Re: opto isolator and capacitors

Dave,
Also check out Doug harrison's folder in the files section
folder name is: "Buffered-Parallel-Port-Intf"
in particular file: Wb8schem.pdf

http://groups.yahoo.com/group/CAD_CAM_EDM_DRO/files/Buffered-
Parallel-Port-Intf/Wb8schem.pdf

-Lee

--- In CAD_CAM_EDM_DRO@y..., "turbulatordude" <davemucha@j...> wrote:
> Thank you.
>
> I had not used a opto isolator before and it seemed that the field
> switch (end switches in my case) just turned on an LED. The LED
was
> powered from some external power source. in that manner, only
light
> connected the 'protected' PC circuits and if one were to slip and
> either ground or hit 220 to the LED, all you would loose is the
> optocoupler. A friend bought a new circular saw some many years
ago
> and one if his first project was to saw the cord off. Always
makes
> think what if...
>
> And I always say that one should learn the mistakes of othes, as
you
> will not live long enough to make them all yourself.
>
> That being the case, The LED only needs minimal lumonsoity.
> Enough for the receiver, but not so bright as to allow one to see
it
> from across the shop. ( I like double or triple powering LEDs and
> pulsing the signal to get a remarkable bright display.)
>
> Thanks, Now I need to play with inverting and non-inverting
signals
> from the parallel port to see what will work with what.
>
> Dave
>
>
>
>
>
>
>
> --- In CAD_CAM_EDM_DRO@y..., "Lee Studley" <indigo_red@q...> wrote:
> > Hi Dave,
> > The optocoupler led is just that, an led like you're thinking.
> It'll
> > need a current limiting resister. I'm making the assumption you
mean
> > the optos are for contact closure sensing going back into the
port.
> >
> > I think your friend is trying to work up a debounce for the
switch.
> > reasons for placing a cap across the switch include:
> > Eliminate arcing to contacts for higher voltages, and not really
an
> > issue at the <20Ma.
> > -or- more likely, he probably means it will cause an RC time
> > constant when the switch is opened and will act to keep the PC
from
> > reacting to contact release 'bounce', if the software could
process
> > it fast enough, and stretch the 'on' time enough so that the
> > PC will see the 'contact' if it happens quickly. The cap's
probably
> > not necessary, but try/test it both ways. Too large a cap will
> micro-
> > weld the contacts upon closure during discharge unless you add
an
> > additional small R in series with the cap across the contacts.
The
> > led series resistor will be on the order of:
> > (5v-2.1vledtypical)/0.015mA ~= 200ohms @ 5v. (330 ohms is a
typical
> > value for ~9mA @ 5v)
> > -Lee
> >
> > --- In CAD_CAM_EDM_DRO@y..., "turbulatordude" <davemucha@j...>
> wrote:
> > > A friend said that I should use a small cap at the switch and
a
> > > resistor in the power line (5 volts).
> > > I assumed that they worked just like a LED. figure current
and
> > > voltage then select a resistor.
> > > Is he correct that a cap should be used too ? And if so,
should
> it
> > > be at the switch or at the board ?
> > >
> > > Dave