Re: Relay question about diodes across the coil

Thanks for all the input:

To summarize suggestions from these posts and other good practices
to minimize noise (before this gets thrown off for being O.T.):

For D.C. Relays: Diodes across the coil

For A.C. relays: Varistor, Transzorb or a resistor in series with a
capacitor

EMI Radiation on A.C. reduced by using a ferrite core
Noise into A.C. reduced by Corcom type EMI filter

A.C. wires should be routed away from D.C. If not possible, should
cross perpendicularly

Shielding should only be connected to ground at one end (usually at
side closest to earth ground

Grounds should be connected to a single common point

D.C. control should be used wherever possible rather than A.C>
control

Input signal wires should be kept seperate from output signal wires.

Low impedance/resistance paths for current should be used.

Analog signals are prone to interference

Try for high signal to noise ratio wherever possible

Important safety considerations:

Proper Grounding!

Local means of disconnecting all power to tool (disconnect switch or
removeable plug in receptacle

Use of a latched relay type motor starter/contactor with three wire
control to Start/Stop buttons (user on one of the machinist sites
just burned out his Bridgeport spindle motor because of a power
outage. Power came on and his phase converter put 220 volt single
phase to his 3 phase motor - with a contactor, this wouldn't have
happened).

--- In CAD_CAM_EDM_DRO@yahoogroups.com, Jon Elson <elson@p...> wrote:
>
>
> Paul Davis wrote:
>
> >I've had it happen in circuits before. Maybe, though, it was the
back emf spike which caused the circuit to re energize the coil? In
any case it's a good idea to have the diode.
> >
> >Paul Davis
> >
> >
> >On Sun, 11 Jan 2004 16:13:56 -0600, Jon Elson wrote:
> >
> >
> >> Paul Davis wrote:
> >>
> >>
> >>
> >>
> >>> Diodes are a good idea for DC even if you don't care about the
> >>> spike caused by the back emf. The relay will remain latched
> >>> until it dissipates it's magnetic field. Assuming it doesn't
> >>> destroy your driving electronics or short it's coils it can
cause
> >>> the relay to remain latched for a few seconds after the input
is
> >>> removed. So for the sake of speed it's a good idea to have the
> >>> diode.
> >>>
> >>>
> >>>
> >>>
> >> Huh? With the diode, the field dissipates slowly, but it is
> >> usually done in tens of
> >> milliseconds. Without the diode, the field dissipates VERY
> >> rapidly, unless it
> >> arcs over in the coil, and the field collapses in microseconds.
> >> So, the diode slows it down, rather than speeding it up.
> >>
> >>
> Oh, yes. Uncontrolled switching spikes are FAMOUS for causing
totally
> unpredictable malfunctions in digital circuits. I have seen
amazing
> interactions,
> where a relay being turned off in one unit triggered bizarre
malfunctions
> in equipment that seemed quite remote to the circuits that were
connected to
> the relay. And, even experts have sometimes been stymied in
finding the
> cause of the problem.
>
> Jon