[talyak]

Look for sony BAs, or find some used and pull yourself, some of those are open / no spout. If you twist the nozzle off a 30095 just right, you can pull the whole side off with it. Test on dead/scrap drivers, it is not a safe method...

I'll give that a go if the sandpaper method proves too inconsistent

 

[piotrus-g]

now i just need UPS to stop holding on to my drivers, so i can actually test what the feds are like.

reading a little more on what subtonic are doing with their slam drivers... they are basically doing exactly what we were talking about the other day with a dual driver setup using them as individual bass drivers with a crossover on each. but they are also venting the sub-woofer, to maximise that extension. they are using knowles dual woofers. my first guess is the rdi-33438, as that seems like the driver that i think most closely matches what they are describing (or possibly one of the dtec pairs), but this is pure speculation. apparently the drivers initially have coupled contacts, so you have no choice but to use a single crossover for both, and i believe thats what they are talking about when they talk about uncoupling. they talk about the mid-bass driver having an internal chamber, but i think this just means that it doesnt have an external chamber and isnt vented. so, my guess is that to achieve the same things they are, the trick will be to vent one of a dual driver unit, and restrict that vented driver to solely sub-bass. allowing the mid-bass to display maximum speed, while having the slow rumble/decay of a vented driver in the sub-bass. obviously i dont have drivers on hand to test this, so its all speculation for now.

I don't know what other brands are doing in terms of electrical wiring so I won't be speculating on that
If I were to use RDI I could see an advantage of using single units of that driver in combination with some clever crossover. RDI is extremely loud to the point I find no use for it in IEMs because it overpowers everything. I think it's about 10dB louder than 38D1XJ. Now probably wiring each half separately could work. A half of the RDI should still be louder than whole 3800 hahaha

 

[swtnate]

even if they get drivers customised from a manufacturer, they still need to get to that point via experimentation. sure 64 audio have an exclusive contract for the tia drivers, but first they found out what they wanted via modifying off the shelf drivers. there's no way we can achieve the quality assurance of mass produced drivers, but there is no reason we can't get say 90% of the performance by modifying off the shelf drivers in the same way.

their designs don't just appear from nowhere. and knowles/sonion arent about to do the research themselves then limit sales to a single brand. we can glean most of how they managed it via marketing materials. both subtonic and 64 audio have written and posted articles on how they achieve their performance. im just making plans to replicate what they said they did, and see if i can get similar results.

I dont disagree with any of that. Its a heck of a lot experimentation, testing, re-testing, etc. and, anything with a 7-way crossover would raise my eyebrows on determining how they are defining that. If you want big sub bass that is deep, full, and has tons of headroom with an immaculate rolloff, the Sonion 38 with acupass/ ferrofluid solves all of that. Ive never been one to “experiment” by potentially damaging a fully functioning driver. Sanding them, expanding back vents, etc. has always been chases down rabbit holes for me. These drivers do amazing things and external dampers its a huge part of the puzzle. Why add all these components to a tight space? Then you’ll start down the electronic phase correction, then acoustic phase correction, followed by polarity reversals, splitting drivers, big tube lengths, short tube lengths, resonators, and the list goes on and on. The lessons Ive learned starting into this from 0 knowledge to where I am now, build from what you know, grow from what youve learned, and establish your own design profile, develop a strategy to hit that, then build. Theres a LOT of gamesmanship and marketing lingo thats customer facing but, the actual implementation plus being repeatable, is the key. Im assuming though that no one would go through the gauntlet of product development and construction to only build it once…. Thats my two cents. i would rather develop builds so people can hear how my brain defines a soundstage.

 

[swtnate]

not to mention the fact that im not trying to replicate their designs 100% accurately. im not trying to make an off-brand u12t or storm. im not trying to have my iems sound the same as theirs. i have different design requirements and am just looking for the best ways to get the best possible performance within those requirements. part of the reason im doing this research is specifically because i cant just throw 12 ba drivers in an iem and call it a day. my design is limited to 4-6. plus i only need a couple of pairs of my end-game design to last me years. so if i can get something that sounds almost as good as a 64 audio iem in the highs and almost as good as a subtonic/symphonium iem in the lows, i don't mind if i end up killing a few drivers in the process of experimenting.

In that case, 6 drivers, 2 sound bores - 1 bore is 2.16mm the other 3ish mm - using a Sonion 3300, Twfk, and an SWFK. Super thump, headroom for days, and a high end extension thats clear and present with distinct separation. Message me and I will give you my 2nd to last beta design, as the final design is proprietary information. You can tweak it how you want from there. My strategy has always been to seek the proper drivers that compliment each other without stepping on each other. Ive never once used a 1st order crossover in any of my wiring. I will use, however, L-Pads and Zobels in conjunction with series resistors and capacitors. @piotrus-g comments saved my sinking ship(s) a couple of times. Im by no means an expert as there are others in this group with so much more build experience that I could never catch up to. I have learned tons of lessons, had tons of failures, and chased sound ideas only to have them crash and burn. Im not by any means trying to discourage you or say your doing anything incorrect. I just know from the 12 or so driver configurations I have designed, then re-designed, the re-imagined, that its very easy to invest so much of yourself into a project only to not attain what you wanted in the end. Its deflating to lose all that time, investment in parts, machinery, documentation, etc. only for it to just be bland.

 

[swtnate]

Its also why I ate a gallon of BlueBlue Chocolate Chip Cookie Dough after hearing the sub response be strangled in the last 10 driver design. I poured over that for months. I still have a little PTSD from the experience.

 

[talyak]

i dont think there is much need to be scared of crossovers when there are such useful tools available for designing them like xsim. and to be honest, they arent exactly complicated to get your head around if you have ever designed a circuit before. so even if i was going to use that 6 driver design, i probs would still give designing a crossover for them a go. plus if you dont have drivers uselessly producing sound in other drivers range, you are less likely to have phase cancellations. (not to mention that having an swfk and twfk im pretty sure would be too big for the completely in canal desing im working on)
eh, regardless, im not gonna restrict myself to easy when experimentation will allow for potentially much more interesting results.

 

[talyak]

worst case scenario, i waste a bit of money, and learn a lot about what you can and cant do with these drivers. better to test ideas and fail, than to never step out of your comfort zone and always be stuck in the realm of mediocre results

 

[talyak]

im working on this project because there is literally nothing on the market that matches my requirements. im not working on this project to build a knock-off dunu sa6 with worse electronics design lol

 

[talyak]

I don't know what other brands are doing in terms of electrical wiring so I won't be speculating on that
If I were to use RDI I could see an advantage of using single units of that driver in combination with some clever crossover. RDI is extremely loud to the point I find no use for it in IEMs because it overpowers everything. I think it's about 10dB louder than 38D1XJ. Now probably wiring each half separately could work. A half of the RDI should still be louder than whole 3800 hahaha

that's good to know!
if i get a pair and attempt it, ill definitely post it here.

 

[stegeoc]

Its deflating to lose all that time, investment in parts, machinery, documentation, etc. only for it to just be bland.

Couldn't agree more with this. The most complex of my builds rarely lived up to expectation and ended up getting recycled or destroyed in the process. There is so much going on and tiny lessons that you learn only by trying things and probably won't read here. There have been times when I've convinced myself that an experiment worked really well, only to eventually concede that it could be be better and won't get used

 

[talyak]

i mean... yeah, that definitely is the nature of experimentation. you are bound to get more losses than wins. but when you do get a win, its that much sweeter.

the 33518s are like 15usd a pair. worst case i lose 15 dollars or so and an hour sanding them down. id love it if it works, but if it doesnt, im not losing that much. the RDIs are more expensive tho. so i would be a little sad if they dont turn out to be worth the effort. but then i can just do what everyone here is saying and just get the sonion 38s with acupass, no biggy. if the experiments work, they open up new posibilities. not just for me, but for anyone who reads about them in future. if they dont work, all that is lost is a bit of time and money. and im willing to eat that if it means i learn for myself more about what can and cant be done. upside of chasing my curiosity, as long as im doing it, im having fun lol

 

[citiesandshapes]

Hi all-looking for some advice. I have recently gone down the rabbit hole of DIY IEM. I've managed to successfully make a pair of custom fit IEMs GK 31732 drivers, and they sound good and fit great, but I find myself wanting more. So I've been diving into the thread looking to get into a more advanced build. I've looked into the CI X TWFK build by @piotrus-g, but I'm having a hard time wrapping my head around crossovers and resistors and what not as I'm new to this world.
So, looking for suggestions on how to get into a more advanced build. I'm a drummer/vocalist and will be using IEM primarily for live monitoring, but would ideally like something that could be used for daily use and monitoring.

 

[talyak]

Hi all-looking for some advice. I have recently gone down the rabbit hole of DIY IEM. I've managed to successfully make a pair of custom fit IEMs GK 31732 drivers, and they sound good and fit great, but I find myself wanting more. So I've been diving into the thread looking to get into a more advanced build. I've looked into the CI X TWFK build by @piotrus-g, but I'm having a hard time wrapping my head around crossovers and resistors and what not as I'm new to this world.
So, looking for suggestions on how to get into a more advanced build. I'm a drummer/vocalist and will be using IEM primarily for live monitoring, but would ideally like something that could be used for daily use and monitoring.

a great tool to get your head around crossovers and shiz is xsim. you can layout the ci twfk circuits and see what they do to frequency responses and impedance curves.

 

[talyak]

@dhruvmeena96

heya. im building an ety fed with a 3d printed shell over the weekend. did you mean series or parallel 75 ohm resistor on this post?

@eunice

Finale 1 is out(design for diy'ers)
And I think most of the people would be disappointed.
But I must tell you, its great.

Direct replacement for ER4 but you guys will need @DannyBouwhuis help

Its a concept, and I think he can do it

FED30048 + 75ohms + a spout

In etymotic shell with etymotic tips
No damper

Without damper, I think it will sound what etymotic should have sounded(still a great pair, but without damper, its a different story)

MASM 7 / Pro 7 Series coming soon

MASM 7 pro would be Pro 7 Rebirth

also, my current 3d model has a 2mm ID 14mm long sound tube. do you have any recommended tweaks to the shape, or should this be good enough?

 

[StephenJames]

Hi all, I'm new around here, but I've been reading and learning from all the information on here for a while.

I have an issue, where I think my ear may be the problem, and not the CIEMS I've made.
I use mine for singing on stage, and even with a good fit (I made sure I used a bite block when getting ear impressions done), I always get a swoosh/phasey thing happening in my left ear. Right ear seems ok.
If I use them for anything other than singing, they're great.

I also have a pair of ACS in ears, and they also do the same thing on my left ear, which makes me wonder if my ear is the problem. I understand there is movement in the ear canal, but I can't really perform when the frequency response is shifting all over the place dependant on how open my mouth is.

I'm really enjoying the process of learning how to making these, but wonder if my time is better used making some universal fits. I've currently gone back to using KZ's with comply tips on stage, because as much as I don't like the peaky sound signature, I can EQ them a bit, and never have the swooshy/phasey problems.

Has anyone come across a situation, where CIEMs just don't work for you, or a client? Regardless of how well everything fits and how well it's made?