crossover board. By minimizing the through hole size, we minimize the loss of cabinet strength created by the hole.

I was delighted to see that Jim Theil, like myself, is a fan of fiberglass insulation. We have both measured polyester fiber fill treatments, typically used by other manufacturers and agree that polyester fiber fill does little to nothing to minimize reflections inside a speaker box. The right amount of fiberglass was used here, and it was located in the right place. It was not lost on me that the fiberglass was skewered in place and held there with solid core copper wire. Audioholics readers who recall Thiels interview and discussion with Gene DellaSala about wires likely remember Thiel prefers solid core over stranded copper for wiring. Perhaps it has less to do with skin effect or material density than it does as being a tool to hold the fiberglass behind the speaker (where it does the most good) and away from the ports (where it will do the most harm). Stranded wire might not hold the fiberglass in place as does the very stiff solid core copper crossover to speaker wires.

My least favorite part of this speaker is the grille. It is nearly impossible to remove without causing some kind of damage to the cabinet or your fingernails. The grille desperately needs a small cloth tab at the bottom or top, so when pulled on, you can overcome the magnetic force that holds the grille to the speaker box. The grille appears somewhat flimsy, and frankly I was surprised that it did not vibrate and rattle during loud passages. It is made from a thin gauge metal, is long, and has an unsupported center causing me some concern.

Overhung vs Underhung Voice Coils

The speaker itself is a two way, with a 6.5 inch woofer, and a 1 inch coaxially mounted tweeter. The speaker cone is not entirely aluminum, but rather an aluminum skin over a polystyrene (plastic) material. The combination is both light and strong. This is a difficult thing to achieve with the shallow woofer cone like the one Thiel uses on the SCS4. The use of a deep cone, the usual approach to achieve high strength with low mass, would adversely affect the off axis response of the coaxially mounted tweeter, hence the unusual but effective approach. The low frequency driver (woofer) uses a non conventional motor. In most woofers, the voice coil is longer than the magnetic gap is high. This is known as an overhung voice coil, and is used by far and away most woofers. The advantages of the overhung approach include material efficiency and subsequently cost. The disadvantages include higher distortion and lower power handling (possibly not definitely). Thiel uses an underhung voice coil, which means the voice coil is shorter than the magnetic gap is high. It also means the voice coil is always completely immersed in the magnetic field – (read less distortion). To further reduce distortion, Thiel uses a copper shorting ring in the motor, to reduce magnetic flux modulation, and voice coil inductance changes with position, two other sources of distortion particularly problematic in woofers.

Another advantage to under-hung voice coils is less frequency response variation of the woofer with high drive levels. A disadvantage to this approach (besides cost) is that if overdriven, the speaker WILL BOTTOM or worse, break. Thiel recommends no more than 200 watts to drive this speaker. I would adhere to that without fail. Minimizing the maximum voltage across the woofer will eliminate the bottoming that will occur with higher than recommended power at low frequency.

While Thiel uses aluminum for the tweeter diaphragm (dome) he does not (thankfully) use it for the surround(edge). While aluminum has its advantages, it also has some negatives as well. If you are a fan of aluminum you know who you are. Those of us who are not, recognize that while it may work for the dome (which should be hard) it can be a disaster when used as the surround edge, which is supposed to be compliant. This latter approach is often found in compression drivers (for horn speakers).

A distinct advantage of putting a small driver in a box of modest size is that one can achieve a very low cutoff frequency. Thiels specifications for this model claim a cutoff frequency of 47 Hz, and my measurements bear out 45 Hz. The disadvantage of the small driver approach is that the sensitivity of the system is low, and the absolute maximum output will also be very low at the lowest frequencies meaning the dynamic range of the system is modest. That said, considering its size, the low end is surprisingly good for a speaker of such small size.

The Coax Driver – For those of you with a greater curiosity about the internal details, I will have to defer my opinion to the vendor. My sample did not include license to take it apart, but Thiel was good enough to provide a sectioned drawing detailing the inner workings of the driver.

Please note the driver itself includes a metal shielding can found in better designed drivers shielded for video applications. That back-can is not shown in the above diagram.