[chinsettawong]

Hi Jesper,

Can you please summarize what you think is good to know from the article?

 

[edform]

Hi Jesper,

Can you please summarize what you think is good to know from the article?

Taking this forum to be an educational facility in which citations from technical papers are appropriate [tongue firmly in cheek here] , it seems reasonable to publish the paper's conclusions...

**********************************
This paper develops an improved model of the high audio-frequency acoustic transparency of ESL assemblies and demonstrates a method for measuring the transparency. The ESL model includes the membrane, the two perforated plate stators, the damping cloth, and easily incorporates any grills and dustcovers, if required. The model development begins in Sec. 2 by collating acoustic models of the individual components: the membrane, the perforated plate, and damping cloth. The model of the perforated plate is improved from that of Baxandall by eliminating an inappropriate approximation for perforated plates with a high open-area fraction, by including the Fabry-Perot resonances in the holes, and by correcting a small numerical error in the approximation for the hole-end effects. The origin of the Wood anomaly and means for avoiding it are also explained.

The response of the complete ESL assembly is complicated and not amenable to a simple algebraic description. It is found that an electrical-analogue model, including lossless-transmission-line components, provides a good description of the complete assembly and is easily solved using SPICE software. The collective behavior of the components is found to be very different from the sum of the effects of each component due to the reflections that occur between the surfaces of all of the components. Therefore, the response of an ESL at high-frequencies cannot be determined solely from the attributes of the components taken in isolation.

The expression for the reactance of the stators provides a good guide for the stator design; the stator thickness should be minimized, the hole size should be small to minimize hole-end effects but larger than 1 mm diameter to minimize frequency-dependent damping effects, and the open-area fraction should be modestly high to ensure that acoustic impedance changes and reflections occurring at the stator surfaces are minimized.

Although the ESL model is one dimensional, and therefore cannot properly model hole-end effects and diffraction effects, comparisons between the model and measurements of two different ESLs show that the model provides a good guide to the high-frequency response. In particular, the high-frequency peaking often observed in ESL responses, and usually attributed to the transformer resonance, may often be due to inter-component reflections.

The overall trend of the high-frequency response of an ESL assembly is determined by the membrane mass, much as given by the low-pass filter response. Superimposed on this response are the complex, almost chaotic, ripples arising from the reflections between the various surfaces in the assembly. The amplitude of the ripples in the response are determined largely by the impedance mismatch between the air and the stator holes and range from about 2 dB for very open stators to about 15 dB for stators with a low open-area fraction. The distances between the various reflecting surfaces, and the overall thickness of the assembly, should be minimized to ensure the peaks and dips in the response occur at the highest frequencies.
*******************************
For information

The Fabry-Perot resonance effect is the source of roll-off and pronounced peaks and troughs above a frequency dependent on the configuration of the plate perforations.

The Wood Anomaly occurs for situations where the wavelength and the grating spacing are the same and gives rise to waves traveling across the surface of the plates allowing zero and first order surface waves to destructively interfere. It produces an initial very sharp dip in output and greatly reduced output for all higher frequencies.

Both effects were first observed in the field of optics but the analogy with acoustical effects is direct and fully explains observed performance of inappropriately configured stators. Headphone radiators, which are small, are only troubled by these effects at very high frequencies since the wavelength of sound at 20Khz is 17mm - rather a big hole spacing for a headphone stator. The other effects discussed in the paper are more important and the general guidance contained in the conclusions shows how to minimize them.

I hope this is helpful.

Ed Form

 

[evalon]

Hi Ed ... Well, I won't add much other than a bit of additional information that I personally find interesting: The free air "band-width" of a 1 um mylar membrane is 97 kHz and it derates linearly with thickness. Thus, a 3.5 um membrane will have a roll off frequency of ~28 kHz. Personally, I also find the article to be most interesting due to the - to me - accessible theory sections which IMHO give a good explanation (and possibly understanding) of why these phenomena lead to any given response.

Cheers,

Jesper

 

[-JFK-]

Hi all,

Based on Young's modulus (rigidity), the FR4 is around 20-30Gpa.
The aluminium is around 60-70Gpa.

I have read many post ago that the aluminium stator as been "aborted" because of the cost and the "finish" done by the process used. (I am right ? I have forgget something ?)
Now, I have my own CNC (24000tr), I'll be able to mill aluminium.
Yes, it cost a bit more, but with a well setted CNC, I should be able to make it "clean".

So, aluminium seem's to be a good choice ?
Thank's

 

[Garuspik]

Thin aluminium + any cnc\or laser cutter + sandblasting = great result.

 

[chinsettawong]

Hi all,

Based on Young's modulus (rigidity), the FR4 is around 20-30Gpa.
The aluminium is around 60-70Gpa.

I have read many post ago that the aluminium stator as been "aborted" because of the cost and the "finish" done by the process used. (I am right ? I have forgget something ?)
Now, I have my own CNC (24000tr), I'll be able to mill aluminium.
Yes, it cost a bit more, but with a well setted CNC, I should be able to make it "clean".

So, aluminium seem's to be a good choice ?
Thank's

Go for it and let us know the result.

 

[Muamp]

Hi JFK,

As you state, aluminium is much more rigid than FR4 (for the same thickness of materials), but flatness is also important. A while back I had some aluminium stators made to the same spec as my FR4 stators, which were PCB fabricated using 0.8mm aluminium substrate PCB + dielectric + 0.35mm Cu single sided. The 0.8mm aluminium stators were rigid, but not usable on grounds of flatness! So I did not use them.
I have since bought 0.8mm and 1mm aluminium sheet that I mean to get round to CNC sometime.

Hi Ed,

Just finally sourced 200 ‘sample’ gold pins for making ‘Stax’ pro plugs and cables. They look as good as the expensive pins I source in the UK, but these are from a company in Shenzhen and very cheap. I will measure the pins tonight to see if they are exact / good enough.... If the pins are acceptable, I should be able to source these pins easily to make plug/cables reasonably cheaply.

David.

 

[chinsettawong]

Hi JFK,

Hi Ed,

Just finally sourced 200 ‘sample’ gold pins for making ‘Stax’ pro plugs and cables. They look as good as the expensive pins I source in the UK, but these are from a company in Shenzhen and very cheap. I will measure the pins tonight to see if they are exact / good enough.... If the pins are acceptable, I should be able to source these pins easily to make plug/cables reasonably cheaply.

David.

Do let me know too, David.

 

[edform]

Hi JFK,

As you state, aluminium is much more rigid than FR4 (for the same thickness of materials), but flatness is also important. A while back I had some aluminium stators made to the same spec as my FR4 stators, which were PCB fabricated using 0.8mm aluminium substrate PCB + dielectric + 0.35mm Cu single sided. The 0.8mm aluminium stators were rigid, but not usable on grounds of flatness! So I did not use them.
I have since bought 0.8mm and 1mm aluminium sheet that I mean to get round to CNC sometime.

Hi Ed,

Just finally sourced 200 ‘sample’ gold pins for making ‘Stax’ pro plugs and cables. They look as good as the expensive pins I source in the UK, but these are from a company in Shenzhen and very cheap. I will measure the pins tonight to see if they are exact / good enough.... If the pins are acceptable, I should be able to source these pins easily to make plug/cables reasonably cheaply.

David.

Good news.

Ed Form

 

[janosch simon]

hey gang did you see the video of mrspeakers voce? looks very nice would love to know what kind of damping materials he uses any idea? also would be great if i sells the cable and conector would fit great to my just finished alpha centaury V6 esl amp and boy does it sound better the my energizer bench setup

cheers and thx janosch

 

[chinsettawong]

Hi Janosch,

I personally don't like to use any damping material, but that's just my personal preference.

Can you show us your Alpha Centuary V6 ESL amp?

 

[janosch simon]

sure thing buddy a propper case is missing but i had to test it the V6 is not available anymore and it had some pcb errors but the designer andreas rauenbühler was so helpfull on the way building it that i can hear now music this diy esl amp project was wayyyyyy over my skills but i grown with the project the amp is fed buy an diy streamer with an soekris dam1021 dac i use the balanced xlr inputs because over rca i got some nasty ground loups next step is including a volume knob cause i cant change the volume digitaly with my picoreplayer no idea why

with this amp i can now go back and realy optimize my esl! lot to do but so much joy
cheers janosch

 

[chinsettawong]

Do you have the schematic diagram? What tubes are you using?

It looks like a simple amplifier to make. Congratulations!

 

[janosch simon]

Do you have the schematic diagram? What tubes are you using?

It looks like a simple amplifier to make. Congratulations!

http://www.high-amp.de/html/hybrid-amp_eng.html shure but its a total diy project and the schematic is very very sensitive nothing for a noob like me D
now im a advanced noob
the v6 (which is show in the picture) uses EL84 tubes v7 which is now online has 6SN7 tubes and in adreas opinion they sound more tube analog like

to build the amp you need also his PWS 5.5 and the add-on for the Alpha Centauri which includes the heating for the tubes

cheers janosch

 

[YVWM47]

Amp looks amazing