Fader connections- cable choice

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

boji

Well-known member
Joined
Jan 6, 2010
Messages
2,375
Location
Maryland, USA
Hello friends,

Need to extend some P&G faders' funky, conductive-epoxied wirepads?!? * about an extra 7" to get  them over to the channel input cards.

Anyone have an opinion on using  2-conductor, (Mogami W2944)  for fader interconnects?
(would use the shield wire for faders'  Agnd)

Or, would you use 3-wire with a proper**  shield?

As always, thanks for your input!
-Boji


**  On the old 79' Wheatstone, the P&G's 3-wires loosely wandered around to channel card backplanes, with no shielding, no cable  management except end connectors. The runs were all  10"-15" lengths.
So... as long as they remain inside the console frame, everything's peachy?


* Have you ever tried to de-solder P&G fader wires right at the thru hole point on the resistive strips...successfully?  Didn't  short-out any traces while scraping at the silvery blob that looks like solder, but acts like epoxy ? :eek:
 
boji said:
Hello friends,

Need to extend some P&G faders' funky, conductive-epoxied wirepads?!? * about an extra 7" to get  them over to the channel input cards.

Anyone have an opinion on using  2-conductor, (Mogami W2944)  for fader interconnects?
(would use the shield wire for faders'  Agnd)

Or, would you use 3-wire with a proper**  shield?

As always, thanks for your input!
-Boji


**  On the old 79' Wheatstone, the P&G's 3-wires loosely wandered around to channel card backplanes, with no shielding, no cable  management except end connectors. The runs were all  10"-15" lengths.
So... as long as they remain inside the console frame, everything's peachy?


* Have you ever tried to de-solder P&G fader wires right at the thru hole point on the resistive strips...successfully?  Didn't  short-out any traces while scraping at the silvery blob that looks like solder, but acts like epoxy ? :eek:
You could easily use 4 conductors or 3 conductors + shield. Doubling up the fader ground leads, allows one to carry the signal current, and the second ground lead provides the fader 0V reference to the following make up gain stage.  Fader kill is an important spec in big dog consoles.

JR
 
1. Soldering directly on to the P&G conductive plastic fader element termination points (silvery blobs).
I have a vague recollection of attempting this, or something similar, some 35 years ago, without success. It was very easy to write-off a fader. As it was already a duff fader and therefore destined for the trash can, it was OK to experiment with. It was an experiement that failed and therefore has not been repeated.
Recommendation: Leave an inch or 2 of original P&G fader wire & extend with screened cable.

2. Mogami 2944 is a 2-core twisted with overall screen.
If you are planning to use this as a combined screened fader send and fader return cable, beware!
High impedance and crosstalk are the enemies.
What is feeding the fader? If it is an amplifier, the fader send will be at low-impedance. Good news :)
The fader return will be at a varying impedance. The fader wiper will be low-impedance at the bottom of the fader travel (ground) and low-impedance at the top (max level & assuming the fader is fed by an amplifier), but along its travel impedance it will be varying.
High-impedance wiring  is susceptible to receiving crosstalk, and the fader send in your twisted core Mogami cable will be having a full-on face-to-face conversation with the fader return wire. Bad news :(
The best solution is to use individually screened figure-of-8 cable, one screened core for the fader send, one screened core for the fader return. You can also then apply John Roberts' advice regarding the use of the return wire's screen as the ground reference for the following circuitry. Do not underestimate the importance of fader kill!




 
I can't lay my hands on it right now but I seem to remember a Neve technical note saying for short lengths ( up to about a foot) you can use a twin screened cable but for longer ones you need to use two separate single core screened cable.

Cheers

Ian
 
Fader kill is an important spec in big dog consoles.

The best solution is to use individually screened figure-of-8 cable

longer ones you need to use two separate single core screened cable.
So glad I asked!  Thank you everyone.


Edit: I just got why you italicized "plastic" ::)
 
If you don't mind, one last check before purchase:  Is the figure-of-8 pair doing something unique that two, shielded, single core wires running parallel to each other could not accomplish?

For example: https://www.redco.com/Mogami-W2368.html

All I could find in figure 8 was 23ga, so going with the mogami I think...
 
Is the figure-of-8 pair doing something unique that two, shielded, single core wires running parallel to each other could not accomplish?

For example: https://www.redco.com/Mogami-W2368.html

All I could find in figure 8 was 23ga, so going with the mogami I think...

Figure-of-8 is not essential, but it is convenient.
Here in the UK it has traditionally been an easy to find and very good value for money type of cable, so for this application, choice #1. (A red core and a blue core also help with conductor identification :) )

Two runs of the W2368 provides the same solution as figure-of-8.
I am surprised at the (seemingly) reasonable price for the Mogami. Over here Mogami has always been "rudely expensive".

The important thing in all of this is that the send and return signals are not enclosed in the same screen.

 
boji said:
Thanks for taking time to explain.  Much appreciated.
I have made many with a common screen, the most important thing is to keep the fader current out of the 0V reference line.

Two separate shielded wires is even better, and probably easier to source than 3 conductor shielded.

JR
 
Thanks JR. 

Please forgive yet another inquiry, but your comment on keeping fader current out of reference has me wondering...would it be a bad idea to route the fader's 0v directly to the backplane agnd bus bar, as opposed to what I planned, which was to have it return to the pcb's fader header that uses an agnd pour?
 
For best 'offness' you need to reference it to the AGND of the input fed by the return which as often as not is the board that sent it.

Cheers

Ian
 
boji said:
Thanks JR. 

Please forgive yet another inquiry, but your comment on keeping fader current out of reference has me wondering...would it be a bad idea to route the fader's 0v directly to the backplane agnd bus bar, as opposed to what I planned, which was to have it return to the pcb's fader header that uses an agnd pour?
You need to reread my chapter about using differential amplifier stages to forward and back reference audio grounds. (Oops I never did write "that" book, but I have posted here on the subject at length).

Real wires have resistance, so the current flowing through the fader, will cause a drop in the wire feeding the fader and the ground wire returning the current. The drop in the feed to the top is inconsequential, but the drop in the ground, can superimpose a voltage on the bottom of the fader, diminishing fader kill.

By using two ground wires (only one carrying current), you can compare the voltage at the wiper, to the actual voltage at the bottom of the fader. As long as there is little/no current in these sense wires you can extract clean audio. This is the same theory behind accurate voltmeters (google "Kelvin 4-wire connections").

Ground is a concept not a voltage, and 0V (another concept, can validly exist in numerous places inside a chassis). Judicious use of differential amps allows us to forward and back reference audio grounds.

In the case of the fader, forget about a brute force ground bus (only one s). Connect the wiper to the + input of the post fader gain stage, connect the ground resistor in the NF network to the fader 0V reference, using proper resistor ratios the ground error will subtract out. In most cases any gain stage can be made differential as easily as adding a couple resistors.. If you study schematics of consoles (especially mine) you will see numerous differential stages sending and/or receiving audio stems around internally. 

JR
 
boji said:
Thanks a bunch JR.  If I was closer, I'd come ninja-mow your lawn for ya.  8)
With my big dog  44" zero turn I can cut the grass in 20 minutes or less... then it takes a day or two to do the trim. Mowed yesterday, not looking forward to trim.  :'(

JR
 
Was looking for some bushing replacements and came across some newer 3220's on the cheap so I purchased them.  Only difference is they are 4-wire, not 3-wire like the older ones.  So, since I have a three pin terminal on the PCB that the fader connects to,  I welcome any suggestions to improve this plan:
(btw not sure how to make Agnd 'differential',  given the pour)
XicvPgH.jpg

 
Interesting topic. I've read up on Ian's Ground101. At the risk of beating an old topic to death, using twisted pair with outer shield (like Mogami W2944) to wire up a fader in my summing mixer. The not great build manual suggested the simple way of wiring the shield to the bottom, and the twisted pair for top/wiper. So in this case, both send and return share the same shield which is also used for audio 0v.

So instead using two sets of individual W2944, only one conductor in the first cable connected to the pcb and nothing else on this side. Other side goes to the top of the fader, and the shield and the second, unused, conductor is connected to the shield + unused conductor of the return cable, all meeting at the bottom of the fader. Return is connected back to the pcb together with the shield of this second cable, ensuring that the signal is shielded from it leaves the pcb all the way through the fader where it is connected to the audio 0v of the busses/summing amplifiers. Is this correct, and best practice?

Maybe easier to get some unbalanced, shielded cable in the same configuration. I think the extra unused conductor is confusing me.

Thanks in advance!
 
I would not recommend running send an receive to/from a dafer inside the same screen because of the danger of HF crosstalk.

Cheers

Ian
 
Hey Ian, thanks for your input, much appreciated as always. Time to rewire the darn thing then I guess :cry: But in regards to the second, unused conductor in the shielded twisted pair in these two fader-cables; do I connect it to shield only at the bottom of the fader, where it together with its own shield meets the shield and unused conductor from the return cable? Hence only connected to shield in one place.

best,

Magnus
 
Hey Ian, thanks for your input, much appreciated as always. Time to rewire the darn thing then I guess :cry: But in regards to the second, unused conductor in the shielded twisted pair in these two fader-cables; do I connect it to shield only at the bottom of the fader, where it together with its own shield meets the shield and unused conductor from the return cable? Hence only connected to shield in one place.

best,

Magnus
Good question. I am not sure why anyone is recommending connecting the spare conductor to the screen and connecting it at one end only. If the length of the cable is less than one metre then core to screen capacitance is not an issue so you might as well just connect the two cores together at each end and treat it as single core cable.

Thinking about it, if you did use one core to send and one core to receive, then, if the cable capacitance is no more than 100pF and you use a 10K fader, the worst case is when the pot is set at 5K. The 3dB point of 5K and 100pF is 318KHz so I do not think this will have any noticeable effect in the audio band.

Cheers

Ian
 
Back
Top