Home-Made IEMs
Feb 13, 2015 at 4:35 PM Post #3,198 of 16,074
Curiosity got the better of me and I picked up a calibrated mic to do freq. response testing. Test results will be coming that compare my old test rig to this one.

Clear tube at front of mic is 2cm^3.


I've built a testing rig as well..somewhat similar to your previous one. A major flaw I saw with your previous rig, and possibly this one as well, is that it is an absolute MUST to get a proper seal for the mic and headphone you are testing. I have noticed that if you don't have a proper seal, your HF readings will increase, and the LF readings will be greatly decreased. 

I'm also using the Presonus Audiobox USB, and the Dayton Emm-6 mic. 

At first I didn't have a proper seal, and the bass readings were really, really low...and the highs were a bit weird. 
 
Not properly sealed:
 

Sealed:

 
I then compared my graph to those on the net for the same IEMs (FXZ200), and they're pretty much spot on
 
Feb 13, 2015 at 6:29 PM Post #3,200 of 16,074
On my first rig, I got a decent seal when I held the cIEM up against the silicon adapter like so:



I made sure that no openings were obvious between the nozzle and the hole.

With my new rig, I'm using a wad of 3M putty, which is basically the same as Blu Tack. I form the putty around the nozzle, squish it to ensure no leaks and then cover the entire opening but don't let the nozzle past the "reference plane". I definitly noticed the bass freqs getting attenuated if there's an unsealed gap between the putty and tube.

But......

I can't get a good match of the freq response between my rig and what's been posted elsewhere for a pair of Metrofi 220's I have laying around. The high frequencies are not even close but things up to around 4kHz are close. So now I'm studying up on the IEM 60711 artificial ear. I did stumble across the the Zwislocki couplet which looks like it might be DIY-able.

It maybe I'll just hire acain to make it for me considering he's got access to some sweet equipment.
 
Feb 13, 2015 at 6:33 PM Post #3,201 of 16,074
I've built a testing rig as well..somewhat similar to your previous one. A major flaw I saw with your previous rig, and possibly this one as well, is that it is an absolute MUST to get a proper seal for the mic and headphone you are testing. I have noticed that if you don't have a proper seal, your HF readings will increase, and the LF readings will be greatly decreased. 


I'm also using the Presonus Audiobox USB, and the Dayton Emm-6 mic. 


At first I didn't have a proper seal, and the bass readings were really, really low...and the highs were a bit weird. 

Not properly sealed:



Sealed:



I then compared my graph to those on the net for the same IEMs (FXZ200), and they're pretty much spot on
Two words, poster tack. Works great.
 
Feb 13, 2015 at 7:03 PM Post #3,202 of 16,074
On my first rig, I got a decent seal when I held the cIEM up against the silicon adapter like so:



I made sure that no openings were obvious between the nozzle and the hole.

With my new rig, I'm using a wad of 3M putty, which is basically the same as Blu Tack. I form the putty around the nozzle, squish it to ensure no leaks and then cover the entire opening but don't let the nozzle past the "reference plane". I definitly noticed the bass freqs getting attenuated if there's an unsealed gap between the putty and tube.

But......

I can't get a good match of the freq response between my rig and what's been posted elsewhere for a pair of Metrofi 220's I have laying around. The high frequencies are not even close but things up to around 4kHz are close. So now I'm studying up on the IEM 60711 artificial ear. I did stumble across the the Zwislocki couplet which looks like it might be DIY-able.

It maybe I'll just hire acain to make it for me considering he's got access to some sweet equipment.

The problem with that rig in the pic is the foam. it will let in ambient noise, and let out some sound produced by the iem.

Something else I have found (which I havent seen being mentioned elsewhere), is that when you are doing the loopback to calibrate your sound card, be sure to switch on phantom power. I have found that it does add a very tiny amount of "noise", which can affect the final readings
 
Feb 16, 2015 at 12:49 PM Post #3,205 of 16,074
 
CI+TWFK
 
Sometime ago I said I would prepare "open source" design and I already have one but I have no time to create the post, so if you want it let me know I will give you directions.


Thanks that would be great, I am good with building the shells I am terrible at crossovers. I used the CI+Ed for one build and it sounded pretty good the mids were too laid back for me.
 
Feb 16, 2015 at 1:29 PM Post #3,207 of 16,074
Thanks that would be great, I am good with building the shells I am terrible at crossovers. I used the CI+Ed for one build and it sounded pretty good the mids were too laid back for me.


If you want a mid forward design you should try out my most recent quad. It is a 3 way using mostly acoustic crossovers.
high is RAB
mid is DTEC
low is CI

You can find the details in some of my previous posts.

It is a very mid forward design that is perfect for Jaz or blues but not so great for rock.

I have been thinking of making another version of it by using WBFK instead of RAB and adding a 2nd CI.
 
Feb 16, 2015 at 3:39 PM Post #3,210 of 16,074
If you want a mid forward design you should try out my most recent quad. It is a 3 way using mostly acoustic crossovers.
high is RAB
mid is DTEC
low is CI

You can find the details in some of my previous posts.

It is a very mid forward design that is perfect for Jaz or blues but not so great for rock.

I have been thinking of making another version of it by using WBFK instead of RAB and adding a 2nd CI.

 
I'm amazed at just how powerful the DTEC is.  During some of my many tests with understanding crossovers,, it was a practice in patients to try and figure out how to get a DTEC to play nice with other drivers. 
 

Users who are viewing this thread

Back
Top