Fader Connection, stupid question

I am just wiring up a bunch of faders, but for some reason they have more wires than I would expect.

They are P&G 3000 Series, exact type is 3220.

I have the wiring diagramm, according to the manual, attached.

Red feeds the signal into the fader
Yellow is the wiper "output"
Green goes to 0V
Black goes to chassis ground

But what the hell am i doing with the blue wire?

Any ideas?

Blue goes together with green to 0V.

To properly optimize fader performance you want to use what is called a "Kelvin" connection...  One ground (I dislike that term) wire is carrying current so will experience a voltage drop due to Ohms law, another purely reference wire is not carrying current, so accurately reflects the voltage at the bottom of the fader. This will affect "fader kill", something that console designers obsess over, when fader feeds a proper differential circuit. Often it is trivial to add the differential to the post fader gain stage.

JR 

[edit... not a stupid question.... not asking would be stupid.. [/edit]

what is the Blue wire attached to? to the metal case of the fader?

Whoops said:

what is the Blue wire attached to? to the metal case of the fader?

Click to expand...

Should be two wires attached to the same bottom of the fader resistive element...  So one wire can carry clean voltage, not corrupted by audio current flowing through the fader and the other wire.

To deliver 90+dB fader kill is not trivial.  It requires an accurate differential circuit.

The case will generally grab chassis ground from the CHASSIS it is screwed to.

JR

It is to do with where the signal current flows and how to ensure you get the best fader off level as JR says. Think of the signal going down from an amplifier to the top of the fader and through it.  Where do we want the return current to go?

Then think of the smaller signal current tapped off the fader slider and fed to an  amp input; where do we want its return current to go?

If you use a single wire from the bottom of the fader then both currents flow in one wire. If you connect this wire to the driving amp, the 0V reference of the input amp is a long way from the input amp 0V reference (the length of the wire plus the distance between the two 0Vs). Any current flowing in this circuitous route creates a voltage drop so the fader is never truly off.

It is a little better if you connect the wire to the 0V of the input amp. The wanted signal current now finds it 0V at the input amp so the off signal is better but the driving amp return current stills flows through the wire and causes a voltage drop so the off is not as good as it could be.

With separate wires, one can be connected back to the driving amp and carry its return current. The other is connected to the input amp and carries only its return current.  Inside the fader, one wire is usually connected to the very bottom of the fader - this is the return for the driving amp. The fader construction often has a short length of bare metal rather than resistive track at the bottom and the  input amp return wire is usually connected to the metal a couple of mm above the the other one.

This the best of my recollection of how we did it 40 years ago at Neve so it may not be wholly accurate but I think the gist is right. In most Neve consoles, faders returned signal to the originating module so the driving and input amps 0Vs were already very close to each other.

Big question is what colour wire is which? If you look at the P&G diagrams you will see one wire is connected a little way up from the bottom. That is the input amp return.

Cheeers

Ian

Whoops said:

what is the Blue wire attached to? to the metal case of the fader?

Click to expand...

Source amp is connected between red and blue wire, output signal is taken from yellow referenced to green IIRC.
Usually green and blue are connected to the same point which is 0V.

I say tie them together.

BBC maybe needed 90dB kill when multiple programs for different services came through the same console. Football and children choir, for example. "YOU" will generally be mixing all one song, same key and beat, so a 40dB kill is infinite-enough. You'll probably get over 50dB even on a bad fader.

PRR said:

I say tie them together.

BBC maybe needed 90dB kill when multiple programs for different services came through the same console. Football and children choir, for example. "YOU" will generally be mixing all one song, same key and beat, so a 40dB kill is infinite-enough. You'll probably get over 50dB even on a bad fader.

Click to expand...

I was surprised how audible a 40dB kill is. With 10K pots its not hard to find ones with just a few ohms or so off resistance and you can achieve 50 or 60dB kill. But when I was experimenting with a balanced passive summer with faders and pan pots, I had to use 1K log pots. I had lots of problems with kill which I initially put down to the 'Neve trick' I was using to make balanced pan pots with just a dual gang pot. Eventually I traced it to the off resistance of the 1K pots. I think I was getting something like 46dB down best case and it was quite audible.

Cheers

Ian

This topic turned out way more interesting than I anticipated.

Now I understand why I saw on some consoles the use of a microswitch and a relais for total cutoff when the fader was down. I can't remember exactly which manufacturer it was, but i think it was either Siemens or Acousta broadcast console. 
The same switch would also do fader start and another one would allow for PFL when the fader was pulled back a bit more... quite a nifty design in the analogue world.

Upon studying the P&G manual, I see that this additional wire is only present on the faders that offer  95dB or 100dB cutoff;  the models with 80dB or 85dB don't feature it.

If I end up with 80dB cutoff I think I'll be more than happy.  I have the habit of muting unuesd channels anyway.

PRR is right that fader kill (and isolation) matters more for broadcast consoles where different programs may be playing at the same time in the same console, but it is a benchmark measurement that separates the men from the boys in console design. It does not cost much to do it right, it is fairly common to make the post fader gain stage differential, so for one extra wire and a couple cent resistor, you get full fader performance.

If it wasn't useful the fader makers wouldn't break out the extra path, but even cheap faders benefit from adding the extra wire, and pc mount faders can make the Kelvin connection using extra pcb traces.. 

JR

PS: L/R separation in old phono cartridges was in the 20s, and not flat with frequency... channel separation and fader kill is likewise worse at HF because simple wire resistance is not always flat..

JohnRoberts said:

PS: L/R separation in old phono cartridges was in the 20s, and not flat with frequency... channel separation and fader kill is likewise worse at HF because simple wire resistance is not always flat..

Click to expand...

And I remember being amazed when stereo cassettes first came out at how much better the stereo was with nearly 40dB of channel separation.

Cheers

Ian

ruffrecords said:

And I remember being amazed when stereo cassettes first came out at how much better the stereo was with nearly 40dB of channel separation.

Cheers

Ian

Click to expand...

Yes, when dealing with recording technology it is always important to see the big picture and not go crazy optimizing stuff that doesn't make a difference (like my several high performance  phono preamp designs, that were way better than the medium  : ).

I find it remarkable to hear the acclaim for vinyl these days, when older (digital) media is empirically better.

JR

JohnRoberts said:

PS: L/R separation in old phono cartridges was in the 20s, and not flat with frequency... channel separation and fader kill is likewise worse at HF because simple wire resistance is not always flat..

Click to expand...

For a halfway decent cartridge It's about 20dB around 20Hz and 15k.Hz  It's more like 40dB  around 1k.