Inverted electrostat inferior in distortion?

Interesting 🙂
I would love to see how, i still can't figure out the way they build the conductive part of their stators.
As i understand it they use the composite resin as part of the whole capacitor, that it has high dielectric constant.
The last time I talked to this guy was somewhere in 2014, I remember he said that several resins are used in the manufacture of these stators and he experimented a lot by mixing these resins, for this he even made a mechanized “hand” to continuously mix with it for several days: ).

The resulting composite had high-resistance conductivity. In the dark, the stator glowed with a barely noticeable blue color.
I made a mistake when I previously wrote about holes (perforations) in the form of tubes, I found a photo of it on an old computer, its stators and holes are in the form of horizontal holes, and also in the photo is its amplifier (direct drive):
 

Attachments

  • P1030589copy-th.jpg
    P1030589copy-th.jpg
    84 KB · Views: 108
  • P1030590copy-th.jpg
    P1030590copy-th.jpg
    78.4 KB · Views: 113
  • P1030593copy-th.jpg
    P1030593copy-th.jpg
    72.5 KB · Views: 106
  • P1030584copy-th.jpg
    P1030584copy-th.jpg
    71.2 KB · Views: 110
The last time I talked to this guy was somewhere in 2014, I remember he said that several resins are used in the manufacture of these stators and he experimented a lot by mixing these resins, for this he even made a mechanized “hand” to continuously mix with it for several days: ).

The resulting composite had high-resistance conductivity. In the dark, the stator glowed with a barely noticeable blue color.
I made a mistake when I previously wrote about holes (perforations) in the form of tubes, I found a photo of it on an old computer, its stators and holes are in the form of horizontal holes, and also in the photo is its amplifier (direct drive):

Nice!
Thanks for digging up the photo's, inspiring build 🙂
 
Some ESL-63 distortion measurements @ 1 meter 75 dB on axis, REW 5.30 MiniDSP UMIK-2
22 sec. sweep 10 Hz - 24 kHz, 96 kHz sample rate

1 kHz - 0.064%
500 Hz - 0.064%
400 Hz - 0.064%
300 Hz - 0.064%
200 Hz - 0.073%
150 Hz - 0.14%
100 Hz - 0.23%
80 Hz - 0.11%
70 Hz - 0.17%
60 Hz - 0.26%
50 Hz - 0.26%
40 Hz - 0.63%
Add to the list above

30 Hz - 0,39%

We now have ALL distortion below the noise floor with our latest improvements on the Quad ESL-63

combi.jpg


And a much improved pulse and step
Overshoot in pulse is only 16% @ 54 µS and a diaphragm stop at 330 µS

pulse step.jpg
 
Last edited:
The listening experience is even more impressive 😎

That change in SPL doesn't make any difference (or at least very very little) for these speakers.
I have done those measurements (on earlier request) too and found only a very small deviation from these numbers.
Then people asked to 100 dB, never done that.

I will look up the numbers or do a new sweep at a higher SPL and post the results for you.
 
Here are the 90 dB measurements
Started low, but a 90 dB full sine at 10 hz started to make the Mylar hit the stator, so no number below 60 Hz , sorry.
To not influence numbers because of the Mylar hitting the stator at very low frequencies and having to settle out, my best result was starting at 60 Hz as lowest frequency to get good measurements
Graph 75 dB versus 90 dB below (

60 Hz - 0,58%
70 Hz - 0,27%
80 Hz - 0,28%
100 Hz - 0,35%
150 Hz - 0,17%
200 Hz - 0,09%
300 Hz - 0,17%
400 Hz - 0,17%
500 Hz - 0,16%
1 kHz - 0,18%

Please note this is NOT a standard ESL-63 by a long shot.

90 db distortion.jpg



DIY.jpg
 
Here is the distortion plot of the hypergraph TL from hobby hifi which I have build as a bass speaker. At midrange it is on par with the quads build by Wout. But at 60 Hz it has higher distortion. For a 13 cm woofer it is not bad at all and lower distortions are achieved by designs incorparating a larger woofer.
Below the 60 Hz the hypergraph produces astonishing deep bass.
My guess is that below 60 Hz distortion of quad quickly rises but that is no shame as most speakers do. The numbers of the quad are actually very good so thumbs up for that. Probably better than audiostatic es100 as well
 
One can conclude that high quality dynamic speakers with sufficient area can have equally low distortion as the quad63 at 60 Hz. This is just harmonic distortion
Dynamic behaviour at low sound pressure is probably better with stats as they dont suffer from having to ' break loose' from its rest position.
 
Sorry but that is just wishful thinking IMHO

  • The measurements don't even show if they are only 2nd harmonics or THD with all even and uneven harmonics 2nd to 9th included as in my graph.
  • The measurements don't show distance.
  • I can't even see if only a single driver was measured or a speaker.

That makes it just a random graph with no context. So it may or may not be completely out of context.

As long as not all parameters are supplied nothing can be concluded at all IMHO.
The measured distortion numbers were only presented here to show what level of low distortion we have reached with modding the old Quad ESL-63.
 
K2, k3 and k5 are shown by hobby hifi and can be compared.

Distance is 1 meter

It is the complete speaker. Hobby hifi also measures individual transducers; they give both. Pick your choice

Please note that your measurements are quite limited because there is no information below 60 Hz. It is below 60 Hz where drivers really start to struggle as they have to make significant excursion. ESLs also show increased distortion. For example saturation of the step up transformers, stifness og mylar etc. I think the limited frequency range is by far the most important issue which makes a fair comparison impossible
 
I just showed the results of my in house measurements after our latest Quad ESL-63 modifications as a result of 10 years of research and testing, because we had never seen such low numbers in any of our earlier measurements on the same speakers,
Not wanted to start a ******* contest with any other speaker brand or model from any historic magazine. I still miss context in print, but that is OK.
These are our numbers, if some did or can do better or equal these numbers that is great. No problem. Not the reason for posting.
 
Last edited: