If you have followed some of my reviews over the years, you know I am a huge fan of planar (flat panel) speakers.
Falling into this category are ribbon, quasi-ribbon, planar-magnetic, and electrostatic (ESL) speakers. They are defined by their flat drivers.
While the first three of the sub-types mentioned above use permanent magnets to push or pull the flat electrical conductor, ESLs use a statically charged membrane (diaphragm) only about 12µ thick (about the size of a red blood cell).
The amplifier power (music) is applied to “stators”, which are perforated metal plates situated very close together. In between is the statically charged membrane. The varying signal voltage applied to the stators pushes and pulls the membrane back and forth, which moves the air, and you hear the sound.
An animated graphic illustrating this is shown below.
The static voltage is produced by a DC bias supply connected to the output of a high voltage transformer. The amplifier power also passes through this transformer, so that the ultimate voltage on the stators is much higher than it was when it exited the amplifier.
One of the main limitations of ESLs is this transformer. Regardless of how powerful the amplifier is, the transformer saturates at a certain point, and no more loudness can be achieved.
However, this tradeoff is balanced by the fact that the membrane, or diaphragm, is so light (low mass), that the sound has incredible detail. Although the tweeter in a conventional speaker has a very light driver (often a small piece of silk), and thus has good detail, the midrange driver cone has significant mass compared to an ESL, and it is in this midrange that the ESL therefore excels.
Bass, on the other hand, is a real problem with ESLs because the membrane cannot move very far, and therefore, cannot move very much air. So, low frequencies suffer. Also, the very low frequencies tend to cancel out because the air from the rear panel movement comes around to meet the air moving out from the front of the panel.
Usually, this is taken care of by use of a conventional woofer driver in the base of the ESL, covering frequencies from about 30 Hz to 200 Hz. But, using such a “hybrid” design introduces the phase delay issues that plague the crossover regions of all conventional speakers. And, the 200 Hz region is very important.
So, the high end ESL market offers “Full Range” products. I put the Full Range in quotes because they are not really full range, down to 20 Hz. Even large panel ESLs really just cannot move enough air to get a flat response down there. The best they can do is about 35 Hz – 40 Hz. To solve this problem, a separate subwoofer is used to get that last low octave. This is the case for the Final Sound 1000i ESL.
Final Sound is located in The Netherlands and was founded in 1992. Until 2002, their products – due to manufacturing costs – were purchased only by the high end (read mucho bucks) market. Then, they came up with a way to build them in modular form, and that reduced the manufacturing costs, but had no effect on the ultimate quality at all. Bottom line on that: still kind of expensive (what isn’t these days?), but now affordable.
The technique really was just a way to make the whole thing lighter. The ESLs I am used to weigh about 100 pounds, but the 1000i, which is Final’s largest unit, is only about half that. So, they are relatively easy to move when cleaning behind them.
One other thing about Final Sound ESLs that is unique, is that the amplifier power goes to the membrane (actually, a conductive center inside the membrane), and the static voltage is applied to the stators. They have patented this technology, and claim that it provides a clearer and more responsive sound.
At more than six foot six, the 1000i is about as tall as a San Francisco 49’ers line backer. But, at only 14″ wide, it is only as broad as that line backer’s leg. So, it’s tall and slender, relatively speaking. Also, it is only a couple of inches in depth.
Being transparent – or perhaps translucent – this tall, slender, thin ESL is a real spouse pleaser because it does not overpower a room with its bulk. This is typical of ESLs, and every female that passes through our labs, upon viewing any ESLs I happen to have on hand, has said, “I want them.” So, for those of you who are having a tough time convincing your spouse that you “need” a pair of really, really nice, but pretty big speakers, your problem may be solved.
Due to the modular approach, Final Sound has been able to design small ESLs too, and these would be ideal for home theater use. One such package is the 90HT ($2,498), shown below. Since these ESLs are pretty small, you need to set the crossover higher, around 120 Hz, and you should put the subwoofer close to the front left/center/right speakers, since sounds from 50 Hz to 120 Hz can be localized.
Specifications: 1000i ST ESL
- Design: Electrostatic (ESL); Full-Range
- Driver: Mylar Membrane
- MFR: 38 Hz – 20 kHz ± 3 dB
- Sensitivity: 86 dB/W/M
- Nominal Impedance: 4 Ohms
- Recommended Amplifier Power: At least
- Dimensions: 78.3″ H x 14.2″ W x 2″ D
- Weight: 50.5 Pounds/Each
- MSRP (USA): $9,999/Pair (Chrome Finish);
$11,000/Pair Piano Black
- Design: Sealed Enclosure; Front-Firing
- Driver: 12″
- Amplifier: 220 Watts RMS
- MFR: 25 Hz – 200 Hz ± 3 dB
- Crossover: Variable 60 Hz – 200 Hz
- Variable Phase
- Dimensions: 15.4″ H x 15.4″ W x 15.4″ D
- Weight: 66.1 Pounds
- MSRP (USA): $999
Even with this small ESL, you don’t have the phase delay problems of convention two-way speakers, because the ESL membrane produces sounds at high frequencies and also the midrange.
Often, with conventional speakers, you might be required to screw on the feet. With the 1000i, you have to assemble the entire base.
First, you remove the main panel from its box, and either lay it flat on the floor, or lean it up against a chair, resting on the bubble wrap packing, like I did for this photo. Notice the protruding metal rod at the base.
Next, you have to assemble the base plate, consisting of an oval-shaped plate and two metal cylinders. Place the cylinders on the black posts.
Notice that the cylinders are in pairs for left and right. Identifying which one goes where, you place each so that the rectangular hole is closest to the outside edge of the oval plate, as shown in the photo below. Then, using a hex wrench, tighten the bottom bolt on the rear of the cylinders, but not so tight that the cylinder can’t move just a bit.
Now, you pick up the main panel, and set it down over the cylinders so that the metal rods drop down into the rectangular holes. Don’t let go of the main panel yet.
Still holding onto the panel with one hand, use the hex wrench to tighten the top two bolts on each cylinder, so that the panel is firmly gripped. Rock it back and forth to make sure everything is tight, then let go of the panel but keep your hand close, to ensure that it is steady. See those little red dots at the bottom of the panel? That’s my blood. I scratched myself on the sharp corner on the panel to the right of the cylinder. So, be careful!
To the left of the speaker binding posts is the socket for the included 12 volt wall wart which supplies the voltage that will ultimately be applied to the stators. The wall warts have very long cables (thin) so they will reach just about any spot the speakers are placed, and in general, ESLs should be out from any wall surface at least a couple of feet (they are dipoles, and just as much sound comes out the back as out the front).
The subwoofer that was paired with the 1000i for review is the S220. It is a 12″, 220 watt, sealed enclosure unit.
The amplifier has several controls that you don’t see on other subs, mainly slider switches that let you tailor the subwoofer crossover to particular Final Sound ESLs. These include switchable high-pass, and Variable Mid-bass EQ (there is also a different control called Variable Bass EQ). For example, if you have the 1000i, one setting is used, while with smaller units, such as the 400i , a different setting is more suitable (high pass is at higher frequency because the ESL is smaller). A table in the instruction manual explains which settings to use with which Final Sound ESLs.
The S220 is not a really big sub, and the idea is that you could get at least two, which would give one for the front left and front right channels. Of course, this would entail some fancy footwork in the processor or receiver bass management setup. You would need to run a pre-out to the sub, then the high-pass line out from the sub back to the pre-in on the processor. This requires that you have a main-out and main-in for the front two channels on the processor or receiver, which some units do have. Otherwise, you have to use the subwoofer-out on the processor, with crossover settings done in the processor. For the 1000i, I used 80 Hz because its frequency response specification is – 3 dB at 38 Hz (I wanted to cross over at a region that was still part of the flat response), and because I wanted the ESL panels not to have to deal with deep bass. Obviously, with higher crossover frequencies, you need to put the sub(s) near the speaker(s) that they are associated with, because above 50 Hz, sound can be localized, and you want all the sound for a particular channel to be coming from the same part of the room.
I tested the 1000i ESLs with a Denon DVD-5900 DVD player, Theta Casablanca III SSP, Classé CA-5200 five-channel power amplifier, and Final Sound 400i ESLs for the center and surrounds. Cables were Nordost. I played music in Matrix (surround sound) and in stereo (two channel). I used the Toslink output from the Denon to the Casablanca to let the SSP handle the decoding.
I placed the speakers about 10 feet apart, toed in towards the listening position, several feet out from the side and rear walls. The subwoofer was placed between them against the wall.
Handel’s Water Music (Virgin 0-94639-13342-0) is some of the greatest music ever written, and a perfect test for the 1000i.
What struck me most was not that the oboes were so distinct from the violins, but how transparent the whole orchestra sounded. This is characteristic of ESLs because they are dipoles.
Also, the instruments sounded very natural, rather than having a characteristic due to an enclosure. Remember, ESLs don’t have an enclosure as such. They are not in a box.
Ein Deutsches Requiem, by Brahms (EMI 0-94636-53932-0), is very somber, as Masses tend to be. But it is beautiful music.
What I listened for here were the soprano voices in the choir, against the violins.
Wow, what a sound! So sweet. So clear and transparent. No harshness. It filled the room.
Haydn’s Symphony No. 88 (EMI 0-94639-42372-9; 2 disc set) is a much different kettle of fish than the preceding two CDs. Here the music is light, airy, powerful, and is not for background listening.
Several octaves of violins sounded like the several octaves rather than mush (they would have been mushy if there were a lot of IM distortion).
The flute and oboe in among all those violins were easy to pick out.
Bass was nice, even when the subwoofer was not on, but the S220 added to the effect. Even though the 1000i is a full range ESL, you definitely want a subwoofer.
200 watts per channel seemed to be quite enough. That does not mean you should not have a bigger amplifier though, just for special transients.
Swan Lake (EMI 0-94639-32432-3; 2 disc set) is not just music, it’s a ballet, and the complete ballet takes up two CDs.
One of my favorite pieces is the waltz in Act I. I was truly amazed at how the 1000i was able to separate the triangle from the flute. Sometimes, a triangle can sound a little strange if the tweeter is not first rate. Well, having played in an orchestra myself (OK, it was the high school band), I know what the triangle actually sounds like, and the 1000i reproduced it perfectly.
The 1000i sounds pretty good off axis too, and that is difficult for a large planar speaker to do. Part of this is due to reflections from the rear side of the ESL off the wall.
There was no hole in the middle. I could place each instrument along the sound stage from one side to the other.
All in all, I was very happy with the sound of the 1000i ESLs, particularly with the mids and highs. Some planar speakers have disappointed me in the past in this area, so I really looked forward to seeing if the 1000i would perform better in this regard. It did.
On the Bench
The microphone was placed 12″ from the middle of the panel for the distortion tests, and 1 meter for the room response test. I used a standard SPL value of 90 dB instead of 100 dB since half the sound was going out the rear, away from the microphone. THD+N measurements were within a 10 Hz to 22 kHz bandwidth.
At 100 Hz, THD+N was less than 4%. This is somewhat more than you might find with a conventional woofer driver, and as I mentioned, ESLs have a tough time with low bass.
At 1 kHz, 0.5% THD+N.
And, at 10 kHz, THD+N was 0.8%. This is less than many other conventional tweeters.
IMD using a SMPTE/DIN standard test, was 0.8%. I could test this (IMD) here because the 1000i panel reproduces both sine waves, rather than a tweeter reproducing one and a woofer reproducing the other.
The room response was reasonably flat down to about 60 Hz. The deep valleys are room issues. It rolled off a bit above 16 kHz.
The impedance stays at least 8 ohms up to about 1 kHz, then goes down to a low of slightly less than 2 ohms at 12 kHz. The large peak is due to the transformer. It stays below 4 ohms between 4.5 kHz and 20 kHz, so you should use a good amplifier that is rated into 2 ohms. As the sensitivity is only 86 dB, I would suggest at least 200 watts per channel.
The second graph shows the same impedance curve in an expanded scale.
The electrical phase stays within ± 60°.
Distortion tests were measured at 12″. The room response was at 1 meter.
At 20 Hz, THD+N was an unacceptable 30.5%. You really need a larger driver (e.g., 18″) or long throw driver, a more powerful amplifier, and a larger enclosure to get a good output at this low frequency.
Maximum output at a combination of 20 Hz, 31.5 Hz, and 50 Hz was 112 dB (1 meter).
At 25 Hz, THD+N approached an acceptable level (10% is considered acceptable by some people in the business).
At 31.5 Hz, we finally reach the standard acceptable level of THD. My own preference is to not have more than 3%, but that’s just me.
At 40 Hz, 4% distortion.
And at 50 Hz, 2%. Overall, fair performance, but not great. The output profile will improve if you use two of these subs. Dr. George Ruben – one of my colleagues at Dartmouth University – uses one Final Sound subwoofer for each of the five channels (he has 800i ESLs all the way around).
The room response with all EQ turned down rolled off below 40 Hz. With the Bass EQ turned all the way up, performance improved significantly.
The Final Sound 1000i electrostatic speakers are truly delightful. I loved the transparency, the clarity, the natural timbre, and the smoothness of the sound. They require a good amplifier, but that is standard for ESLs. I would suggest getting more than one of the subwoofers, or as an alternative, one subwoofer with a larger driver and more amplifier power.