Re: Twisted Pairs

Hi,

You're right. The edges are coincident and of opposite direction, so
there is no problem there.

Regards to coupling noise into the power supply. This is not a
problem because supply lines are heavily bypassed with capacitors to
ground. Figuring 30 pF capacitance per foot and 300 uF of bypass
capacitors, a 10' cable would couple 5 mV into the supply on every
logic edge. This is not a problem.

Mariss

--- In CAD_CAM_EDM_DRO@y..., Alan Marconett KM6VV <KM6VV@a...> wrote:
> Mariss and the list,
>
> Good morning!
>
> How about a differential pair? But then I guess that would be the
same
> signal!
>
> Alan KM6VV
>
>
> mariss92705@y... wrote:
> >
> > Hi,
> >
> > I was talking to someone yesterday who was using a 10' twisted
pair,
> > sheilded cable for his encoder. He had used one pair for channels
A
> > and B, the other for +5VDC and ground. He called to say things
> > weren't working very well.
> >
> > I suggested one pair be channel A and +5VDC, the other channel B
and
> > ground. He rewired it and called back to say everything was now
> > running very well.
> >
> > The moral of the story is only one wire of a twisted pair should
> > carry a signal; the other wire should be ground or supply.
Otherwise
> > cross-talk will get you.
> >
> > This is true also if twisted pairs are used for step and
direction,
> > limit switches or any other I/O use. It can be the cause of many
> > mysterious malfunctions.
> >
> > Mariss