IPB

Welcome Guest ( Log In | Register )

Help With ATH-m50 recable, Question about cable shield!
tehprairiedog
post Mar 2 2013, 23:49
Post #1





Group: Members
Posts: 6
Joined: 30-March 12
From: Midwest
Member No.: 98200



I have a question regarding recabling my Audio Techinica ATH-m50. Don't worry, I'm not trying to "hear the air between the guitar strings" or anything, heh. The cable has separated from the 3.5mm plug at the end. I saw some pictures (about halfway down http://www.head-fi.org/t/484744/crosstalk-...0#post_6601078) on recabling them and thought it looked like a fun project, but I want to try to do dual entry. I thought of drilling a hole in the right earcup and using Mogami W2534 qaud cable with a DIY y-split and terminating with a 1/4" TRS plug. Does this sound like an okay idea? So my questions are about the shielding. Is it okay that the wires won't have shielding between the headphones and the y-split, and should I connect the shield to the ground on the TRS plug? Some people say connect only one end of the shield, some say both ends, and the guy who started that head-fi thread said to not connect either end of the shield! I'm quite new to DIY, so I'm sorry if this is a stupid question, but I'm confused and haven't been able to find a very good answer through searching on Google. Any help would much appreciated!
Go to the top of the page
+Quote Post
 
Start new topic
Replies (1 - 7)
alexeysp
post Mar 3 2013, 15:49
Post #2





Group: Members
Posts: 130
Joined: 3-April 09
Member No.: 68627



Why not just buy a twin-wire cable with separate shields (e.g. something like this) and use the shields as the return wires? And you can split it as much as you want without introducing additional joints.
Go to the top of the page
+Quote Post
tehprairiedog
post Mar 4 2013, 02:35
Post #3





Group: Members
Posts: 6
Joined: 30-March 12
From: Midwest
Member No.: 98200



Thanks for the reply! That would be simpler, but I want to try it with whole y split and stuff. It sounds like a challenge happy.gif The thing that is tripping me up is this, "Combining the ground wires with the shield allows anything picked up by the shield to flow to the ground, even if the other end of the shield isn't connected to anything." and "Using shielding as ground channel is not recommended, and will surely cause crosstalk." Is this true? If the shield is only connected at one end, there won't be any current through the shield, right? How can this "surely cause crosstalk"? I know this is from a website that I don't completely trust when it comes to objective truth, but I'm curious if there is any truth to these statements, or if it's more audiophile talk. Is there any truth?
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Mar 4 2013, 14:39
Post #4





Group: Members
Posts: 3649
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (tehprairiedog @ Mar 3 2013, 20:35) *
Thanks for the reply! That would be simpler, but I want to try it with whole y split and stuff. It sounds like a challenge happy.gif The thing that is tripping me up is this, "Combining the ground wires with the shield allows anything picked up by the shield to flow to the ground, even if the other end of the shield isn't connected to anything." and "Using shielding as ground channel is not recommended, and will surely cause crosstalk." Is this true? If the shield is only connected at one end, there won't be any current through the shield, right? How can this "surely cause crosstalk"? I know this is from a website that I don't completely trust when it comes to objective truth, but I'm curious if there is any truth to these statements, or if it's more audiophile talk. Is there any truth?



General truths don't apply to all situations. Headphones inherently use a shared ground because that's how all of the headphone jacks are made. As long as the ground conductor(s) have reasonable conductivity a shared ground doesn't cause unmanageable amounts of crosstalk.

A shared ground with relatively high resistance can definitely cause at least measurable amounts of crosstalk. But measurable doesn't necessarily mean audible, not by 3 or more orders of magnitude.

Furthermore, if you know how much crosstalk the ear is blind to, you get a real-world perspective on the problem which is that given reasonable care, the wiring doesn't matter.

IME most headphones that I have taken apart and tried to service used separate grounds down to the plug.

If you are not already an expert equipment repairer/restorer, this whole discussion is probably way over your pay grade.

If this your first soldering project, get another headphone cable and cut the two pieces (old and new) to the desired length(s) and splice them. Wiring 3.5 mm jacks so that they are reliable isn't trivial, so using a robust cable with one molded on is a good solution.
Go to the top of the page
+Quote Post
alexeysp
post Mar 5 2013, 10:51
Post #5





Group: Members
Posts: 130
Joined: 3-April 09
Member No.: 68627



QUOTE (tehprairiedog @ Mar 4 2013, 04:35) *
The thing that is tripping me up is this, "Combining the ground wires with the shield allows anything picked up by the shield to flow to the ground, even if the other end of the shield isn't connected to anything."


If the shield is used as ground wire, its other end is necessarily connected to the load. While this indeed prevents it from properly performing the EMI shielding function, for the audio signal under question this is generally irrelevant.

In situations when it is really necessary to provide the appropriate EMI shielding, two conditions should be met: no noise voltage (relative to the signal reference potential) must be present on the shield, and it must not carry any current. In practice this means that the shield must be connected to the signal reference point (the "signal ground") in the close proximity to the actual signal source.

QUOTE
"Using shielding as ground channel is not recommended, and will surely cause crosstalk."


In most (all?) shielded cables the effective cross-section of the shield strands is at least equal to, and usually larger than that of the actual wires. Besides, as Arnold mentioned, in practice seemingly huge amounts of crosstalk are likely to pass unnoticed by ear.

QUOTE
I know this is from a website that I don't completely trust when it comes to objective truth, but I'm curious if there is any truth to these statements, or if it's more audiophile talk. Is there any truth?


If you want to learn more about the theory and practice of shielding, I recommend to read through this two-part article from Analog Devices:

AN-346: Understanding Interference-Type Noise
AN-347: Shielding and Guarding

The reading will require some effort though if you're not familiar enough with the electrical engineering basics and terminology.

Also note that these papers are written in the perspective of precision measurement instrumentation design. Although the described concepts are universal, their relevance to audio equipment is limited. Of course this doesn't mean that it's fine to abandon any shielding in any audio gear altogether; it only means that you have to think before you do, and keep in mind that the shielding done improperly can potentially cause more harm than good.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Mar 5 2013, 14:34
Post #6





Group: Members
Posts: 3649
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (alexeysp @ Mar 5 2013, 04:51) *
The reading will require some effort though if you're not familiar enough with the electrical engineering basics and terminology.


Just a friendly reminder that the most likely source of cross talk is not related to shielding, but rather relates to resistive drops across the common ground.

For example, if the common ground has the same resistance as the load, then half of each channel's output voltage is dropped across the common resistance, this reflects back to the other channel and the channel separation is actually zero presuming a low source impedance. The opposite channel earphone receives the same signal but inverted, which would give a strong out-of head experience.

Could we do this intentionally and sell it as a feature, not a bug? ;-)

My first estimate of the channel separation if the common lead as 1/10 the resistance of the load would be 26 dB, which is not too good. Should I make a spread sheet?

If the common mode resistance was 1/100 of the load which is probably not all that uncommon, we are in the area of 46 DB which is probably not that bad practically speaking, but also still sufficient to raise significant concerns in many circles.
Go to the top of the page
+Quote Post
tehprairiedog
post Mar 5 2013, 20:19
Post #7





Group: Members
Posts: 6
Joined: 30-March 12
From: Midwest
Member No.: 98200



Thanks Arnold and Alexeysp! This really clears up a lot of things! And thank you for the links about shielding. I'm a freshman in Electrical Engineering, so I'm trying to get my head around all theses concepts and I appreciate all the info I can get smile.gif
Go to the top of the page
+Quote Post
alexeysp
post Mar 6 2013, 13:51
Post #8





Group: Members
Posts: 130
Joined: 3-April 09
Member No.: 68627



QUOTE (Arnold B. Krueger @ Mar 5 2013, 16:34) *
Just a friendly reminder that the most likely source of cross talk is not related to shielding, but rather relates to resistive drops across the common ground.


Yes, crosstalk can be considered a specific case of the "common impedance noise" described in the AN-346 linked above (although technically crosstalk can also be caused by capacitive and inductive coupling between the wires, but for audio signal in a typical audio cable these effects should be negligible).

QUOTE
If the common mode resistance was 1/100 of the load which is probably not all that uncommon, we are in the area of 46 DB which is probably not that bad practically speaking, but also still sufficient to raise significant concerns in many circles.


Resistance of 3 meters of a 25AWG copper wire is about 0.3 Ohm, for 30AWG it's about 1 Ohm. Typical full-size headphones are usually 30 to 100-something Ohms. So yes, even in a cable with relatively thin wires the wire resistance is likely to be about 1/100 of the load resistance, or less.

Providing separate returns for left and right channels and connecting them together only at the plug sleeve will diminish the common impedance even further, virtually eliminating any crosstalk that could be possibly induced by the cable.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 28th July 2014 - 18:30