Loudspeaker assembly kits are just what is needed when new loudspeaker chassis elements first see the light of day. The kits are the only thing that provides a practical application for them. We can feel fortunate with the Eton because we have Philipp Vavron on our team, who is known for building excellent drivers. His assembly plans, presented decades ago in a hi-fi journal, also contributed to our own development – they clearly showed how Eton loudspeakers should sound. That is when the “Eton sound” was created, which has now gripped many people and never lets them go. But as is all too often the case with true connoisseurs, even a Philipp Vavron cannot lend a hand everywhere he is needed. His day only has twenty-four hours too, and like most of us he needs to sleep at night. So Eton needed to hire someone who was capable of developing loudspeaker assembly kits and delivering halfway decent results. The perceptive Eton Managing Director quickly turned to the 40 years experience of the German Eton distributor for his choice.
Fresh from the production line, the tiny new components – barely 10 cm tall and with ceramic membranes – were placed on our table. The box read Eton 3-400/A8/25MG, and we soon discovered what that meant. An aluminum bracket under the 25-mm voice coil, 8 ohms and a magnesium membrane. The neodymium magnet with a pole-piece hole, and the many ventilation openings under the raised centering piece, we could see for ourselves. Philipp explained a couple of other features, like the two-sided membrane coating with keronite despite the weight of just 3.3 grams, along with the 55-mm ring diameter and 3-mm height of the magnet, and the 7-mm wrapping height with a 4-mm pole-piece thickness. This mid-range speaker would form the basis for a new box, possibly paired with one of the new eight-inch basses. In any case, the 26 HD 1 had already been chosen as the tweeter; anything else would have had a larger diameter than the pair of white “giants” that we were thinking of using for the desired volume of about 85 dB once the crossover was connected, and because of the separation around 200 Hz. It was packaged accordingly, and I can give this much away: thanks to the small series spread, you can also use the two 3-400s as twins without any additional selections. By that evening we had a few new chassis elements that the world had never seen before.
Next, we let the chassis elements “age” for 24 hours, keeping them in constant vigorous motion with a sine wave of 10 Hz and 5 volts. That is necessary in order to make the mechanical parts of the driver work as smoothly as they would after a “normal” playing time of 100 hours or more. After this time, the parameters stay in a constant relationship for the lifespan of the chassis, and the calculated cabinet is the right size. Once it was broken in, we screwed the 3-400 onto our measuring wall to create the usual diagrams. The impedance measurement was used to calculate its parameters. Interested do-it-yourself developers and simulation-program users can find the data here as a zip file, as usual.
Item No.: 1381316
Measurements as Zip-file
|Membrane||magnesium, keronite coated||Pole plate thickness||4||mm|
|Basket||die cast||Magnet diameter||55||mm|
|Pole piece hole||yes||Mounting holes||6|
|Centering||pot spider, ventilated||Outside diameter||4||mm|
|Magnet shielding||yes||Installation opening||77||mm|
|Voice coil||25 mm||Installation depth||37,5||mm|
|Voice ciol former||aluminium||Milling depth||4||mm|
|Frequency response and phase||Impedance||Frequency response under 0/ 30/ 60°|
|Distortion for 90 dB||Step response||Waterfall|
Naturally we performed the same procedure for the 8-502, which has now replaced the 8-800 in the BlueNote and was planned to be put in as a teammate. Up to that point, there had been no reason not to use it. But there is a good explanation for why its data doesn’t appear here after all. It needed 65 liters, which it got from my quickly assembled test cabinet. The two 3-400s were happy with a shared 3 liters, and generously let the lone tweeter sneak in between them. Before every crossover development, we measure the chassis in the box to discover the separation frequencies and any disruptions from the cabinet’s geometry. What the two mid-range speakers showed us was very impressive, staying close to the 90-dB line – we had expected something more like 87 dB below 1 kHz. The bass couldn’t keep up, and it stayed at 84 dB in the relevant range. There were three ways to adjust the volume levels for both branches. We didn’t want to use a series resistor, and installing only one 3-incher rather than two conflicted with our shared plan. So the only option left was to double the bass – but that meant we would need a horrifying 130 liters for each box. To visualize the whole thing, we drew the cabinets in SketchUp and 3-D. The design with the large basses and small mid-range speakers turned out looking so strange that we didn’t waste any more time thinking about that solution.
Smaller basses? We had already successfully built two of the new #17s into the Phase/Dusty 34, where they were convincing even without a mid-range speaker, which is normally a requirement for this size of chassis. Still, how else could we get out of this tricky situation? More help came from Philipp Vavron, who told us about his little construction with #17 and 26 HD 1, where the bass mid-range was only given about 15 liters – a capacity that even LSPCad thought was sufficient. On “paper” on the monitor, the whole thing looked pretty decent, even though the five chassis elements and the additional reflex tube meant there were more than a few components positioned on the front. The successful appearance took care of our faint bellyache, thinking of all the mid-range speakers that had been used with two 7-inchers. We had nothing to lose – we still had a 7-302 and 7-300 in the warehouse, both of which could be used in the Eton box. And why shouldn’t the mid-range speakers enjoy being handed off to a specially designed component instead of being modulated by the bass?
So we quickly bought some new wood. To avoid spending too much time thinking about the surface design, we once again used black MDF for the framed walls and glued beechwood for the side walls. As with the Excel 22 DXT, we had the MDF boards cut 8 mm wider and added a 4-mm-deep groove all the way around the glued-wood boards. Before assembly, we cut out the holes in the baffle board, with one person operating the camera. In the end, we had two nice cabinets and a couple of pictures showing the gluing process. To make sure no one misses anything this time by not clicking on the photos, we skipped the usual underlying texts.
The assembly plan is stored here as a SketchUp file for creative do-it-yourselfers; everyone else will have to manage with the simple drawing. The wood list includes 22-mm MDF as the construction material. It will need to be converted to a 19-mm width for the stained MDF boards and the glued beechwood.
The wooden boards had one night to get used to being together; once they had latched on to each other tightly, it was time for the sanding and painting work. Three bags of Sonofil needed to be distributed evenly in each box. The HP 70 bass reflex tube goes into the bottom hole, and then the chassis elements can be screwed in. Naturally, you need to solder on the cables first and feed them out through the reflex tubes. For the basses, we chose the 7-302/C8-32Hex.
Finally it was time to start developing the crossovers, a process that normally starts with the bass. Its volume drops off slightly below 200 Hz due to the lack of contact with the ground. If it is placed on the ground, this area is filled out. If you don’t take this into account, you will be punished with a much too strong bass foundation. A large Ferrobar coil and a smooth electrolytic capacitor cause the basses to steadily lose volume above 180 Hz. The mid-range speakers have five components connected upstream, two of which form a high and low-pass filter. The resistor lowers the volume by the 2 dB by which the two ceramic chassis elements exceed the basses. The tweeter is also guided into place using a second-order filter and a potentiometer. The two #17s are connected out of phase with the rest of the equipment. An impedance correction, which is only needed for tube amps, completes the crossover. The separation frequencies are at 400 and 3600 Hz.
Once all of the measurements were finished, we also installed the 7-300/C8-32HEX into the boxes with the same crossover just out of curiosity and received identical results, as expected. After an initial private listening test, we were able to introduce the boxes to the Etons, and finally to start using them in our listening studio. The box is now available for anyone who is interested in hearing it.
Still, we did have to build some new cabinets in a hurry because Mr. Kröner from Eton insisted on having a set for his own rooms. That gave us the opportunity to test another seven-incher with the same construction, with an aluminum bearing in place of the carbon for the voice coil. The only differences for all of the models were found in the upper mid-range, but that is no different than in the two-way versions. Now we needed to give the box its own name, one that seemed very appropriate for our box hierarchy given its sound: the Little Princess. Initially it was very hard to believe what our ears clearly heard. Regardless of the music, the Little Princess showed the same familiar effortlessness as our Duetta. Loud or quiet, classical or jazz, rock or electronica – the delicate box put everything it was handed onto a plausibly deep and wide stage that seemed fully freed from the speakers. Nothing indicated the ceramic tweeter, which is described in other combinations as moving more toward the front. The bass depth and the clear details it provided were also familiar from the Phase and Dusty; here they continued into the mid-range, which makes the somewhat spectacular style of the two-way speaker more homogenous. Duetta can go deeper than that, and maybe even louder, but in exchange it needs a cabinet that is three times bigger, which doesn’t exactly help with the Female Acceptance Factor. “Isn’t there some way to hide that with a clever design?” is a common question when girlfriends and wives come along for the listening session. The Little Princess doesn’t need to be hidden, but at the same time not too much is lost compared with the Queen.
Better than any sound description that we could laboriously type out, here are a couple of emails describing the effect that the boxes have on visitors:
“I like the name. But I would have left off the Little. The sound is big,” writes Carlo, who was the first unsolicited owner of the Little Princess.
“Where the Duettas were relaxed, the new speakers create a stage that is just as detailed and full of tone colors without seeming too loose. Congratulations,” Holga posted on the forum. And confirmed the darling daughter of the “Queen of Blues” or a queen-murderer? .
“..You also showed us your new Little Princess, which I and the other visitors especially liked. In particular, the impression of fresh, clear air (as if after a storm) made by this box stayed with me for a long time,” was Jochen’s summary of his listening experience.
We will stop patting ourselves on the back now, as soon as we share Joachim’s email: “It’s still hard for me to believe what an amazing sound the Little Princess had. The sound that came out of this relatively small cabinet is truly incredible compared to the other loudspeakers. At first I thought you were demonstrating a ‘cheaper’ loudspeaker for the benefit of the couple who came in later. I was just thinking, ‘Is that really necessary after the outstanding Blue Note?’ and then I was quickly converted. Our amazed faces probably spoke volumes about how impressed we all were…”
Yes, we were happy not to be alone in our amazement over these little boxes. But let’s leave it there for now. Describing sound coherently in words is a vain labor of love, and it is never as good as hearing it for yourself. We also have to concede that it’s almost more fun when visitors come into the studio with high expectations and then freely admit afterward that their expectations were exceeded.
|Chassis||2 x 7-200/A8/32HEX||Wood list in 22 mm MDF:|
|2 x 3-400/A8/25MG||105,0 x 26,4 (2x) sides|
|1 x 26 HD 1||105,0 x 18,0 (2x) front/ back wall|
|22,0 x 18,0 (2x) lid/ floor|
|Sales||Intertechnik, Kerpen||30,0 x 18,0 (1x) Midrange chamber back|
|Construction||Udo Wohlgemuth||10,0 x 18,0 (3x) Midrange chamber buttom/|
|Nominal impedance||4 Ohm||Milling depth:|
|Damping/ insulatiuon||3 bags Sonofil||Woofer: 6 mm|
|Terminal||K 42 AU||Midrange:4 mm|
|Tweeter: 5 mm|
|Approx. cost per box:|
|750 EURO/USD||Wood cutting: 30 EUR/USD|
|Frequency response and phase||Impedance||Frequency response under 0/ 30/ 60°|
|Distortion for 90 dB||Step response||Waterfall|