My DIY electrostatic headphones - Page 42

I initially was blown away by the sound stage, but after a bit of listening and then going back to my HD600's, I realized they really are almost completely lacking in bass. This is well explained by the FR.

I rebuilt it just to see if more tension would fix it, since at this point I have no clue what's wrong. It did not help the FR to have more tension. Although they are somewhat different, in the grand scheme, they're both largely qualitatively the same in that they have rolloff below 1.5khz.

The elvamide is working well, it's just a more resistive coating than others and it takes a while for them to charge up. I don't mind this, I was just checking if perhaps that was somehow contributing to the frequency rolloff. It also had no effect on the FR with regard to the rolloff problem.

So, basically I have no idea what's wrong. The first electrostatic headphones I made had almost flat frequency response right down to <50hz (unfortunately these headphones are long gone and scrapped, I don't really keep old ones together), then every new model I've built have had more and more rolloff below 1.5khz. Unless I figure out what I must be doing wrong, pretty soon I'll just be building electrostatic high pass filters!

Quote:

Originally Posted by arnaud 


I can see the diaphragm resonances in this graph (for instance what I suppose is the second radial mode in the diaphragm shifted from ~175Hz to ~250Hz when you increased the tension. I am pretty confident I can reproduce this with simulation (see my previous post about your old can simulation). It doesn't address the roll off below 1kHz but I may be able to see if this is an issue with the airpad cushion and/or earcup volume (or lack there of). It takes me a while to get to this but if you provide me all dimensions, I can give it a shot. Is this an omega 2 pad?

The rolloff does seem to get quite a bit better, although it's still there a little, when using soft uncovered foam earpads. This graph was using the SR007 earpad.

My only thought is that perhaps it has something to do with the fact that a large portion of the diaphragm is hidden behind earpad cushion, although I'm not sure exactly what acoustic effect that'd have. It would surely add some frequency-dependent damping since higher frequencies would see a volume under there but lower frequencies wouldn't.

Quote:

Originally Posted by chinsettawong 

How is your FR compared to what Purrin measured on mine?  

 

As for the coating, it makes a lot of difference in sounding.  My headphones would charge up instantly.  

 

If you are sure of alvamide, let the diaphragms charge up for about an hour and take another FR measurement.  If you see any difference, then it's the coating problem.

 

Wachara C. 

Mine overlayed on yours is as follows:

As you can imagine, not much bass there, and way too warm midrange.

I did let the elvamide charge up, overnight even... it did not change the FR. I also tried both soap and PVA-graphite (again, I know both are bad for longterm... just doing shortterm testing here to get my FR working) and both charge almost immediately but have the same rolloff. So it's got to be acoustic dynamics.

I second altering the coating and I would also look into backwave cancellation.  I've been working on the HE90 with Lambda drivers project over the last two days and one of the sets I tried just sounded odd with a strange reverb in the bass.  Very little low level bass too which quickly led me to find a small gap in the glue holding the driver to the baffle creating this odd effect.  If the earchamber isn't completely sealed it can lead to some very odd behavior. 

Btw. Wachara, the HE90''s sound amazing with Lambda Nova Signature drivers.  Some of the odd imaging from the HE90 remains but it's a minor effect and quite similar to the HE60.  The rest is just brilliant. 

I really like elvamide because it doesn't go unstable even when blown onto a stator with the same tension where other coatings would immediately go unstable upon plugging in. I assume this is because it's higher resistance at least in the way I apply it (takes a few minutes to charge up, and a few hours to reach full sound quality.

It also doesn't seem to affect the sound signature in my experience, and it doesn't affect the FR in any major way (if it does, it's below the threshold of repeatability in my measurement techniques)

Quote:

Originally Posted by chinsettawong 

 

One should not have to plug and wait an hour before using his headphones.  While high resistivity is good, but if it's too high, your diaphragm might never get enough charge at all.  When the charge isn't optimal, you won't have good bass.

 

It took me at least 2 years after so many trials and errors to find out this cheap and good coating material.  I just wish could you give it a try.

 

Wachara C.

Yes, I have some in the mail arriving tomorrow. I'll give it a shot this weekend and report back on it.

I'm also planning to make a pair of stators to test that have holes restricted to the center of the area where there is earpad opening. I've noticed that Stax appears to do this, and preliminary testing with duct tape indicated that it may help and/or solve my rolloff issue.

A few interesting findings this weekend:

First off, I tried out the anti-static cleaner. Not a chance, will never use that. In about half a second, went unstable to a degree that I had to literally peel it off the stator. And that was on the highest tension diaphragm I've ever made. Love Elvamide, and will probably use it forever.

More importantly, back to the key issue with my new headphones. I made a pair of stators with holes that approximately line up to the opening of the earpad. Makes a huge difference! Here are the FR's of the two channels, one built with the old stators and one with new stators. Everything else (tension, coating material, and sweat screen configuration) is identical between the two. Red is new stators with fewer openings, blue is old stators:

This is no measurement glitch, highly repeatable and listening to them definitely shows that the bass is approximately identical between the two ears, but one has painfully strong midrange. It's nothing subtle, just a few minutes of listening causes physical pain to me having that imbalance in the midrange.

I can only stipulate that it must be some sort of resonance forming in the volume between the diaphragm and the backing of the earpad. Would be very curious to hear any thoughts on what's really going on. I know Stax claims having some extra copper area with no holes increases the force on the diaphragm, but that shouldn't really cause this sort of issue I wouldn't think.

Also of interest is the dip at ~6khz. Again, it is definitely real and not some result of my measurement setup. I can hear a pretty big dip in that area when running a frequency sweep. Any idea what causes this to form?

Quote:

Originally Posted by chinsettawong 

Hi dude_500,
I'm very interested to hear from you what the problem with the antistatic cleaner is that makes you think it is so bad. If I were you, I would increase diaphragm tension even more until the diaphragm is stable, listen to it, then judge if it's good or bad. My point is that with electrostatic cleaner as a coating, your electrostatic field on the diaphragm is much stronger. That's what it should have been in the first place.
Without good efficiency, you won't get good enough sound.
Wachara C.

Well, the thing that was bad was the instant instability. It required probably about a half a pound of laterial force to get if unstuck from the stator. I can't even imagine how much tension that would require to keep stable. I doubt my tensioning jig is capable of making such a tension. I put a bit thicker of an elvamide coating on and now those charge up in about 10-15 seconds.

Quote:

Originally Posted by arnaud 

Dude500: very very interesting! Besides stax comment (to increase the force), another thing that comes to my mind is the increased rigidity of the stator. This is the reason why I asked you the dimensions and properties of the stator, but as usual I am lagging behind with the simulation... That or increase damping in the DS gap which severely affects mids/highs response. I have hard time imagining this relates to the earpad covering part of the driver (but what do I know eh ).
The 6kHz dip requires investigation but are there any dimensions within 50-60mm range like the inner pad diameter? Else a dimension in 25-30 mm range? What is the perforated diameter in your revised stator?

Many of the earpad opening dimensions are at least on the order of 50-60mm, although I haven't found any measurements that are dead-on the wavelength of 6khz. There is still that void between the now-covered part of the stator and the back of the earpad. I filled that in partially and the dip lowered in magnitude slightly and went up to 7khz. I'll cut some plastic parts to fully fill it and see if that changes anything. Also possible that void is related to the 2khz dip, perhaps a quarter-wavelength issue. At least in electrical engineering, quarter wavelength is a short circuit.

The new stators have an opening diameter of about 2.5in.