Home-Made IEMs
Feb 10, 2019 at 2:50 AM Post #8,195 of 15,910
How is everyone joining two or more drivers/acoustical to one tube?
I bought a tool to stretch the opening of a tube, that way you can fit two drivers in one 2mm ID tube.
This question has been asked multiple times in this thread with no clear answer. Search for y-tubing in this thread.
I will ask a friend to 3D print a Y adapter for me, he also built an adapter for my 2cc coupler which works very good.
 
Feb 10, 2019 at 3:07 AM Post #8,197 of 15,910
I bought a tool to stretch the opening of a tube, .

I use a soldering iron on low to warm the tubing, then take a chopstick and push it in the tubing until it cools. I repeat the stretching if needed or for three tubes. I use glue to affix the 1.5 tubes (obviously). Then the 1.5 tubing OD fist perfectly inside that larger tubing. So it’s ok but time consuming and I worry about long term use.
 
Feb 10, 2019 at 3:18 AM Post #8,198 of 15,910
I use a soldering iron on low to warm the tubing, then take a chopstick and push it in the tubing until it cools. I repeat the stretching if needed or for three tubes. I use glue to affix the 1.5 tubes (obviously). Then the 1.5 tubing OD fist perfectly inside that larger tubing. So it’s ok but time consuming and I worry about long term use.
I use heat gun at slow speed

Use tweezers to expand the tube slowly while rotating it.
 
Feb 10, 2019 at 5:22 AM Post #8,199 of 15,910
My flagship
IMG_20190210_152531.jpg

Will start working on this soon

Collecting all the design I can come with
Already simulated on comsol

This is my top of line project

But now I have to create custom shell and PCB with SMD components.(I mean huge shell with PCB here and there)


SMD inductor is also available with the value I have given in there.

OK what did I tried to do

HODVTEC

acting as a true state of art subwoofer(inspired by @piotrus-g scary crossover's CI). The zobel circuit in this driver kills all HF present. The L-pad eats up voltage and drops driver by 10dB, so impedance is now mostly flat with no HF or any sort of artifact peak coming from 0dB in graph.

Now the 72Hz crossover acts as a true low pass.

RAB-32063 and SWFK

I used the driver due to better impedance compared to vented counterpart. This driver also has lower distortion. Since HODVTEC was handling Bass, this RAB in series parallel arrangement with zobel will act as a perfect midranger.

SWFK is one of my favourite driver due to its figures. Harshness can be controlled by any measures but technicality it throws is just sheer enjoyment to work on with.

SWFK is crossed at 11kHz by single capacitor.

Now, the circuit down there is impedance voltage eater. I saw that on speaker builds

This is actually better than series notch as it doesn't require the factors and doesn't even need Le.

This is an impedance voltage divider.

This will fix the resonant peak of RAB and SWFK(the tiny one) with SWFK rising impedance at the same time, plus correcting speaker phase issues.

11SMD components
8driver
2Bass + 4mids + 2high

The mid ranger will require @Ivan TT reason for mid RAB.
No dampers for this

The problems is length of tubing for now


Plus if somebody wants to remove HODVTEC from equation

Just get RAB 32257 and its zobel
4.6uF and 41.25ohm

And remove the peak notcher from below
 
Last edited:
Feb 10, 2019 at 9:52 AM Post #8,201 of 15,910
@Ivan TT @IvanNOON @Rabid86
Ivan tt, actually

I did some simulation about series driver concept of yours. You were right about a magical sound.
I was just keeping one impedance factor in mind and not inductance of driver.

Well, I got to this very interesting junction.
Putting driver in series, will not only double the Impedance (z) but also inductance

So when RAB 32257x2 is in series, at 500Hz we have 66ohm impedance and 15.6uH inductance.

This leads impedance graph going flat till 11kHz compared to single RAB.

But that 11kHz will be straight wall compared to steep cliff from 5kHz.

So impedance will be flat till 11kHz which means no phase shift, smoother treble and amazing clarity.

On 4xRAB, things get even more flat. The wall becomes a wall stop at 16kHz. Means no reproduction after 17kHz. The impedance rise nearly becomes full stop(imagine near infinite impedance). Well music details stop at 14kHz, as some people on other thread say, but microphone noises and other factor begin after that. So no air hiss, sound on your setup was complete black(if I am correct??).

The impedance end point will get sudden peak at that point and will drop dead. Rising impedance is what creates variable peaking.

But series driver will make resonant hump more peakier too.

192ohms 3kHz is way to high for amp to handle properly, making the phase dip apparent. It can be a 90degree phase shift. That's why I suggested you series parallel.

OK that was mentioned by JBL acoustics in 1982 about series driver fixing the impedance. Now I get it.

4xRAB in series is not recommended if HF extension is in mind and 2xRAB is good enough for series.

Parallel series arrangement will keep the 2x series impedance structure at 11kHz but the impedance wall becomes a cliff again, due to voltage division per driver. The amp thinks it is a single driver with 33ohm and 7.8mH at 500Hz but actually it is two parallel with 66ohm and 15.6mH at 500Hz.

Paralleling driver also helps in flattening of resonant peak due to same factors.

Full range is so fun to play with. Easier than crossover for understanding.
 
Last edited:
Feb 10, 2019 at 4:41 PM Post #8,203 of 15,910
I use a very soft 1mmID 2mmOD at the driver (length approx 2mm) . Then 2mmID and 3mm OD through the canal.

I have the impression it reduces some resonances if I do it like that.

Regarding glue: what to use then? I also have UHU which is better for gluing wood but worse for gluing the tube to the drivers. UHU also creates the white flume sometimes. I tried SK47 silicone glue which doesn’t create any flumes but also is not sticky enough to attach the tubes.

I will try the method that Bassiklee told us, besides Loctite I believe some people use Epoxi, so that will be my next try if needed.
Weird that SK47 doesn't work properly, and good to know because a local store here recommends this product for gluing tubes from what I've read.

Thanks for the tubes infos by the way!!
 
Feb 11, 2019 at 10:44 AM Post #8,205 of 15,910
@dhruvmeena96 Is it usally true that higher ohm damper put on woofer side makes more thump and lower ohm damper put on tweeter side gives more bright highs?
For GV, a lot of people recommend to use green on tweeter, and red or orange on woofer. (from Here). I am wondering if Brown instead of Green and Orange will have more V-shaped sound. Or is there any reason why people usually put greens on tweeter?

Also I have driver from W3 that I am currently building. I am thinking about putting white damper on TWFKs and red damper on CI, Should it be okay?

Thanks
 

Users who are viewing this thread

Back
Top