Home-Made IEMs
Aug 11, 2019 at 5:58 PM Post #9,781 of 15,974
If you read it properly and see FR, it was normal tuning. He mentioned itself.

But still a hard experiment to play of.


What do you need from your sound.

BS6Z is more than enough, if you believe me.


MASM7 adds midrange, and prevent treble roll off to give it the airy sound. Bass is also prevented from bass roll off but midrange is attention here a slight more, so bass seems more flatter compared to @eunice BS6Z. I love BS6Z in itself.

Well, I was going to build the BS6+ zobel but then I had some extra RAB 32257 so I thought; what the heck - MASM7?
I really like the MASM3, but it could do with a touch more in the lower end.
I also made som Y-tubing I was quite pleased with (2+1Id) which will fit the BS6 nicely.

Does this seem right?

Skjermbilde 2019-08-11 kl. 23.39.55.png
 
Aug 11, 2019 at 8:01 PM Post #9,782 of 15,974
Well, I was going to build the BS6+ zobel but then I had some extra RAB 32257 so I thought; what the heck - MASM7?
I really like the MASM3, but it could do with a touch more in the lower end.
I also made som Y-tubing I was quite pleased with (2+1Id) which will fit the BS6 nicely.

Does this seem right?

Yes it is right

Just put green damper first and white damper afterwards

And the moment you add resonator to MASM7, it becomes different beast as soundstage profile expands
 
Last edited:
Aug 12, 2019 at 2:45 AM Post #9,783 of 15,974
So I finally received BS6 drivers and assembled them with the suggestions by @eunice. However I noticed a bit of weird thing. Zobeled BS6 hisses quite a lot. It is not super sensitive and my DAP output is silent, so there must be something with the setup or my drivers.

My idea is maybe zobel values are a bit off. One time I remember experimenting with different resistors on another zobel and when I used quite lower resistor value I noticed the same hissing. Increased the resistor - hissing disappeared. Did anyone notice something similar? Technical data for this driver is missing so I wonder if zobel value is correctly calculated..

Oh and I went with brown damper on high frequency drivers and orange on the woofer. Tried with white on HF and red on LF, did not like it too much.

BS6 sounds really good, nice bass punch, a lot of details, sound is very layered and feels like a one unity and not different drivers playing. Really like this one :)
 
Aug 12, 2019 at 3:47 AM Post #9,784 of 15,974
So I finally received BS6 drivers and assembled them with the suggestions by @eunice. However I noticed a bit of weird thing. Zobeled BS6 hisses quite a lot. It is not super sensitive and my DAP output is silent, so there must be something with the setup or my drivers.

My idea is maybe zobel values are a bit off. One time I remember experimenting with different resistors on another zobel and when I used quite lower resistor value I noticed the same hissing. Increased the resistor - hissing disappeared. Did anyone notice something similar? Technical data for this driver is missing so I wonder if zobel value is correctly calculated..

Oh and I went with brown damper on high frequency drivers and orange on the woofer. Tried with white on HF and red on LF, did not like it too much.

BS6 sounds really good, nice bass punch, a lot of details, sound is very layered and feels like a one unity and not different drivers playing. Really like this one :)
Well the situation is slightly same on some tracks.

The value if zobel is still spot on.

If you lower resistor, you increase damping factor so hissing should be silent. Increasing resistor reduces damping.

The hissing is due to TWFK nature. Cannot help it, BS6Z demands a more cleaner source for some reason.

The major improvement on BS6Z is 1kHz distortion and crazy hissing, and by huge margin.


This song shows the hissing on BS6Z.

BS6 normal(no zobel) turns that hissing into grain, making it sound slightly rough.

BS6Z turns the grainish texture(rough vocals) into smooth sound with silent background but also have residue of hissing here and there which comes out as distracting.

But I love it how complex music is layered properly. Black background music has this small issue but its okay



You can increase TWFK damper with one colour value
 
Last edited:
Aug 12, 2019 at 4:07 AM Post #9,785 of 15,974
Well the situation is slightly same on some tracks.

The value if zobel is still spot on.

If you lower resistor, you increase damping factor so hissing should be silent. Increasing resistor reduces damping.

The hissing is due to TWFK nature. Cannot help it, BS6Z demands a more cleaner source for some reason.

The major improvement on BS6Z is 1kHz distortion and crazy hissing, and by huge margin.


This song shows the hissing on BS6Z.

BS6 normal(no zobel) turns that hissing into grain, making it sound slightly rough.

BS6Z turns the grainish texture(rough vocals) into smooth sound with silent background but also have residue of hissing here and there which comes out as distracting.

But I love it how complex music is layered properly. Black background music has this small issue but its okay



You can increase TWFK damper with one colour value


With 30 ohm resistor hissing is less obvious, only in silent passages as you say. But with 25 is was quite obvious and annoying to tell the truth. I do not hear any difference in sound signature when going from 25 to 30, it may be more visible with measurements. Unfortunately I do not have measurement equipment.
 
Aug 12, 2019 at 4:22 AM Post #9,787 of 15,974
I don’t have any hissing on my BS6 + Z (using dragonfly black or Audio-gd nfb11 or fiio Bluetooth cable)

May I ask which caps you did use?
Hmmm, strange. It was hissing both on my phone and ibasso DAP.
I used Vishay MELF resistor and Taiyo Yuden ceramic capacitor. I used those on the other setup and they work just fine.
 
Aug 12, 2019 at 4:32 AM Post #9,788 of 15,974
Hmmm, strange. It was hissing both on my phone and ibasso DAP.
I used Vishay MELF resistor and Taiyo Yuden ceramic capacitor. I used those on the other setup and they work just fine.
I use hung lo cheap chinese ceramics. I measure them to find similar pairs as they have very bad tolerances.

But I do think hissing must be an issue of your source and high sensitivity of the dual TWFK highs. Maybe try a different source?

If you were to buy a dragonfly black for under $100 you would not regret it.

Edit: other threads on head-Fi do report hissing with ibasso dap and sensitive IEMs. So I think it’s your source.
 
Last edited:
Aug 12, 2019 at 5:00 AM Post #9,789 of 15,974
With 30 ohm resistor hissing is less obvious, only in silent passages as you say. But with 25 is was quite obvious and annoying to tell the truth. I do not hear any difference in sound signature when going from 25 to 30, it may be more visible with measurements. Unfortunately I do not have measurement equipment.
I don’t have any hissing on my BS6 + Z (using dragonfly black or Audio-gd nfb11 or fiio Bluetooth cable)

May I ask which caps you did use?
Hmmm, strange. It was hissing both on my phone and ibasso DAP.
I used Vishay MELF resistor and Taiyo Yuden ceramic capacitor. I used those on the other setup and they work just fine.
I use hung lo cheap chinese ceramics. I measure them to find similar pairs as they have very bad tolerances.

But I do think hissing must be an issue of your source and high sensitivity of the dual TWFK highs. Maybe try a different source?

If you were to buy a dragonfly black for under $100 you would not regret it.

Edit: other threads on head-Fi do report hissing with ibasso dap and sensitive IEMs. So I think it’s your source.

It is the issue of two TWFK.
That just loves to sound more hiss-y due to drop low impedance structure on TWFK side if you see the impedance graph. The impedance on TWFK is very low and TWFK likes to hiss and run sensitive. Even Knowles counterpart(if any...i dont think so) will hiss too.

Just use as clean as a source you can or damp the TWFK side a little more
 
Last edited:
Aug 12, 2019 at 5:37 AM Post #9,790 of 15,974
I use hung lo cheap chinese ceramics. I measure them to find similar pairs as they have very bad tolerances.

But I do think hissing must be an issue of your source and high sensitivity of the dual TWFK highs. Maybe try a different source?

If you were to buy a dragonfly black for under $100 you would not regret it.

Edit: other threads on head-Fi do report hissing with ibasso dap and sensitive IEMs. So I think it’s your source.
I used to have Zorloo zuperdac as a dac for my laptop, but after several upgrades I remained with DX150. Really good dap. And it was hissing a bit with super sensitive IEMs like Andromeda, but not with any I've manufactured. Will have to test it with AMP8, it is supposed to have blackest background.
 
Aug 12, 2019 at 9:36 PM Post #9,791 of 15,974
I know what is EST problem..

Hahahaha

Downside of first order crossover is
Distortion(and a crazy one)

a first-order filter doesn't attenuate as abruptly as high-order filters, a driver used with a first-order crossover has to exhibit good performance two octaves beyond the specified crossover point--

For example
in a two-way speaker with a 2.3-kHz crossover point, a tweeter must be able to handle signals as low as 575 Hz, and a woofer must be able to handle signals as high as 9.2 kHz. This is especially hard on a tweeter, producing distortion at high volume and sometimes leading to driver failure.
It may also excite the woofer's "breakup modes," or distortion-producing high-frequency resonances.

So, and EST doesnt have two octave beyond 10kHz, so it produce a distortion producing high frequency resonance and weird chuffing.

Crossing at 12kHz seems where EST starts producing good result, 14kHz is sweetspot

So, EST is not be E25 or SWFK replacement. It is a super tweeter for new gen High Res audio.
 
Aug 13, 2019 at 2:01 AM Post #9,792 of 15,974
I know what is EST problem..

Hahahaha

Downside of first order crossover is
Distortion(and a crazy one)

a first-order filter doesn't attenuate as abruptly as high-order filters, a driver used with a first-order crossover has to exhibit good performance two octaves beyond the specified crossover point--

For example
in a two-way speaker with a 2.3-kHz crossover point, a tweeter must be able to handle signals as low as 575 Hz, and a woofer must be able to handle signals as high as 9.2 kHz. This is especially hard on a tweeter, producing distortion at high volume and sometimes leading to driver failure.
It may also excite the woofer's "breakup modes," or distortion-producing high-frequency resonances.

So, and EST doesnt have two octave beyond 10kHz, so it produce a distortion producing high frequency resonance and weird chuffing.

Crossing at 12kHz seems where EST starts producing good result, 14kHz is sweetspot

So, EST is not be E25 or SWFK replacement. It is a super tweeter for new gen High Res audio.

But there are some IEMs that have treble extension up to 20kHz (twfk and swfk can do this, though it's not easy to keep extension and reduce nasty peaks of lower treble, 4kHz to 9kHz lets say). Dynamic drivers are also very good at extending up to 20kHz, but they are mostly not as precise/fast compared to mentioned BAs.

Given these reasons combined with new problems that EST introduces, I woulnd't call it the next gen just yet. Perhaps a viable alternative with possible totl implementation that we have yet to see, one that isn't polarizing to people first and then all else. I do have to admit that Sonion's documentation about EST is very different compared to most BAs which is exciting in its own, perhaps it will be a shift towards actually usable documentation as an industry standard. Looking at you Knowles...
 
Aug 13, 2019 at 2:20 AM Post #9,793 of 15,974
I know what is EST problem..

Hahahaha

Downside of first order crossover is
Distortion(and a crazy one)

a first-order filter doesn't attenuate as abruptly as high-order filters, a driver used with a first-order crossover has to exhibit good performance two octaves beyond the specified crossover point--

For example
in a two-way speaker with a 2.3-kHz crossover point, a tweeter must be able to handle signals as low as 575 Hz, and a woofer must be able to handle signals as high as 9.2 kHz. This is especially hard on a tweeter, producing distortion at high volume and sometimes leading to driver failure.
It may also excite the woofer's "breakup modes," or distortion-producing high-frequency resonances.

So, and EST doesnt have two octave beyond 10kHz, so it produce a distortion producing high frequency resonance and weird chuffing.

Crossing at 12kHz seems where EST starts producing good result, 14kHz is sweetspot

So, EST is not be E25 or SWFK replacement. It is a super tweeter for new gen High Res audio.

That would certainly explain why all the EST ones I heard sounded so weird. I guess you gotta use it together with an SWFK rather than replacing one :\
 
Aug 13, 2019 at 6:41 AM Post #9,794 of 15,974
But there are some IEMs that have treble extension up to 20kHz (twfk and swfk can do this, though it's not easy to keep extension and reduce nasty peaks of lower treble, 4kHz to 9kHz lets say). Dynamic drivers are also very good at extending up to 20kHz, but they are mostly not as precise/fast compared to mentioned BAs.

Given these reasons combined with new problems that EST introduces, I woulnd't call it the next gen just yet. Perhaps a viable alternative with possible totl implementation that we have yet to see, one that isn't polarizing to people first and then all else. I do have to admit that Sonion's documentation about EST is very different compared to most BAs which is exciting in its own, perhaps it will be a shift towards actually usable documentation as an industry standard. Looking at you Knowles...
When you cross a tweeter, you see the frequency going at 0Hz side(I mean the left side of FR chart) not toward the upper side. Same with the woofer.

I meant octave on the low frequency side and EST cannot do lower frequency at all. Distortion rise become so significant that it shuts lower frequency it self(I mean upper mids in simple terms).
 
Aug 13, 2019 at 12:17 PM Post #9,795 of 15,974
EST only use till now in my setup is to prevent roll off in treble region


New Build

Woofer
CI-22955 + 10ohms
HODTEC-31230 series + zobel

zobel
Cz=1uF
Rz=203.5 ohms

These two driver shares the tube
14mm length
damper at the end(keep dampers 1mm distance away from end of the tube)
damper is Single orange

Can try yellow and red too




Tweeter and Full range
ED-29689 center tap + 2uF High pass
Update
RAB is removed, as it is unneeded, this setup works actually better even though Impedance is now a little swingy(still tight, but not near straight line)
zobel
Cz=5uF
Rz= ~32.125 ohms

12mm length
both driver share the same tube
Brown damper(same at the end with 1mm distance clearance for safety, damper popping inside ear)


All drivers in parallel


I am still experimenting
 
Last edited:

Users who are viewing this thread

Back
Top