Re: [CAD_CAM_EDM_DRO] Re: Parallel port relay blues

On Sunday 30 January 2005 03:22 pm, caudlet wrote:

> > This doesn't address the issue of the port being expected to source
> > current, though, rather than sinking it...
>
> Most likely the 'source' for current is a pullup to plus 5 of something as
> high as 4.7K. If that is the case, the max current would be about 2.5Ma even
> if he connected the base direct. Given that the Beta of the transistor is
> something like 50, that means that it would drive a realy coil of about
> 125ma or less. The problem with this circuit is that it defaults to on, so
> when you fire up your PC during bootup, there may be circumstances where the
> ouput of the pins is undefined. Once the port initilizes during the Windows
> load it may well let the pin go high. Nothing like having all your 'stuff'
> come on during bootup!

Powerup issues are something that I haven't addressed yet in my thoughts on
where I want to go with this stuff, but it's surely something that I'm going
to have to deal with. My thinking at this point is that I'll have some sort
of a global enable signal that'll keep anything from happening right off,
until it's supposed to. If there's a spare pin available maybe I'll use the
charge pump approach that I've seen referred to.

> Doing active low for triggering the relay is a good approach although
> you may well be beset with some of the same bootup issues. One of
> the simplist ways to deal with the transistion is to use a chip like
> the ULN2003 series that has the drive to handle several relays.
> There are other open collector chips as well. I grabbed a 74541 non-
> inverting octal buffer and it worked to make the signal transitions
> from a PP to optos nice and clean.
>
> PNP transistors are easy to use. Ground the collector and connect
> the emitter to the Coil -. Put a diode across the coil with the
> cathode to + and the anode to -.

You put the coil in the emitter circuit and the base will have to go that much
further voltage-wise. I'd probably put it in the collector circuit.

> Put a 1K resistor in the base to the parallel port pin and use active low to
>turn things on. The port will sink enough current to turn the pnp hard on.
> The 2N2907 is the complemtary transistor to the 2N2222. Any PNP that will
> handle the max current of the relay will work. You can also run loads that
> need higher voltage. 12V relays are more common.

Yep, though there are some 5v ones out there. I even have a couple that are
in a TO-5 can, with driver transistor and diode! Some sort of "Teledyne"
markings on them, if I'm remembering right. And these are DPDT! Smallest
ones I've seen. I also seem to have *way* too many 24V and even some 48V
relays...

> If the relay needs to drive a big load you may find that it's coil exceeds
> the current capability of the TO92 style plastic transistors. YOur choice
> is to use a small relay to turn on the big one or go to bigger transistors.

Yep. Lots of choices there, and when I build something of the sort I'm sure
I'll find much use for the stuff I already have on hand...