[Jedrula1]

I am trying to figure out how to implement the Sonion EST due to the very low output. Is it necessary to tame the rest of the drivers so that it is really audible? Or it is designed to sound this extremely dim?

Yeah, that is the main issue regarding EST drivers. In the past, Sonion released a second revision of dual EST called EST65DB01 with a lower impedance transformer to improve output by +2.5 dB but unfortunately, it did not solve the problem in its entirety.
In our flagship model Aurum we had to create a very complex crossover as well as a special acoustical chamber to match the very low level of EST with other drivers.
Of course, you can use quad EST65QB02 but IMO it is overkill.
As an alternative, you can consider using the new construction of super tweeter from Knowles - RAU/RDU series

 

[saldsald]

Yeah, that is the main issue regarding EST drivers. In the past, Sonion released a second revision of dual EST called EST65DB01 with a lower impedance transformer to improve output by +2.5 dB but unfortunately, it did not solve the problem in its entirety.
In our flagship model Aurum we had to create a very complex crossover as well as a special acoustical chamber to match the very low level of EST with other drivers.
Of course, you can use quad EST65QB02 but IMO it is overkill.
As an alternative, you can consider using the new construction of super tweeter from Knowles - RAU/RDU series

Thanks for the info. Your Aurum is definitely very interesting!

 

[musmecca]

Yeah, that is the main issue regarding EST drivers. In the past, Sonion released a second revision of dual EST called EST65DB01 with a lower impedance transformer to improve output by +2.5 dB but unfortunately, it did not solve the problem in its entirety.
In our flagship model Aurum we had to create a very complex crossover as well as a special acoustical chamber to match the very low level of EST with other drivers.
Of course, you can use quad EST65QB02 but IMO it is overkill.
As an alternative, you can consider using the new construction of super tweeter from Knowles - RAU/RDU series

I always shorten my tube lengths when using an est and combine it with another tweeter. Gotta use a high pass as well or it will crap out on lower frequencies. This works for me as attenuating the low end to match the est yields a very dull low end. I only use est for extra "air" or sparkle...I don't expect it to cover all treble duties.

 

[Wgibson]

Hi @Jedrula1 , I was hoping I could pick your brain for a second regarding the Sonion EST's, please and thank you

Have you noticed any EMF or other "placement problems" with the transformers relative to other drivers?

Is it feasible (or is it already being done) to put a crossover between the transformer and EST?

Thanks in advance, but if this is proprietary info then no worries, had to try!

 

[swtnate]

GV has no resistor on HODVTEC

Im fairly certain there is some added resistance internally on the HODVTEC in the GV configuration…. Also, I would recommend running 2 tubes as opposed to the coupler. The sound I always registered from the GV was pretty dark when combining to one tube. Even with 2 tubes the low end can still over power the TWFK. if youre looking for some advice on taming it, I would use an L-Pad followed by a zobel, and use either a yellow or orange damper. The TWFK I usually setup to have a 0.82uf to 1uf capacitor going to the WBFK. If youre using a straight up GV, you will need to rely on tube length, width, and dampers to tame the HOD.

 

[stegeoc]

Hi all, I’ve barely had any time recently for diems so my latest build took forever to get just right, but the results were well worth the wait. The plan was to build something more bassy than my other iems, with tame enough treble that allows it to be cranked up loud without any harshness. I’m very happy with what I ended up and it is a very clean sound with good clarity and powerful 38xj bass (albeit a bit more upper-bass than originally intended), but one thing that surprised me was the best sound stage I’ve ever heard from an iem.

That was more luck than anything else but do you guys have any tips on what factors into a good sound stage? I presume good left/right matching, perfect fit and seal. Do more tubes help? And their exit angles perhaps? Or is it just a function of the frequency response? Cheers

 

[stegeoc]

For the greater good I decided to sacrifice a GV driver to determine once and for all the values of said components.

1. There are no resistors in this quad. Just capacitors.
2. I believe but not 100% the TWFK used is a 60173 which I will post a pic of, and I believe it already has one capacitor attached which is the same value as the one on the GV. Obviously the positive thermal wire would need re-routed to the config in my drawing.

TWFK-60173 available to purchase at digikey. Spec sheet is here.



The HODVTEC wiring is crossed, which I have attempted and it does work.

The capacitors looking at the pic below are as follows.

Right capacitor reading was 5.13uf - there could be a tolerence so it is likely a 5.2uf or a 5.0uf.

Left capacitor reading was a 0.52uf - there could be a tolerence so it is likely a 0.47 uf as that is common.

 

[swtnate]

Hi all, I’ve barely had any time recently for diems so my latest build took forever to get just right, but the results were well worth the wait. The plan was to build something more bassy than my other iems, with tame enough treble that allows it to be cranked up loud without any harshness. I’m very happy with what I ended up and it is a very clean sound with good clarity and powerful 38xj bass (albeit a bit more upper-bass than originally intended), but one thing that surprised me was the best sound stage I’ve ever heard from an iem.

That was more luck than anything else but do you guys have any tips on what factors into a good sound stage? I presume good left/right matching, perfect fit and seal. Do more tubes help? And their exit angles perhaps? Or is it just a function of the frequency response? Cheers

#1 - making them identical
more tubes CAN create more problems BUT when done correctly, it makes for a cleaner delivery of the frequency spectrum and less resonance. Getting the tubes to lay down as identical as possible really does do wonders. Sometimes one ear will have a harder turn going into the canal than the other. That big turn will alter the response from the other ear, just as an example. Lastly, having the canal depth the exact same is just as, if not more, important than exact tube lengths.

 

[stegeoc]

#1 - making them identical
more tubes CAN create more problems BUT when done correctly, it makes for a cleaner delivery of the frequency spectrum and less resonance. Getting the tubes to lay down as identical as possible really does do wonders. Sometimes one ear will have a harder turn going into the canal than the other. That big turn will alter the response from the other ear, just as an example. Lastly, having the canal depth the exact same is just as, if not more, important than exact tube lengths.

Interesting, thanks. I've noticed the filtering effect on 3D printed tubes with tight bends and now copy/paste them to make them as similar as possible. Maybe that is what helped this time.

What you said makes sense logically, but I guess there must also be some characteristics in the frequency response too? Otherwise very simple (but well matched) universals would have better sound stage than they do. Or is it possible to have a good sound stage with any tuning (warm, bright, U, V etc.)?

 

[swtnate]

Interesting, thanks. I've noticed the filtering effect on 3D printed tubes with tight bends and now copy/paste them to make them as similar as possible. Maybe that is what helped this time.

What you said makes sense logically, but I guess there must also be some characteristics in the frequency response too? Otherwise very simple (but well matched) universals would have better sound stage than they do. Or is it possible to have a good sound stage with any tuning (warm, bright, U, V etc.)?

3D printed frequency pathways typically yield much less resonant frequencies and usually prevent acoustical phase/ frequency cancellations. Its much easier to engineer the sound tunnels and much longer lengths with varying widths along the soundpath. Now, thats all contingent on your printer and resin(s) having minimal scale fluctuations in your prints.

 

[swtnate]

When you cant get the old school method out of your system but you also LOVE new technology that makes the shell so much more pristine. I absolutely love the Pro3dure CD-2. Nothing comes close to beating its versatility.

 

[swtnate]

Interesting, thanks. I've noticed the filtering effect on 3D printed tubes with tight bends and now copy/paste them to make them as similar as possible. Maybe that is what helped this time.

What you said makes sense logically, but I guess there must also be some characteristics in the frequency response too? Otherwise very simple (but well matched) universals would have better sound stage than they do. Or is it possible to have a good sound stage with any tuning (warm, bright, U, V etc.)?

Its all a matter of personal taste when it comes to the FR graph. Thats a completely separate data point than the human ear listening and HEARING the clarity of whats coming through the ear canal. Its a special piece of art that can make the Philharmonic sound impecable while at the same time radiates an 808 for an electronic/ techno musician. A darker soundstage will offer more subdued with a clear, precise low end that isnt muddy. While a brighter signature will provide airy-ness or breath. Im trying really hard not to sound like a wine aficionado trying to describe what they hear. The listening experience is separate from the graphs, testing, and formulaic laden components.

 

[swtnate]

For those of us using Citubox or Lychee Slicer for 3D printing, I randomly came across an article discussing .ctb files. Inside of Lychee Slicer, for example, once you have imported your model, oriented it for the least amount of islands and added all the necessary support structures you come to the export tab. Click on the name of your printer under within the Export Slices window. From there, click on the picture of your printer at the top. This will open up the image, volume, screen, and other parameters unique to the printer you use. At the very bottom there is a drop down menu for “CTB VERSION”. For my elegoo Saturn, it was set to V2 by default. However, the Saturn, Mars, and many other printers that support the CTB format, allow for the newest, V4, to be used.

once you select V4, close that box and click to edit settings for your selected resin. THERE IS A SIGNIFICANT UPGRADE using V4. You are able to utilize a gamechanger setting, TSMC (Two Stage Motion Control). Now you can have the machine slow down the build plate for a short distance then increase the speed to 150 mm/m. So awesome when you read up on how to set this up correctly. Now, the plate wont rip the model off by lifting or retracting too fast too quickly. Also, cut my print time from 4 hours to 2 hrs and 15 minutes.

 

[Wgibson]

I think you need ultra high and ultra low for soundstage. No real need for explanation, like a regular movie vs Imax.

I've got some 38D2XJ's (not a typo) and some Sonion EST's on hand, and a few planars. Little short on free time unfortunately.

 

[BotByte]

Hey ya'll. I started DIY IEM stuff after loving DIY earbuds. It's a totally different beast compared to the plug+play ease of earbuds. So much knowledge from everyone on this board, so thankful for it all!

I have a stupid question: Does SMB capacitor size matter? I ordered resistors and capacitors in 805. Resistors work perfectly fine. However, after wiring up a capacitor, nothing. I tried testing continuity, but nothing. I see everyone using 1206 caps, so I'm confused if the size matters in terms of voltage they can take, or something.