Headphone ext cable help

[MorningDew]

Let me preface my question with a disclaimer that this is my first attempt at building any sort of cable, soldering, etc. So, thank you in advance for your help and patience.

I’m attempting to build a simple headphone extension cable with Canare star quad and Neutrix /4in on either end. I tried to follow the below image of the cable and it’s instructions which called for the following connections:

R-, L-, and the cable shield to the ground/TRS shield
R+ to the ring
L+ to the tip

Image with yellow labeling is of my attempt.

I’ve confirmed with a multimeter that this is the outcome of what I built. However, upon its first use I’m getting terrible STATIC. Sounds hollow, unusable. If someone could help me correct my work I would be very grateful. My apologies if my attempt/question is not well described. Thanks!

 

[PaulD]

You do not need star-quad cable for a headphone extension cable.

A star-quad cable is used to maximise the CMRR of the input preamp with longer microphone runs, by ensuring any interference is equally distributed on the conductors. When wired, the blue wires are connected together at each end, and the white wires are connected together at each end. This can improve noise rejection in balanced cables, sometimes useful for low-level signals such as microphones.

A standard TRS headphone cable operates at higher level signals, unbalanced, and only needs three conductors, the shield and R+ and L+. Wiring is simply tip-to-tip, ring-to-ring, and shield-to-shield.

If you want to use star-quad cable (because maybe you already have it), keep the blue wires together at each end (maybe connected to the tip on each end), white wires are connected together (maybe to the ring on each end) and the braided shield is connected to the shield on each end.

Hope that helps!

 

[MC_RME]

I disagree - this is an unbalanced headphone extension cable. To minimize crosstalk the ground in it must be as low resistance as possible. Using the StarQuad as explained in the pics and in the OP's post is the absolute right way to go.

@MorningDrew: Hollow sound means out of phase, so you must have overlooked something. Did you plug a TRS into the female connector to measure the whole path 1:1? From your photos the most likely error might be that you used the wrong pins at the female connector of the cable.

 

[solderdude]

When you hear the hollow sound is it a bit lower in volume as well and is stereo info a bit more mono?
In this case, most likely, the sleeve is nog making proper contact with either the headphone plug of the amplifier socket.
Wiring seems correct.
I assume you checked tip is connected to tip, ring to ring, sleeve to sleeve and there is no short between ring, tip and sleeve ?
The latter is easy to happen in your cable usage.

 

[MorningDew]

Success! Thank you all for your help. So after trying all of your recommendations and triple checking the connections I found that it was correct. Confirmed with a headphone cable attached and testing the L/R Sennheiser pins then down at extension male end that everything was correct. So, with then extension cable soldered but not assembled with the caps I plugged it in and got clear and full audio. I then began reassembling each plug while still listening and it turns out the screw-on section at the male end is causing the issue (apparently in the left channel) just before it’s all the way tight. My guess is that the strain relief area as it gets pinched must be causing some unwanted contact. As long as that screw-on section is backed off a couple of clicks from super tight it works 100%. Thanks again for your time and expertise!

 

[PaulD]

I disagree - this is an unbalanced headphone extension cable. To minimize crosstalk the ground in it must be as low resistance as possible. Using the StarQuad as explained in the pics and in the OP's post is the absolute right way to go.

You can disagree, but do you think it will make an audible difference?
Let's look at the facts. I will attach the spec's for the cable

So the 2 cases are:
Shield in parallel with 2 conductors, versus shield alone (conductors in parallel for signal). Let's just worry about the return (ground) resistance only and not concern ourselve with any potential improvement in the signal transfer resistance..
Let’s be simple and suggest a 3.33m headphone extension cord:
So we have 0.01 ohm || with 0.03 ohm (that's both conductors in parallel) = 0.008 ohms return current resistance
versus
0.01 ohm for the 3.33m extension cable.

Do you really think there is much difference between 0.01 ohms and 0.008 ohms for the ground return of a 3.33m headphone extension cable? It could well be rounding error! I would need a reel of the cable to check it with measurements, and I am not at work at the moment... (Yes, I studied enough science to know that I should not be comparing 0.008 with 0.01, they are the same when we take significant digits into account, that is why I would measure it if I twas at work... maybe post COVID).

Yes I will grant you that 0.008 ohms is slightly better than 0.01 (it's about 1dB in power transfer terms if we ignore the difference in significant digits). But in the practice of a headphone extension cable will it make any difference?

Does it outweigh the simplicity of wiring a star-quad cable the way it is meant to be wired?

 

[AnalogSteph]

Let’s be simple and suggest a 3.33m headphone extension cord:
So we have 0.01 ohm || with 0.03 ohm (that's both conductors in parallel) = 0.008 ohms return current resistance
versus
0.01 ohm for the 3.33m extension cable.

Given the attached datasheet specs, I have no idea how you arrived at these values.

For the L-4E5C, if you parallel two conductors of 130 mOhm/m each and a shield of 25 mOhm/m, that gets us approx. 18 mOhm/m, or 60 mOhms (0.06 ohms) for a 3.33 m cable. With contact resistance at one end included, maybe 80-90 mOhms total?

I will agree that with the shield doing most of the heavy lifting, it makes little difference whether one or two conductors are used in parallel. It's about 20 mOhm/m with just one.

 

[solderdude]

Another question is which Sennheiser is used (the impedance specifically).

0.43Ω per 3.3m wire.
0.083Ω for the shield (3.3m)

So either common return wire (only screen) would be 0.083Ω or (with 2 wires in parallel) = 0.06Ω = 28% reduction in wire resistance.
Indeed let's leave out solder and contact resistance.

Even when the headphone in question were 32Ω crosstalk differences, in this case, are -52dB (no parallel wires) or -54.5dB (with wires in parallel)

Of course these numbers will get better when the HP in question is 300Ω -71.1dB (no parallel wires) or -74dB (with wires in parallel)

 

[MorningDew]

So since this is a discussion of impedance, are there considerations I should make in either materials or layout when planning DIY cables for 300 ohm HD6xx or 37 ohm (and current-hungry planar) Sundara? I’m hoping to use more of the same Canare L-4E6S star quad for both. If so, when splitting the cable at the Y what should be done with the outer braided shield since it was mentioned that it’s doing the “heavy lifting”? At the male TRS end, solder it with the two negatives to the sleeve (like was done with the extension cable)? Thanks again everyone.