[Carlosfandango]

Bilavideo I've only just discovered this thread and have only got up to page 12 or 13. But just wanted to whizz along to the last page to commend you in your efforts and dedication to this task.

Absolutely awseome altruistic thread, you information sharing probably helps to single this site out as such a good source of information for people seriously dedicated to hifi.

I kneel at your feet

 

[Bilavideo]

Quote:

Originally Posted by Carlosfandango /img/forum/go_quote.gif Bilavideo I've only just discovered this thread and have only got up to page 12 or 13. But just wanted to whizz along to the last page to commend you in your efforts and dedication to this task. Absolutely awseome altruistic thread, you information sharing probably helps to single this site out as such a good source of information for people seriously dedicated to hifi. I kneel at your feet

Carlos, you honor me with the kudos. Thanks. As you'll find as you continue with the thread, there were a lot of smart people who jumped in and filled it with terrific research and suggestions. I had a question and some genuine curiosity. I quickly found myself surrounded by friends whose background and expertise turned this thread into something amazing.

 

[Bilavideo]

Quote:

Originally Posted by marozie /img/forum/go_quote.gif I was waiting for a nice sunny weekend to try out the UV lacquer, which ended up being today. It worked like a charm. It cured in a matter of minutes in direct sunlight. Unlike the nailpolish, which has a very tacky surface even after it dries which made it at times a bit of a challenge to get the IEMs in my ears, with this stuff on them the slide right in. If you're willing to pay the premium for this stuff it solves a lot of problems with potential toxicity/allergy to your primary plastic substrate; this stuff is even hypo-allergenic.

I think you've found it. I think I just need to go back and re-read where you found it.

 

[marozie]

Quote:

Originally Posted by Bilavideo /img/forum/go_quote.gif I think you've found it. I think I just need to go back and re-read where you found it.

Warner Tech Care, the DIY IEM builder's superstore. It's made by Dreve. They refer to all of the products for hearing aids/IEMs as otoplastiks. It's expensive though, $30 for 20mL (compare that to nail polish which is about $5 for 15mL).

And I don't think I ever expressed my gratitude to you myself, Bilavideo. Kudos! Without you none of this would have happened!

Now, for my next project I wanted to make an Sonion based IEM. I took apart some Klipsch Custom 2s that I had and was going to use those drivers (manufactured by Sonion) as the midrange and add a woofer and a tweeter to it. Sonion drivers can be purchased from Colsan Micro which is in the UK. My thought was to make a 6 driver (3 dual drivers), 3 way design using the 4400 as tweeter, the Klipsch dual driver as the midrange, and the 2800 (which is a monstrous beast). The 4400 series is the super tiny dual driver that someone way back in this thread had used as a tweeter, which I would have thought would be a good choice as well.

I reached out to them explaining what I was looking for and this was their reply, which is pretty interesting I think:

"Hi Mark

This is what I have been advised:

We would advise 2015 or 2091i as a woofer and a 2389 as a tweeter. The 2091i is more difficult to drive but has bigger bass. The 2389 could be driven "half coil" to enhance high frequencies [I don't know what this means, can anyone explain?]. The receiver for Klipsch custom 2 earphone is not generic available. We advise to use a 17A003 as midrange instead. (better sounding) In general we would advise to use two way designs over 3 way designs. Designing a good sounding 3 way is very difficult and almost impossible without proper measurement equipment. For a 2 way the cross over network could consist of a simple capacitor (1-2 uF, preferably ceramic type and not tantalum) and some acoustic dampers for acoustic low passing of the woofer. A 3-way is more complicated, one of the reasons being balance armatures are highly inductive and difficult to filter electrically. We would think a combination of 2389 and 2015 or 2091i will outperform a combination of 2015/klipsch midrange/2389 by far.

4400 makes not much sense for musician IEM because of low headroom. It has too high impedance for use as tweeter."

I understand what they're saying about the low impedance, and the 2389 is the lowest impedance driver they make, but the frequency response isn't very impressive, especially when compared to the 4400. Also, this business about them not recommending a 3 way design... I don't know, the IEMs I made are 3 way and they sounds pretty damn good. And I didn't think of this when I was making those, but couldn't you use a cap and a high impedance acoustic filter on the midrange to achieve both high and low pass filtration? Thoughts?

 

[SilverCans]

[URL="http://www.head-fi.org/forums/f6/home-made-iems-430688/...but couldn't you use a cap and a high impedance acoustic filter on the midrange to achieve both high and low pass filtration? Thoughts?

You probably could. The thing is with a 3 way design, is that trying to achieve a 3-way using only electrical parts (caps/resistors/inductors though inductors are not really feasible for such a project) is tough to do on the mid-range without measuring the freq & impedance curve of the drivers. The reason being because the mid-range driver needs to have bandpass electrical slope (both a high and low pass cross over). However, as you eluded to above, I'd imaging you could use an acoustic filter to simplify things and give you the high freq roll off but, again, you would kind of be guessing without proper frequency measurements. Personally, I'd say go for it, you have nothing to loose given you are making the shells on your own

 

[Bilavideo]

I realize engineers are trained - and paid - to work things out mathematically, which is economically preferable to trial and error. But given the number of variables at work, there's nothing quite like trial and error to test - and dismiss - a great many preconceptions about the relationship between a proposed design and the unvarnished reality of performance in the real world.

Lots of people will tell you what you can't, or shouldn't, do. Only experience will settle the argument.

 

[marozie]

Quote:

Originally Posted by Bilavideo /img/forum/go_quote.gif I realize engineers are trained - and paid - to work things out mathematically, which is economically preferable to trial and error. But given the number of variables at work, there's nothing quite like trial and error to test - and dismiss - a great many preconceptions about the relationship between a proposed design and the unvarnished reality of performance in the real world. Lots of people will tell you what you can't, or shouldn't, do. Only experience will settle the argument.

Dude. You're like a Head-Fi sage. I think that should be the next level after Headphoneus Supremus.

(I have a grammatical bone to pick with that title, while we're on it. If headphone keeps its 'e' in Headphoneus why doesn't supreme? Bilavideo, maybe you can get that changed since you're in the club...)

 

[marozie]

Alright, I got an explanation about driving the tweeter half coil. I mean, I still don't really understand what "half coil" means, but this is how you do it:

They sent me some other information that might be proprietary to Sonion so I don't want to post it, but with the two drivers they recommend, in this configuration, the tweeter goes well past 10kHz, to almost 12kHz, and the woofer is almost 115dB all the way down to 20Hz. That's pretty damn inpressive.

Also, they recommend green and yellow dampers for tweeter and woofer, respectively, whereas I had used white and red.

 

[marozie]

OK, I got pricing info. The Sonion drivers are a lot less than the Knowles drivers, to my surprise. Since I have to have them shipped from the UK I thought I would ask anyone if they're interested in a small group by to save on shipping. Also, I know this is a DIY forum and all, but I'd be more than willing to make IEMs for others if they're intimidated about the process, or just don't feel like doing it. I'll post a new thread about this as well.

 

[marozie]

I apologize for momentarily hijacking this thread but I have something to get off my chest. I had a run in with the Head-Fi police yesterday. As I stated in my previous post, I had been thinking for a while that it would be fun to make IEMs for other people on Head-Fi, not even for the purpose of trying to make money, but because I like building things, and I only need so many pairs of IEMs for myself. I've amassed a lot of knowledge regarding this process, where to source supplies, etc., why not use it? So I put up a post in the "for sale" section and got like 10 replies within a couple hourse before I got a PM explaining that I was breaking the rules and that my post had been deleted. I was told that if I wanted to procede I would have to be classified as a "Member of the Trade" which would require me to pay Head-Fi advertising fees (which would mean that I would have to charge people more than cost to cover those fees), and would severely restrict the kinds of things I could post -- for example, everything that I've posted in this thread would be deleted because MOT can't post anything about their own products in the DIY section, they can't make any comments about their own products, they can't make any comments about anyone else's products, etc., etc., etc. I understand why they want to be cautious about letting people make and sell things, and in fairness any website that is used as a portal for you to sell things to other people charges a fee for that service, but I wasn't trying to start a business, and the primary difference between this and selling something on eBay or Etsy or something is that those sites don't restrict your attempt to turn a profit. I mean, it makes perfect sense that if you're going to use a website to try and make money that the website should get a piece of it. But I was going to sell these more or less at cost, which as far as I see it is providing a benefit to the Head-Fi community since comparable custom IEMs from actual businesses cost 3, 4, 5 times what they cost to make.

In addition, you can't post a thread about a group buy without first getting it approved by the administrators??? What is this crap? I can't see what possible reason they would have for restricting a group buy post. Protecting their members from getting scammed? I don't think that's Head-Fi's responsibility. The whole point of having a community is that people get to know one another. If I organized a group buy and scammed people they would talk about it, I would be blacklisted, no one would interact with me anymore. As some point you have to let the community function as a community, and any community has empirically determined do-s and don't-s, a collectively, if also unspoken, agreed upon standard of behavior. Call it ethics, morality, whatever. But they don't have to be shoved down people's throats by the governing body. I don't mean to go off on too much of a tangent, but this is exactly why communism always fails; communism attemps to force a synthetic ethical ideology on everyone with out letting it evolve naturally.

Am I wrong in thinking that this is a bit of an unreasonable, heavy-handed overreaction, or am I just not getting it? Because this seems kind of absurd to me. I'm a big kid, I can take care of myself, I don't need Head-Fi to protect me. I would like to think the same is true of others...

 

[cyberspyder]

No, but I don't make the rules here...life isn't fair, but there's nothing you or I can do about it.

 

[FYJX]

Quote:

Originally Posted by marozie /img/forum/go_quote.gif Warner Tech Care, the DIY IEM builder's superstore. It's made by Dreve. They refer to all of the products for hearing aids/IEMs as otoplastiks. It's expensive though, $30 for 20mL (compare that to nail polish which is about $5 for 15mL). And I don't think I ever expressed my gratitude to you myself, Bilavideo. Kudos! Without you none of this would have happened! Now, for my next project I wanted to make an Sonion based IEM. I took apart some Klipsch Custom 2s that I had and was going to use those drivers (manufactured by Sonion) as the midrange and add a woofer and a tweeter to it. Sonion drivers can be purchased from Colsan Micro which is in the UK. My thought was to make a 6 driver (3 dual drivers), 3 way design using the 4400 as tweeter, the Klipsch dual driver as the midrange, and the 2800 (which is a monstrous beast). The 4400 series is the super tiny dual driver that someone way back in this thread had used as a tweeter, which I would have thought would be a good choice as well. I reached out to them explaining what I was looking for and this was their reply, which is pretty interesting I think: "Hi Mark This is what I have been advised: We would advise 2015 or 2091i as a woofer and a 2389 as a tweeter. The 2091i is more difficult to drive but has bigger bass. The 2389 could be driven "half coil" to enhance high frequencies [I don't know what this means, can anyone explain?]. The receiver for Klipsch custom 2 earphone is not generic available. We advise to use a 17A003 as midrange instead. (better sounding) In general we would advise to use two way designs over 3 way designs. Designing a good sounding 3 way is very difficult and almost impossible without proper measurement equipment. For a 2 way the cross over network could consist of a simple capacitor (1-2 uF, preferably ceramic type and not tantalum) and some acoustic dampers for acoustic low passing of the woofer. A 3-way is more complicated, one of the reasons being balance armatures are highly inductive and difficult to filter electrically. We would think a combination of 2389 and 2015 or 2091i will outperform a combination of 2015/klipsch midrange/2389 by far. 4400 makes not much sense for musician IEM because of low headroom. It has too high impedance for use as tweeter." I understand what they're saying about the low impedance, and the 2389 is the lowest impedance driver they make, but the frequency response isn't very impressive, especially when compared to the 4400. Also, this business about them not recommending a 3 way design... I don't know, the IEMs I made are 3 way and they sounds pretty damn good. And I didn't think of this when I was making those, but couldn't you use a cap and a high impedance acoustic filter on the midrange to achieve both high and low pass filtration? Thoughts?

That is why i never agreed to people saying "impedance is too high" or "it's too bright" or "to harsh" to be something something something. The TWFK is a good example. No one at Knowles would agree that the TWFK is a good tweeter, but that totally wouldn't be true as Unique Melody has managed to crank out a huge amount of performance out of them in highs and high mids. Also, i would also agree that a 3 and 4 way is much harder to make compared to a 2-way, but if you get it right. It's hard to compare. The 2389 is a very good high freq driver, but it just doesn't extend to those frequencies we all want to go to...yet people seem to ignore this when suggesting the driver.

 

[pdupiano]

Sorry mate, but those are the rule of the forum. I was surprised too by the group buy issue, and I would assume that it is on behalf of protecting head-fiers from past issues. Additionally you really shouldn't look at head-fi as preventing you from doing what you want, more like head-fi would end up getting reprimanded by its sponsors because you would be getting free advertisement for what those guys pay quite a bit of money for (to sell items on head-fi and post information about their products).

As for Half-coil, based on the schematic, it looks like its running a dual driver into a single driver configuration. Think of it as a variable resistor with a full coil and only using half of the whole range. Presumably this would reduce interaction issues with all 3 drivers, limiting yourself to two drivers but obtaining the benefit of using a dual driver's better specs. It is possible that when making dual drivers, a company's tolerance is tighter on each coil making it better to use those dual drivers (at half coil) rather than using a single driver with the same design.

As for the 3 driver design and lack of measuring equipment, and use of experience instead, I'd have to go with Sonion, Knowles, and I would have to say everyone else in this case over Bill's encouraging Trial and error approach. Trial and error can only serve you well if you have the ability to make a dozen of these with all of the different variables. At the moment even if you have say 5 drivers you want to test (assume dual driver), 3 filters (acoustic not caps), 2 different tube lengths, you would have to build 60 iems to test all possibilities, even more with dual drivers then with caps and resistors. Even without the different tube lengths thats 30 iems and that's before the cross over issues. I think finding some measuring device to test frequencies would do you far more good than trial and error. Trial and error is fine if you're working with something feasable (usually cheap, quick, and dirty) but it is rarely the case for electronics equipment. Additionally trial and error would lead you to make conclusions that your ears may not detect because you aren't looking for it. Its like those "where's waldo" books, if you're not looking for red and white stripes and goofy pair of glasses, you'll never find waldo -you wouldn't even think he's in the picture. You can argue that you hear what you hear and would therefore not need any sonic equipment. All I can say is try listening to the same track using the same gear, focusing on different aspects of the music (highs, bass, etc... or even a particular instrument) and you'll find that its possible to listen to something different each and every time -while using the same gear!

I'd still push for the measuring equipment, primarily to have the ability to play with 3 drivers and have measurement data, just seems more fun that way. At the moment if these companies caution people against dual drivers.... I wonder what they say to 8-driver iems.... good luck, and don't wear yourself out making and testing 60 iems, thats what testing equipment is for.

 

[piotrus-g]

Marozie, I'd also advice you using 23xx Sonion series over 4400. 2323 f.e. which goes to 18kHz with ease. I believe that Phonak PFEs are based on 2300 propietary and must say that highs outperforms TWFK by far.
 
If it comes to woofer, you can use dual driver or single. CI22955 is better than 3300 UE-propietary (TF10). However 2015i/2091i should be better than CI.
 
It's harder to make 3way, I agree, but if you made it good you're home. As a midrange driver I'd suggest using 2015i over 1700. Series 1700 is rather entry-level driver. 2000 is said to be more refined and sweeter.

 

[tsn141]

3way systems are very difficult because of high inductance of receivers
if you desire design a 3way system, you must use semi conductors-needs high level of ee engineering -(otherwise you must use very big caps )
i asked that to my electronical engineering student friends and also show w3's crossover circuit theysaid this is very difficult and it needs semiconductor knowledge ,although it seems very simple
secondly there is a more difficult problem , the iem needs a tunning but it has lots of veriable whereas two way no need only 2 circuit variable
also both sonion's and knowles' advise two way more ideal
however i really interest the 3way systems maybe we can design a 3way system
sorry for my bad english