[newzild]

Hi guys,
 
I have a question about custom IEMs, specifically this: if you buy a second-hand set of custom IEMs and have them remolded to fit your own ear canals, does this affect the IEM's sound signature?
 
The reason I ask is that I've been considering buying a JH13 or JH16 Freqphase second-hand and having it remolded for my ears through Null Audio (Singapore) or Rooth (Japan).
 
The key property of Freqphase is that it's been designed so that all frequencies reach the ear at the same time, and presumably this is done by placing the drivers within the shell so that they are equidistant from the eardrum.
 
Presumably, when a third-party manufacturer remolds the IEMs it uses its own shell, and the drivers will be set in new positions, thereby disrupting the "Freqphase" property.
 
Any ideas if this is true?
 
Thanks
SM

 

[bacon333]

I'm in the exact same scenario. Were you able to get a response on this?

 

[Ivabign]

I have been told that many of the higher quality re-shell companies make sure that the length and diameter of tubing are kept the same in a re-shell. Just like a JH13fp can be made for any ear size or shape by JH Audio, it shouldn't make too much difference if the placement is a bit different inside the shell as long as the distances are maintained. If you look at a pre-fp JH16, you will see the high frequency BA sits almost at your ear, right up at the end of the nozzle - in the fp version, the high frequency BA is set much further back in the shell.
 
It seems like it would be a simple swap of innards, but until we have a measurement from a before and after pair, we will never really know. I do know that mine sound awesome and every review and impression I've read of the JH16fp applies to my set (for the most part)
 

 

[goodvibes]

tube length matters but probably not crazy critical since there's more to the freqphase than that. Even if slightly off, it should only have minimal effect when you consider wavelengths vs crossover frequency and how little cancellation or blurring would occur. A mm or 2 may be negligible in the grand scheme. Obviously, you want it as good as possible but even in the original, we don't know that they are all the exact same length. Only that their lengths are arranged for pressure arrival to match at common frequencies. I'm sure that reshellers try very hard to maintain those relationships.
 
It's funny. This came up in another thread where a poster though equal tube length is all there was to freqphase. You don't correct 90* of phase shift at 2kHZ by moving a driver 1/4". You'd need to move it 7 times that.  6.75"/4 = 1.7", basically the length of the 1/4 wave (90*) of the full wavelength of 2kHZ.(360*). 45* at 1.5Khz needs over an inch.

 

[newzild]

.

 

[newzild]

  I'm in the exact same scenario. Were you able to get a response on this?

No, I haven't been able to get a definitive response on this. It would be good to hear from one of the major reshellers as to whether or not they think they can preserve the FreqPhase properties.

 

[newzild]

  tube length matters but probably not crazy critical since there's more to the freqphase than that. Even if slightly off, it should only have minimal effect when you consider wavelengths vs crossover frequency and how little cancellation or blurring would occur. A mm or 2 may be negligible in the grand scheme. Obviously, you want it as good as possible but even in the original, we don't know that they are all the exact same length. Only that their lengths are arranged for pressure arrival to match at common frequencies. I'm sure that reshellers try very hard to maintain those relationships.
 
It's funny. This came up in another thread where a poster though equal tube length is all there was to freqphase. You don't correct 90* of phase shift at 2kHZ by moving a driver 1/4". You'd need to move it 7 times that.  6.75"/4 = 1.7", basically the length of the 1/4 wave (90*) of the full wavelength of 2kHZ.(360*). 45* at 1.5Khz needs over an inch.

"tube length matters but probably not crazy critical since there's more to the freqphase than that."
 
That's interesting. Can you be more specific about what other properties are associated with FreqPhase other than optimal placement of drivers?

 

[goodvibes]

Crossovers are designed to work in conjunction with that tube length. Tube lengths are dissimilar so it's more complex than equal tube length for all the drivers to match phase and group delay.