[Furco]

Start small. You need to be successful at creating shells which is a big hurdle to get over. Next you need to know your limits with regards to soldering miniature components. As was said 10 times before, DIY IEMs are not cheapest route but it can be the most rewarding. Follow the Instructables guide for the methods the guy used for the different build phases. For the actual drivers, start cheap with a single driver and then as you gain confidence get more advanced.

--cheers

Hey Guys,


I 'm pretty new to this thread, but as I was searching on the web for instructions to building a CIEM, I came across this thread at instructbles.com :

http://www.instructables.com/id/DIY-In-Ear-Monitors/


It sounds like as if it is a pretty good home made CIEM, but since I am not familiar with the settings used, can you take a look at it and see if it is a appropriate setup?

 

[piotrus-g]

Hey sports fans!
I took some time to measure the previous design. Turns out I was a bit wrong regarding driver connection lol.
 
Anyways here's updated scheme plus graphs. This will be definitely bass heavy IEM with V shaped sound, but CSD looks pretty clean so the resolution should be quite good IMHO.
Measurement into IEC711 with 0.1 output impedance with design on the scheme.


 

 

 

[tranhieu]

That's one hell of lower treble boost lol

 

[piotrus-g]

Alright here's the second one more fancy, more neutral (still quite fun! but more on the sub-bass side)
It's 4 way design for those who don't fear soldering since it's more complicated.



My recommendation ... don't use two CI-22955 if you plan on using only one TWFK

 

[pc27618349]

Yeah, I've discovered the heavy bass of the first design LOL
 
I ended up using decently high resistor values - 82 ohms on both CI drivers, with a 2.2 uF cap on the TWFK and a 22 uF cap on the filtered CI. This yields a cutoff point for the filtered CI of about 88 Hz, which makes it pretty much a legitimate subwoofer hahah. The resulting signature is pretty balanced with a sub-bass boost. The mids are still a bit wonky (snares don't really sound natural) but are pretty decent nonetheless. 
 
@piotrus-g What program do you use to simulate these graphs? Or did you actually build the setup and run measurements on it?

 

[MuZo2]

Its from ARTA software and measured using iec711 coupler.

 

[piotrus-g]

 
Yeah, I've discovered the heavy bass of the first design LOL
 
I ended up using decently high resistor values - 82 ohms on both CI drivers, with a 2.2 uF cap on the TWFK and a 22 uF cap on the filtered CI. This yields a cutoff point for the filtered CI of about 88 Hz, which makes it pretty much a legitimate subwoofer hahah. The resulting signature is pretty balanced with a sub-bass boost. The mids are still a bit wonky (snares don't really sound natural) but are pretty decent nonetheless. 
 
@piotrus-g What program do you use to simulate these graphs? Or did you actually build the setup and run measurements on it?

yeah, my second design has cut of at 75Hz, but you still need to add resistance past the LP to attenuate CI even more. Also, check the new wiring scheme, because in the one I posted few pages back, there was a phase issue which is why the mids are probably what you are describing. Those designs above sound pretty decent unless you after like super-hi-fi sound, though in such case the driver selection may not be right here.
Overall 2x CI-22960 (easy to buy) could be much better choice here.

 

[pc27618349]

  yeah, my second design has cut of at 75Hz, but you still need to add resistance past the LP to attenuate CI even more. Also, check the new wiring scheme, because in the one I posted few pages back, there was a phase issue which is why the mids are probably what you are describing. Those designs above sound pretty decent unless you after like super-hi-fi sound, though in such case the driver selection may not be right here.
Overall 2x CI-22960 (easy to buy) could be much better choice here.

 
I see. I just ordered the parts and I'll try out the new crossover. The mids just don't sound right in the first design. Also, I learned my lesson and bought through-hole components rather than trying to squeeze SMD components in a non-SMD application... soldering them was a nightmare.
 
I'm guessing that the CI-22960 are better because of their higher impedance?
 
I already put in the order for the new parts, but would simply flipping the polarity for the TWFK (along with the high-pass cap) on the first design help the odd mids problem?

 

[wjp007]

My latest iteration. After at least five tries, I think I hit the best sound yet. I can honestly say this is the first time I feel my diy customs really sound better than my universals. I used a four driver setup. I never believed anything more than three drivers would make a difference, but I'm surprised how good they sound.
1) CI for base, 10ohm, and two filters (yes two) red 3300, yellow 4700
2) Combo driver from ADDIEM (ED and WBFK) 4.7uF
3) WBFK 1uF

I forgot to put a filter on the WBFK. So they are a bit harsh. My next iteration will be to add a grey filter for the WBFK.
.

 

[ForceMajeure]

My latest iteration. After at least five tries, I think I hit the best sound yet. I can honestly say this is the first time I feel my diy customs really sound better than my universals. I used a four driver setup. I never believed anything more than three drivers would make a difference, but I'm surprised how good they sound.
1) CI for base, 10ohm, and two filters (yes two) red 3300, yellow 4700
2) Combo driver from ADDIEM (ED and WBFK) 4.7uF
3) WBFK 1uF

I forgot to put a filter on the WBFK. So they are a bit harsh. My next iteration will be to add a grey filter for the WBFK.
.

Nice, Do you use measurements for tuning? 
What kind of sound did you achieved, how do they sound like, do you have measurements? 
How many bores are there? hard to tell from the pic if 2 or 3?
What material did you use for the shells and faceplates? 

 

[wjp007]

I had some thin rosewood veneer I used for the faceplates.  The shells are made of UV adhesive I got on ebay (Kafuter).  Measurements are below.  Fairly flat.  
 
 

 

[ForceMajeure]

That look good, the rolled of bass is due too lack of seal with the coupler?

 

[wjp007]

That look good, the rolled of bass is due too lack of seal with the coupler?

yes, I had to use blue tack for the finished units.

 

[pc27618349]

HOLY COW @piotrus-g the new crossover works wonders. Everything has a realistic timbre to it now - the snare especially sounds much more natural. The soundstage has expanded greatly now and there's a lot more air. The bass has been pulled back, but still has great impact and texture while letting every other part of the spectrum sing - and really being authoritative when called on. The mids are in perfect balance - not recessed yet still giving the bass and treble room to rumble and shine. Thanks for your help!!
 
The soldering needed for this 4-way was quite insane:

 

[piotrus-g]

Yay! great you were able to solder it. It looks hardcore indeed. I use SMDs so it didn't look as scary
 
Anyways I hope you'll enjoy it!