Home-Made IEMs
Oct 2, 2020 at 11:28 PM Post #11,673 of 15,974
Here is the schematic for the Heavy 2 version.....I kept the 22 Ohm resistor originally on the ED and used the 2.2 Uf Cap as described to get the high pass. Let me know if I've missed anything, particularly on the polarity....I reversed the ED.heavy 2.png
Plus no need of adding 22ohms parallel to ED
Just make it CEnter tap 2.2uF
 
Oct 3, 2020 at 8:40 AM Post #11,674 of 15,974
For those interested in a 10 drivers with a "large" U shape sound with subbass 10-100hz; flat mids 120hz - 2khz and sparkle highs (great for singers in live situations)

Dual Sonion 38AM
R1=100 OHM

Tubing 25mm in total
10mm large tube for both spouts (3mm OD/2.5mm ID) (yellow & orange damper facing driver 3mm from spout)
9mm skinny tube 2.5 mm OD
10mm steel tube 1.5 mm OD
___________________________________
DTEC 3008 (polarity inversed)
Bandpass Filter
C1=22uF
R2=68ohm

C2=2.2uF
R3=24ohm

Tubing 35mm in total
7mm skinny tube silicone
7mm large tube (yellow damper)
17mm skinny tube silicone
10mm steel tube
___________________________________

Dual SWFK 31736

C4 = 1.5uF

Note* : You can add a R4 = 10ohm (connected in series for further attenuation)

70mm Tubing in total
10mm large tube 3mm OD/2.5mm ID silicone for both spouts
54mm skinny tube 2.5mm OD / 1.5mm ID silicone
10mm steel tube TOTAL
(length 6mm -1.5 mm OD with
Length 6mm 2mm OD/1.6mm ID hornet

All tube inserts 2mm.
Custom 10 Driver.jpg
10 drivers scheme.jpg
 
Last edited:
Oct 3, 2020 at 12:34 PM Post #11,676 of 15,974
Good eye! I missed the CT the first time. I assume the 2.2 cap is direct to CT since that is constant and I will reverse polarity as needed to correct phase? Here is corrected:

heavy 2 corrected.png
Reverse 33AJ too
I mean you have to check the phase in retrospect to HODVTEC

The HODVTEC damper is double yellow or 4700ohms foam damper by knowles
 
Oct 3, 2020 at 1:03 PM Post #11,677 of 15,974
Reverse 33AJ too
I mean you have to check the phase in retrospect to HODVTEC

The HODVTEC damper is double yellow or 4700ohms foam damper by knowles
I saw the yellow in the description of the first iteraton ( posted a few back...you possibly didn't see it) but wasn't sure if it was a typo. I assumed the tube for the Dtec is a 1 mm because you wanted it as long as possible. If so, would I put the damper(s) right at the dtec spout?
 
Oct 3, 2020 at 1:40 PM Post #11,678 of 15,974
For those interested in a 10 drivers with a "large" U shape sound with subbass 10-100hz; flat mids 120hz - 2khz and sparkle highs (great for singers in live situations)

Dual Sonion 38AM
R1=100 OHM

Tubing 25mm in total
10mm large tube for both spouts (3mm OD/2.5mm ID) (yellow & orange damper facing driver 3mm from spout)
9mm skinny tube 2.5 mm OD
10mm steel tube 1.5 mm OD
___________________________________
DTEC 3008 (polarity inversed)
Bandpass Filter
C1=22uF
R2=68ohm

C2=2.2uF
R3=24ohm

Tubing 35mm in total
7mm skinny tube silicone
7mm large tube (yellow damper)
17mm skinny tube silicone
10mm steel tube
___________________________________

Dual SWFK 31736

C4 = 1.5uF

Note* : You can add a R4 = 10ohm (connected in series for further attenuation)

70mm Tubing in total
10mm large tube 3mm OD/2.5mm ID silicone for both spouts
54mm skinny tube 2.5mm OD / 1.5mm ID silicone
10mm steel tube TOTAL
(length 6mm -1.5 mm OD with
Length 6mm 2mm OD/1.6mm ID hornet

All tube inserts 2mm.

10 drivers custom.jpg
It appears the negative for the Sonions are passing through the positive leg of the Dtec..am I missing something? Would love to see the graph of this build if you have it as well as a gut shot...interested to see if it can fit in my ear!
 
Oct 3, 2020 at 2:21 PM Post #11,679 of 15,974
It appears the negative for the Sonions are passing through the positive leg of the Dtec..am I missing something? Would love to see the graph of this build if you have it as well as a gut shot...interested to see if it can fit in my ear!
My bad on that...it was the first time that i used an app to make the schematics. Will edit it asap
 
Last edited:
Oct 4, 2020 at 9:39 AM Post #11,680 of 15,974
That is a matter i wanted to discuss as well. To make a truly flat IEM for you you hve to adjust for your own hearing. I asked my sister which is an audiometrist, and she replied that you likely have a hardness of hearing in the mids and highs. Which acutally makes me wonder why most builds sound bright and harsh for you.

Edit:
I replied too fast! She said that this "small, little" hearing hardness can cause exactly what you described. Apparently humans have two borders/limits in hearing. One is the level at which we can hear a sound which in someone with perfect ears would be something around 0dB (which we can see in your diagramm and most frequencys) the other level is the discomfort level at which the sound becomes uncomfortable to listen to. This level doesn´t shift! So in the highs and mid you basically have a small window of enjoyable soundlevel and thats why many builds sound harsh to you.
I've been thinking about this...the test I showed was ability to discern a particular frequency once it reaches a particular DB. You are correct, the TOP limit wasn't checked...is that what a "tinnitus" test does and if not does your sister know of a test that does that?
 
Oct 4, 2020 at 10:59 AM Post #11,681 of 15,974
Oct 5, 2020 at 5:23 AM Post #11,682 of 15,974
Hello everyone, I would like to ask a few question regarding low pass and high pass filter. How do one make low pass and high pass filter? I do find calculator online but I am not too sure what value to put into the calculator. Lets say I want a 22955 low pass at 500 hz, and 29689 high pass at 500 hz as well, how do I calculate them?

link to 29689 specification https://www.knowles.com/docs/default-source/model-downloads/ed-29689-000.pdf
link to 22955 specfication https://www.knowles.com/docs/default-source/model-downloads/ci-22955-000.pdf

1601889768513.png

29689
1601889790688.png

22955


Thank you very much!
 
Oct 5, 2020 at 9:37 AM Post #11,683 of 15,974
Two ways.

One is experimentation with just a single cap on the ED and a combination of a resistor(really doesn't act as a LP filter though) and dampers on the CI.

The other, which is slightly more complicated, involves building an RC filter for each pass. The good news is that the values of the components are the same. You start by figuring out how much attenuation (if any) is needed on your (hopefully) low freq driver, and build that into the RC filter.
 
Oct 5, 2020 at 9:45 AM Post #11,684 of 15,974
Two ways.

One is experimentation with just a single cap on the ED and a combination of a resistor(really doesn't act as a LP filter though) and dampers on the CI.

The other, which is slightly more complicated, involves building an RC filter for each pass. The good news is that the values of the components are the same. You start by figuring out how much attenuation (if any) is needed on your (hopefully) low freq driver, and build that into the RC filter.
Hello! Thank you for the reply! What are the exact values that I should be looking for and taking from the specification to calculate the RC filter? Also by attenuation do u mean by adding resistor in series to reduce the spl of the lower frequency driver?
Thank you!
 
Oct 5, 2020 at 10:35 AM Post #11,685 of 15,974
Hello! Thank you for the reply! What are the exact values that I should be looking for and taking from the specification to calculate the RC filter? Also by attenuation do u mean by adding resistor in series to reduce the spl of the lower frequency driver?
Thank you!
I use the first calculator located below for crossovers. I plug in values until the crossover point is found....whether it is LP or HP depends on the configuration. For 500HZ I came up with 17.7Ohms and an 18UF cap. Then you may have to adjust if there is not a readily available value, or use resistors/cap in parallel or series to get the desired point. Regarding Zobels and padding, I use the second one, though there is lots of discussion as to which value for the impedance to use, as BA's impudence changes in relation to the frequency being applied. I usually put in an average of the datasheet values and guesstimate as to the DB drop I'm looking for. I take it you have a test rig, as it is still a lot of trial and error?

http://www.sengpielaudio.com/calculator-RCpad.htm

http://www.sengpielaudio.com/calculator-Lpad.htm
 

Users who are viewing this thread

Back
Top