New Posts  All Forums:Forum Nav:

Home-Made IEMs - Page 128

post #1906 of 2032

I've used an adaptation of the James-Baxandall tone control network, which has a bass/treble control. Interestingly, I've managed to somehow cut-down the number of components used by experimenting in pSpice. The left and right set of two resistors are actually variable resistors of 10ohm. The output is after the 22ohm resistor, or between the right two resistors.

 

 

The green line is something close to what I'm using to filter the input signal, but note that the output still goes through filters before reaching the drivers. The red line is just another example of possible tuning, but the treble is out of control there. Top circuit is green, bottom is red.

 

I was testing the earphones against shure 846s and JH customs (JH5 and JH11) at Jaben. With this control, I can adjust the response and get something fairly close to a Shure 846 signature, but at the expense of volume. On my sansa clip, I can drive the shures at -40dB, while my earphones need around -10dB to get the same volume.

 

The trick now is to shrink this plus the filters into a circuit small enough to fit into my earphones. At the current count, I have 14 different resistors/capacitors per ear.

 

Incidentally, the Shure doesn't have very nice highs, but really impactful bass. The JHs doesn't have the same kick to the subbass, but more forward mids.


Edited by vector1 - 7/5/14 at 5:22am
post #1907 of 2032

Funny enough I gave up on Pspice after using it for awhile. I couldn't get the sound right by using its simulation but maybe that's just me.

 

Btw 14 components per side are a bit too many. The TG334 for example only has 1 cap.


Edited by tranhieu - 7/5/14 at 9:34am
post #1908 of 2032
Quote:
Originally Posted by tranhieu View Post
 

Funny enough I gave up on Pspice after using it for awhile. I couldn't get the sound right by using its simulation but maybe that's just me.

 

Btw 14 components per side are a bit too many. The TG334 for example only has 1 cap.

 

The simulation gives me a baseline to tune from, and I tweaked the bass/treble controls while comparing to actual IEMS (shure 846) directly at the shop, to get the same sound (roughly).

 

The drivers interact in unexpected ways with the circuits. For example, running a large resistor in front a a driver in the 3 way parallel crossover means almost no current goes into that driver. That means a theoretically workable filter using 1uF and 1kohm won't work at all. It needs to be something like 100uF and 10ohm for example. All of my filters are running at ~5ohm, and I tune via capacitor now. It's something that I only found out after a few days of trial and error. I noticed it when using 100ohm variable resistors, but never actually thought about why it was happening since the sound still comes from the other drivers.

 

I'm going to check out some prototyping facilities locally. If they can handle 0201 SMD (0.6 mm × 0.3 mm) components, then it's quite possible to get all the components on a 3x3mm board, or smaller even. If I have to do it with 1206 (3.2 mm × 1.6 mm) SMD components manually, it's still possible, but I'll need to expand the case slightly. I'm currently nearly 5mm thinner than the shure 846s, so I have some room to grow my earphones. Length and height are similar.


Edited by vector1 - 7/6/14 at 2:32am
post #1909 of 2032

May be you can put them in cable like Shure E5c

post #1910 of 2032

does anyone have a recommendation regarding mounting of tubes to BAs?

post #1911 of 2032
Quote:
Originally Posted by MuZo2 View Post
 

May be you can put them in cable like Shure E5c



I shall see about that, but since my cables are already custom made, I have to order a new set if I go down that route. I might as well implement the entire passive EQ/tone control if I actually do that...

 

Also an interesting point to note, when everyone order parts, try to order extra and as low tolerance as possible (1% etc) and test them to make sure they actually are as close to spec as possible.

 

I can easily detect a 0.5ohm out of 10 ohm difference in my tone control circuit as a 3% channel imbalance towards my right ear. After some tweaking to reduce it to approximately 0.1-0.2 ohm difference, I can't detect it anymore. As such, anything more than 1-2% might show up pretty easily.

post #1912 of 2032
Quote:
Originally Posted by tomtomtom View Post
 

does anyone have a recommendation regarding mounting of tubes to BAs?

I have had good luck both using UV acrylic gel, and also with using superglue. Superglue is dangerous though because you can easily ruin the BA driver if you use to much or get it in the wrong place. UV gel works great but you have you hold it together under the UV light, thus exposing your hands to UV which is not exactly good for you. Skip hot-glue entirely, it doesn't stick strongly enough to the smooth metal surfaced of the BA to be useful. 

post #1913 of 2032

The UV thing worked nicely. Super glue is either too runny or takes too long to stick if it isn't runny. :P

post #1914 of 2032
I've been wanting to try and make a custom Bluetooth IEM. I had some time this weekend and here are the results. I took apart a Bluetooth stereo iem style headset I got from eBay. I had to make the shells bigger to fit the electronics. One side holds the battery and the other the electronics. I'm actually surprised at how good they sound.

Edited by wjp007 - 7/7/14 at 6:35am
post #1915 of 2032
Quote:
Originally Posted by wjp007 View Post

I've been wanting to try and make a custom Bluetooth IEM. I had some time this weekend and here are the results. I took apart a Bluetooth stereo iem style headset I got from eBay. I had to make the shells bigger to fit the electronics. One side holds the battery and the other the electronics. I'm actually surprised at how good they sound.

great idea, did you re-use the drivers that came with the original IEM or did you implement your own design? I have been thinking this sort of thing should be possible but I hadn't thought of cannibalizing an existing Bluetooth device.

post #1916 of 2032
Quote:
Originally Posted by CMOS1138 View Post
 

I have had good luck both using UV acrylic gel, and also with using superglue. Superglue is dangerous though because you can easily ruin the BA driver if you use to much or get it in the wrong place. UV gel works great but you have you hold it together under the UV light, thus exposing your hands to UV which is not exactly good for you. Skip hot-glue entirely, it doesn't stick strongly enough to the smooth metal surfaced of the BA to be useful. 

 

Thanks! I ordered some UV curing adhesive and will see if it works well!

post #1917 of 2032
No, I used a GQ Knowles driver. The original dynamic driver was awful. Sensitivity was too low and the driver had a port in the front which resulted in ambient noise leaking in. Also I needed the space inside the shells for the electronics.
Quote:
Originally Posted by CMOS1138 View Post

great idea, did you re-use the drivers that came with the original IEM or did you implement your own design? I have been thinking this sort of thing should be possible but I hadn't thought of cannibalizing an existing Bluetooth device.
post #1918 of 2032
Quote:
Originally Posted by tomtomtom View Post
 

 

Thanks! I ordered some UV curing adhesive and will see if it works well!

 

I used the same UV gel for attaching the tubing that I used to make the shells

post #1919 of 2032
Quote:
Originally Posted by wjp007 View Post

No, I used a GQ Knowles driver. The original dynamic driver was awful. Sensitivity was too low and the driver had a port in the front which resulted in ambient noise leaking in. Also I needed the space inside the shells for the electronics.

 

I have a set that I made with the GQ driver as well, with no tuning it was extremely bright and harsh, with white dampers it was no longer harsh but it became flat and boring and I was almost ready to give up on it but I was finally able to get my hands on some more green dampers (they had been on back order at mouser) and now they sound amazingly sweet and balanced. The base does not have that strong resonance that I get from my GK set but the mids are noticeably better, I guess the BK driver part is just a much better mid driver than the FK driver part that handles mids in the GK. 

 

How did you tune yours?

post #1920 of 2032
I used the white dampers. I thought the GQ was a WBFK paired with an ED. I do agree the bass is lacking. I'm thinking of adding a CI for my next design as a three way.
New Posts  All Forums:Forum Nav:
  Return Home