I’ve had experience with ED29689 only, other models are hard to get. I do wanna try Sonion 2389, except they are soo fucking expensive. Also want to try 28uap (micro woofer) and E50 series, but i heard they tend to sound dry?
The 23xx series from Sonion is what I would do. Fuck price, sound is all that matters.
i’m a graph head, I tried a subwoofer cut, but it blended back in
I tried another subwoofer cut but, it makes the mids seem thin…
Edit:
A preview of what’s to come.
That’s ED29689 after impedance correction
an absolute resistor correction
The build is nearing its end. I’ve reconfigured the system.
The truth about Impedance with an RC system, The impedance is not really flattened at the beginning, so to work with that, The base impedance is centered around the Mids driver and i cannot exceed 5.6ohms of the ED29689, the sound will change starting with 1.5k boost. So the way it is right now, the branch impedance at best will end up near 12ohms?
so total iem impedance when correcting and connecting each branch, all ends up at 4ohms with a +/-0.1ohms across the FR impedance. The thing is, at this point the iem will be a tad less sensitive, like 1~2dB below the usual variations, Add another 4ohms, that will drop by 6.5dB or so…
Edit: We are looking at a 7dB loss over variations original sound for an immune impedance system
I’ve been an advocate for safe impedance on iems. In this case, the amps and sources my client is surrounded by are less than 1ohms output impedance, so I will go ahead with a 4~5ohms build that’s equally sensitive to variations in terms of dB/mV, it will be about 3~4x more current hungry tho….
That is indeed a very good series.
Here is a diy i made with the Sennheiser IE800 (clone) shell with a custom 3d printed part to fit the 2356 sonion BA in it.
Oh and here is the same concept but with the E-Audio EB-32257, which iirc is a clone of the Knowles RAB-32257.
This was mainly to test a concept i had, if its possible to have a comfortable BA only iem with vented shells. Since most of BA only iems I have seen dont have a vent and if it does it doesnt do much to help with pressure build up…
This actually is very comfortable.
The only real way to do a vented IEM is with either large DDs, Planar, or just a driver with extreme bass. Around 16-20db of subs is enough to vent the piss out of them down to near 0 and still actually hear a lot of them.
Yeah that does indeed seem to be the case if you do want something without rolled sub.
Compared with a single BA in non-vented setup though.
That F3100 is absolutely dogshit though.
Guess what moondrop did to high pass the mids BA
Vented out enough bass to have the mids BA playing post 150hz
DD also got vented out near 6~7dB or even more
There’s vent damper, the crossover space and then the faceplate exit vent,
bass is vented out, isolation is retained
Explains why the subs feel weak down low.
I thought you realized that. I changed the vent damper to a low value and left the dampers off the DD entirely muhahah. Then I drilled a hole in the back plate and put a nice grill on it. If you dont do that the DD doesnt ‘open back’ enough.
The back plate is hardened steel so you need a bit and oil. JFC, annoyed me.
I was half ok with DD having to fight the vent, The BA however, when I saw the loss, i was just disappointed. Variations doesn’t have low mids, I thought it was the crossover, but the BA loosing out to the vent? That’s crazy, BAs already have a hard time doing bass, the Mid-Bass even with the tuck could have been better if it wasn’t for the damn vent.
Fuck, I’ve been off in calculations, No wonder Helios is that crazy low sensitivity to drive. Jeez. I could achieve flat impedance to a resistor level on BA or any other individual drivers, but for the whole iem, the loss is fucked up.

remember all the sound is vented through that vent, dont matter what it is. anyways you can just make the DD do more, like I did if you want
Nah, I decided to double down on the vent, I’m going for clean extension, and its going all the way to the infras. The vent has double 600 stacked, 1200 it is.
I’m still confused as fuck as to how anomaly gets a sudden infra boost.
The driver has 16+db of subs is how. My venting is precisely calibrated to wavelength, so as the frequency drops below the vent diameter (0.4mm x 2) IE, longer waves, it cant escape. Phantom subs. The approx vent has to be about 4.5-5mm long and 0.4mm wide. It took me massive iterations to get it right.
That, scientifically makes sense…
it is an ellipse?
the shape doesnt matter, the length and air volume does
One last thing, quite a bit of high mids and treble can in fact go through the vent.. they dont entirely see a sealed chamber. Also this vent is AFTER the final damper so it is not normal. It’s exactly like an earbud or something that cant achieve a full air seal.
Did you create this?
The reason why I’m struggling is 3Ohms isn’t a good number, sensitivity falls as low as a Helios but even worse whole iem impedance.
from this i can guess that they use higher impedance drivers inside. one of the drivers is CI-22955 (rumour) that is 47ohms and above.
The problem when it comes to the current drivers
DD - 16ohms
ED - 3.8ohms baseline 5.6ohms flattened treble impedance
Planar - 15.5ohms
The damn ED Branch is dragging everything down to hell.
Flat impedance single BA is easy asf
Flat impedance iem such that even crossovers can’t shift is pushing numbers so low. It is insanity.









