or Connect
New Posts  All Forums:Forum Nav:

Home-Made IEMs - Page 116

post #1726 of 5835
Quote:
Originally Posted by piotrus-g View Post
 

Yes you can, but have to have resources to make it work correctly.

What kind of resources do you mean? I have prototyped active filters with opamps before, and I understand the theory behind the passive filters. Equipment wise I can etch circuit boards and solder on the SMD resistors/caps so I can fit in a more complex crossover.

 

I'm still considering MuZo's words of wisdom, and perhaps go back to a TWFK/DTEC combo like what's in my old W4. It's just so tempting to experiment with the CI for the low end.

 

The main thing is that CI+WBFK = TWFK in terms of cost, and the downside is added complexity in crossover design.

 

I've already read all the documents from the other thread, and come to the conclusion the crossovers are really just for cutting the response very roughly, and not for tuning. The acoustic properties of the material I use, sound tube length, and tube diameter seem to impact the response even more than what I can implement with passives. I'll finalize a design and post it up here for comments, then see about ordering everything next week.

 

I'll attempt something I saw a few pages back with a separate active filter/amp in the future.

 

*One thing though, where can I find this acoustic tubing for the drivers from Asia?*


Edited by vector1 - 5/31/14 at 5:18am
post #1727 of 5835
Quote:
Originally Posted by vector1 View Post
 

What kind of resources do you mean?

You need measuring equipment. Working with TWFK especially with DTEC+TWFK is really tricky if you use regular DTEC-30088 with 3300 Ohm damper and slap 1-2uF on TWFK-30017 it will create gap in frequencies due to phase canceling. It's not really easy to work with those drivers, Especially TWFK is really hard to get right - to avoid harshness and sibilance but still get decent highs.

 

BAs don't have flat phase or impedance and they will change dramatically depending on tubing, damping etc.

 

Quote:
 The main thing is that CI+WBFK = TWFK

Buy GK unit and you'll be set, there's on board crossover and all you have to do it do some acoustic tuning and you'll get good sound for $100/pair.


Edited by piotrus-g - 5/31/14 at 7:52am
post #1728 of 5835
Quote:
Originally Posted by piotrus-g View Post
 

You need measuring equipment. Working with TWFK especially with DTEC+TWFK is really tricky if you use regular DTEC-30088 with 3300 Ohm damper and slap 1-2uF on TWFK-30017 it will create gap in frequencies due to phase canceling. It's not really easy to work with those drivers, Especially TWFK is really hard to get right - to avoid harshness and sibilance but still get decent highs.

 

BAs don't have flat phase or impedance and they will change dramatically depending on tubing, damping etc.

 

Buy GK unit and you'll be set, there's on board crossover and all you have to do it do some acoustic tuning and you'll get good sound for $100/pair.

Regarding phase canceling, what I understand is that a 180deg shift is needed, but if I'm only implementing a 1st order, maximum phase change I'll see per filter is 90deg past the cut off frequency. Will the output be so significant that with a 6db/decade rolloff it'll still cause issues?

 

I could shift the cutoff frequencies away from the regions of issues, especially near the measured peaks from the literature.

 

I won't only use acoustic filters, rather a network of first order filters.

 

I am guessing that I should be able to exploit the voltage divider in the filters to help attenuate each driver relative to each other, but I'd have to buy quite a few values of capacitors if I want to maintain the same cut-off frequencies when I'm testing.

 

It's not that I want a simple earphone, as I could just go buy a set of $200+ earphones. It's more on the lines of a side project for me so I want to go the distance :)


Edited by vector1 - 5/31/14 at 10:20am
post #1729 of 5835
Quote:
Originally Posted by vector1 View Post
 

Regarding phase canceling, what I understand is that a 180deg shift is needed, but if I'm only implementing a 1st order, maximum phase change I'll see per filter is 90deg past the cut off frequency. Will the output be so significant that with a 6db/decade rolloff it'll still cause issues?

It's the 1st order filters that will create those canceling. As I'm MOT I cannot talk about other manufacturers but check out Rin Choi's blog to see what I'm talking about here

 

Quote:
 I won't only use acoustic filters, rather a network of first order filters.

Then it's going to be a real tough-y

 

Quote:
 It's more on the lines of a side project for me so I want to go the distance :)

I totally understand just know the distance it ten times longer than you expect. Trust me on this. Working w/o measuring equipment vs. working with measuring equipment is like night and day difference. Without measuring rig you simply don't know what you are doing because you are unable to verify your work, and hearing can be VERY deceiving.


Edited by piotrus-g - 5/31/14 at 10:38am
post #1730 of 5835

I am using the GK driver in my build, I can tell you one of the main frustrations with this driver is the distance between the 2 sound ports is about the same as the thickness of the wall of standard acoustic tubing making it hard to fit 2 tubes over the sound ports but they are far enough apart that I can't fit them both into a single tube.

 

I have resorted to sanding down the tubing walls in order to make them thin enough that I can fit both tubes over the driver ports but it is a tight squeeze and difficult to get the tubing fully flush with the drivers.

 

If anyone knows a source for smaller tubing I would be grateful.

post #1731 of 5835
Quote:
Originally Posted by vector1 View Post
 

Regarding phase canceling, what I understand is that a 180deg shift is needed, but if I'm only implementing a 1st order, maximum phase change I'll see per filter is 90deg past the cut off frequency. Will the output be so significant that with a 6db/decade rolloff it'll still cause issues?

 

I could shift the cutoff frequencies away from the regions of issues, especially near the measured peaks from the literature.

 

I won't only use acoustic filters, rather a network of first order filters.

 

I am guessing that I should be able to exploit the voltage divider in the filters to help attenuate each driver relative to each other, but I'd have to buy quite a few values of capacitors if I want to maintain the same cut-off frequencies when I'm testing.

 

It's not that I want a simple earphone, as I could just go buy a set of $200+ earphones. It's more on the lines of a side project for me so I want to go the distance :)

- BA's phase response in general is very unpredictable, unlike dynamic drivers. A 45 deg shift is more than enough to create a trough in your frequency response. In all honesty you should get yourself a measuring rig to fine tune the BAs.

 

- Avoid using acoustics dampers as much as possible when it comes to tuning the tweeter. Try to move your cutoff frequency further upward instead.

 

- Also avoid using resistors to dampen your woofer as they also cut down your low bass decay considerably. Unless you're into PFE 112 kind of sound signature. 

post #1732 of 5835
Quote:
Originally Posted by CMOS1138 View Post
 

I am using the GK driver in my build, I can tell you one of the main frustrations with this driver is the distance between the 2 sound ports is about the same as the thickness of the wall of standard acoustic tubing making it hard to fit 2 tubes over the sound ports but they are far enough apart that I can't fit them both into a single tube.

 

I have resorted to sanding down the tubing walls in order to make them thin enough that I can fit both tubes over the driver ports but it is a tight squeeze and difficult to get the tubing fully flush with the drivers.

 

If anyone knows a source for smaller tubing I would be grateful.


There's something like tubing expander I'm pretty sure it's quite common tool though.

http://www.westone.com/store/index.php/professional-supplies/tubing-cements-supplies-accessories/tubing-expander-pliers-curved-2442.html

 

using this tool you'll be able to put one 2mmID tubing on two drivers' spouts.

 

I'm pretty sure GK is designed to be used with single tubing or in universal IEMs where they are firing into designed silicone sleeve (like UE uses) or in plastic body (like Westone/Shure).


Edited by piotrus-g - 5/31/14 at 11:47am
post #1733 of 5835
Quote:
Originally Posted by piotrus-g View Post
 

It's the 1st order filters that will create those canceling. As I'm MOT I cannot talk about other manufacturers but check out Rin Choi's blog to see what I'm talking about here

 

Then it's going to be a real tough-y

 

I totally understand just know the distance it ten times longer than you expect. Trust me on this. Working w/o measuring equipment vs. working with measuring equipment is like night and day difference. Without measuring rig you simply don't know what you are doing because you are unable to verify your work, and hearing can be VERY deceiving.

 

That was interesting, since it uses TWFK and DTEC per W4 I had.

 

I think it's something that I have to feel the pain of working without measurements before I'll understand the need ^^

 

But the thing is, I thought the calibration of the IEM ultimately comes down to the individual, so as long it's not too harsh/bloated for me, I hope it'll turn out fine anyway.

 

Even so, could I trouble you to direct me to any resources/shops that I can possibly get these measuring equipment from? And is the cost going to be more than the headphone drivers it self?

Quote:
Originally Posted by tranhieu View Post
 

- BA's phase response in general is very unpredictable, unlike dynamic drivers. A 45 deg shift is more than enough to create a trough in your frequency response. In all honesty you should get yourself a measuring rig to fine tune the BAs.

 

- Avoid using acoustics dampers as much as possible when it comes to tuning the tweeter. Try to move your cutoff frequency further upward instead.

 

- Also avoid using resistors to dampen your woofer as they also cut down your low bass decay considerably. Unless you're into PFE 112 kind of sound signature. 

1) I can see why my friend was talking about getting a measuring rig first. It seems to be the first thing everyone says when doing DIY.

 

2) I understand that, but I might still get a selection of filters to test if they're not too expensive.

 

3) I'm not sure how do I simulate that in pSpice :(. Do you mean resistors in series? Because I was referring to the impedance mismatch between input, filter and woofer that can be used as a voltage divider if the response is known for that frequency range. I know that it ought to be a very low ratio in order to 1) reduce attenuation 2) improve response.

post #1734 of 5835
Quote:
Originally Posted by piotrus-g View Post
 


There's something like tubing expander I'm pretty sure it's quite common tool though.

http://www.westone.com/store/index.php/professional-supplies/tubing-cements-supplies-accessories/tubing-expander-pliers-curved-2442.html

 

using this tool you'll be able to put one 2mmID tubing on two drivers' spouts.

 

I'm pretty sure GK is designed to be used with single tubing or in universal IEMs where they are firing into designed silicone sleeve (like UE uses) or in plastic body (like Westone/Shure).

that is a great idea. I will experiment and see if I can expand the tubing a bit with the tools I have on hand before ordering a specialty tool though. I have already spent far more than I intended on this project.

 

I think you are correct about GK being designed to work with a single tubing because (at least to my ears) I have not been able to significantly change the sound signature by using dampers (white on the TWFK and red on the CI,) I only succeeded in making it quieter.

post #1735 of 5835
Quote:
Originally Posted by piotrus-g View Post
 

It's the 1st order filters that will create those canceling. As I'm MOT I cannot talk about other manufacturers but check out Rin Choi's blog to see what I'm talking about here

 

Then it's going to be a real tough-y

 

I totally understand just know the distance it ten times longer than you expect. Trust me on this. Working w/o measuring equipment vs. working with measuring equipment is like night and day difference. Without measuring rig you simply don't know what you are doing because you are unable to verify your work, and hearing can be VERY deceiving.

Hi Piotrius,

 

Great advice. I`m currently working with dynamic drivers and then move on to BAs. My plan is to start with oneway drivers and then build on that experience (2-way, 3-way etc.). Do you have any recommendations (maybe there is a thread already?) for a simple and cheap measurement system to start working on BAs? All these ear simulators are very expensive.

post #1736 of 5835
Quote:
Originally Posted by tomtomtom View Post
 

Hi Piotrius,

 

Great advice. I`m currently working with dynamic drivers and then move on to BAs. My plan is to start with oneway drivers and then build on that experience (2-way, 3-way etc.). Do you have any recommendations (maybe there is a thread already?) for a simple and cheap measurement system to start working on BAs? All these ear simulators are very expensive.

That's good to start off with 1way IEMs, it lets you understand basics of how BA changes on acoustic terminations.

 

As for measurement - you can buy some some condenser microphone, use cut syringe (to 2cc) as coupler. However I've never done it so you should do some research on it. ARTA software is free (in free version you cannot save your measurements).

post #1737 of 5835

For measurement rig you can use

http://www.johncon.com/john/wm61a/

post #1738 of 5835

DO NOT USE SUPERGLUE! I've found it nearly impossible to use it without clogging up something on the driver, be it the nozzle or the side opening, plus it leaves a white film over everything. If the sound is bad even when amped, you may have a cold joint , loose tubing, or the solder pad heated up too much and burned off a little, but not all the way. i tried making my own first pair and completely destroyed them and wasted $300 due to soldering issues and the use o superglue.

post #1739 of 5835
Quote:
Originally Posted by ThatOneGuy37 View Post
 

DO NOT USE SUPERGLUE! I've found it nearly impossible to use it without clogging up something on the driver, be it the nozzle or the side opening, plus it leaves a white film over everything. If the sound is bad even when amped, you may have a cold joint , loose tubing, or the solder pad heated up too much and burned off a little, but not all the way. i tried making my own first pair and completely destroyed them and wasted $300 due to soldering issues and the use o superglue.

that is probably good advice but I have already completed my first assembly using superglue and for me, it solved the problem I was having with the tube detaching from the driver. 

 

My intentional mistake with this build was using 2 tubes one one side and only 1 tube on the other. I wanted to see how much (if any) of a difference it would make to the sound. initial impressions, 2 tubes has a much stronger base response and SPL is higher overall. 1 tube seems to have a more flat response. I need to listen to them for a while longer before I decide which one I prefer.

 

You can see my finished CIEM in my avatar picture.

post #1740 of 5835

Be careful with superglue. I think it leave some fumes so superglue for attaching tubes to shell or something like it might not be safe.
 

New Posts  All Forums:Forum Nav:
  Return Home