[lkos98]

Please do not forget that lkos98 is using "inside out" drivers: there is one stator in the middle between two membranes, and it is driven by a single ended amp.
I have built this configuration myself. It works well except one thing: it has way to much bass compared to the highs.
If this segmentation solves this, I can just congratulate to Ikor98 for the idea.
I have seen this kind of segmentation only on electrostatic speakers (like: ).

Thanks for clearing that out. This design is so called fully inverted and it's definitely boosting the bass. Middles and highs are quite normal.
I think I found the reason for it - bass boost is caused by the fact that membranes are charged opposite and attract each other all the time. This makes them closer in the middle (and to the stator), therefor ES force is higher, but opposing elasticity force is lower. This creates bigger air volume in low frequencies. Middles and highs are not affected that much, because even light, membranes are much heavier than air and can't vibrate with high volume too fast. Another prove of this is that when I increase the bias voltages (membrane centers get closer), the bass boost is increased even further.
Splitting the membrane in half is reducing the bass, in my opinion exactly what I needed. As seen from the FR, segmented phone is almost flat unlike before. I made the second channel and I'm very, very happy how they sound now.

 

[chinsettawong]

That’s an amazing news. I’ll have to give it a try. By the way, what bias voltage and spacer thickness do you use?

 

[lkos98]

That’s an amazing news. I’ll have to give it a try. By the way, what bias voltage and spacer thickness do you use?

I started with +- 330V, but it was too quiet, so I made a voltage doubler - about +-630. It made it much louder, but I noticed the huge bass boost. Since this was a new model, I was still not aware that bias voltage affects the bass boost, so I made voltage trippler and increased the serial resistance up to 6.6MOhm. It's now around +-750-800 (it's not a reliable measurements because of the multimeter resistance, so it could be +-900V). Fairly said I did not notice big difference in loudness between 600 and 800.
The spacers are 0.7mm (I ran out of 0.55mm FR4), but there is another space there - my stator is made of 2 x 0.55mm soldered copper to copper, so the membranes are facing the non-laminated side of it (I was tired of the sparks and varnishing, etc.) This makes the spacing 1.2-1.3 mm, but since FR4 has about 4 times higher dielectric constant than air, the distance can be expressed around 1mm in total. I guess this is quite a lot, that's why I increased the bias so much.
If you like - try it, but bear in mind - you have to start from scratch and it might be difficult to modify your amp (need to use just one of the differential outputs), one stator - double sided (or like me- 2 single sided) and so on, so on.
What I like about it:
First - no moisture caps (coating is inside the cavity created by the membranes and is not affected much by humidity)
Second - extremely low phone capacitance - I was really surprised when I measured it directly at the phone and it showed 0! (it's probably 2-3pF, which my multimeter could not measure).
Third - the membrane frame can be made as thick as you like, because membrane is not facing the stator from the side it's glued.
Four - four cables to the phone ( I use 6 just for convenience).
Disadvantages in my opinion - need for high and dual biases and high voltage amplitude at the output. Need to take care of the hum, noises and crosstalk, because they don't cancel each other like in the normal 2 stator design. Need for protective grills on each side ( I guess this is also advisable for normal design)
There are some more I guess, just cannot recall any at the moment.

 

[chinsettawong]

Ok, now I understand how you can get so much bass to begin with. With the spacer being more than 1 mm thick, the tension on the diaphragm can be set at a lot lower than my 0.5 spacer thickness. To be efficient, the bias voltage has to be at least 1KV. In your case, are you feeding one diaphragm with +700-800V and the other diaphragm with -700-800V? How much is voltage swing on your amp?

 

[lkos98]

Ok, now I understand how you can get so much bass to begin with. With the spacer being more than 1 mm thick, the tension on the diaphragm can be set at a lot lower than my 0.5 spacer thickness. To be efficient, the bias voltage has to be at least 1KV. In your case, are you feeding one diaphragm with +700-800V and the other diaphragm with -700-800V? How much is voltage swing on your amp?

Unfortunately low - it's +-300 (600 p-p) at max. It's not easy to build single ended amp with much more. Most of differential output amps also have around 600V p-p on one arm, so the combined is 1200V p-p. To achieve such swing with a single ended amp is almost impossible. My amp is made mostly with n-channel mosfets, but to shift the levels I need to use at least one pnp and all I can find is with 400-500V C-E. I ordered a 900V pnp from Aliexpress and I got a npn, not even knowing it's parameters. On top of it, my power rails are +-320V. Still I can't complain about the phones loudness - it's not extreme, but quite enough at normal listening levels.
Tension did not change the bass much, I have made a lot of experiments trying to reduce it and even with a very high tension it's still there. In my experience the ear pad sealing and active area play major role for the bass. May be with standard design it's different, the logic tells me not, but I haven't tried it and can't confirm.
The other option is to reduce the space, but I ran out of 0.5mm FR4.
Anyone know where I can find thinner FR4 like 0.2-0.3mm? I can still solder 2 of those face to face adding some sturdy structure in between.

 

[lkos98]

I'm a bit disappointed with the graphene coating:
https://www.aliexpress.com/item/100...o.order_list.order_list_main.5.686c1802zBbn14
It seems that after 3-4 months it's either drying out or rip off of the mylar. I noticed my phones were extremely quiet and decided to recoat them. Immediately the loudness came back with normal levels. Otherwise the coating is very nice - it's light, sound is clear and balanced, FR is flattened.
@chinsettawong , Wachara - wondering if you experience the same issue or the one that I bought is just a low quality product? May be with your humid climate it's better?
I will wait another month or so and check the levels again. I hope, I don't need to go back to the softener.

 

[chinsettawong]

Hi Ikos98,

You actually beat me to report this problem. However, I found that the coating is still good but it just takes a bit more time to charge up. On some of my headphones, it takes 3-5 minutes now before they sound loud and balanced. The strange thing is that some of my other phones don’t have this problem.

I still prefer this coating to my previously used floor cleaner. All of my phones now are a lot more stable and I don’t have a problem of the diaphragm sticking to the stator that much anymore.

 

[lkos98]

Hi Ikos98,

You actually beat me to report this problem. However, I found that the coating is still good but it just takes a bit more time to charge up. On some of my headphones, it takes 3-5 minutes now before they sound loud and balanced. The strange thing is that some of my other phones don’t have this problem.

I still prefer this coating to my previously used floor cleaner. All of my phones now are a lot more stable and I don’t have a problem of the diaphragm sticking to the stator that much anymore.

You are right, I noticed that charging behavior too, but it happened a lot sooner, may be after 3-4 weeks. I have to confess, after these 3-4 months I did not have patience to wait so much time. So mine coating is probably worse, or may be air here is a lot dryer than with your location.

 

[Jan H]

Hi all,
I'm so deeply impressed of the work being done in this thread, WOW! I wish I had the skills to build one.

Anyway I was watching this video on youtube, it might be this way Stax does the stretching.
Most likely using a Teflon piston instead.

 

[chinsettawong]

Hi all,
I'm so deeply impressed of the work being done in this thread, WOW! I wish I had the skills to build one.

Anyway I was watching this video on youtube, it might be this way Stax does the stretching.
Most likely using a Teflon piston instead.

That's too complicated, IMO.

 

[Ulfar4]

After long time I made new E-stats, this time with oval shape. The dimmensions of the cup are 110x90mm.
Active area of the diaphragm is 75x55mm.


Stators are FR-4 as always, this time 0.6mm thick with ENIG surface. Spacers are also 0.6mm. I tried some graphene nanoparticles treatment for cars as the coating but it never worked for me. So I ended up with rubbing graphite powder. This solution is working for almost 3 years without degrading on my previous headphones, but only downside is that the headphones are much more susceptible for farting.



Cups are printed from abs and this time I tried cheap silicone cable from Alliexpres and did my own plug.



There is frequency response from miniDSP with SBAF compensation, I was really impressed how good was the matching on this pair. Overall the sound is pleasant but I still prefer my old 110mm circular E-stats.

 

[chinsettawong]

Very nice!

 

[chinsettawong]

Where do you get the cable from? Could you please share with us?

 

[Ulfar4]

Sure.

https://www.aliexpress.com/item/100...1!sea!CZ!810190061&curPageLogUid=WEpbol3xr4Jm

 

[chinsettawong]

Nothing shows up on the link. What search words are you using?