Re: ParPort output low - hi-Z or ground?

Check out the 7414 Hex inverters, have schmitt triggers in them
cancelling out noise. Port logic can be as high as 12 volts. This
converts it to 5v two ended ttl. Generates a powerful square wave to
clean up signal and allow faster drive operation.

My first wiring job on my bridgeport ran away. I used gen purpose
switching/amp transistors (3904? don't remember now) to flip port
logic from positive to negative.

This chip requires NO other parts, no resistors, no other parts but a
radio shack ic board. solder in the parallel cable or get fancy and
add your own D-sub connector.

That's how mine works anyways.

I added a opto-22 board, positive logic, it works with the software to
turn on and off all the other controls. See pictures posted at the
DIY-cnc site..

Hope this helps.

David






--- In CAD_CAM_EDM_DRO@yahoogroups.com, "caudlet"
<info@t...> wrote:

> --- In CAD_CAM_EDM_DRO@yahoogroups.com, "ghidera2000"
> <ghidera2000@y...> wrote:
> > Spent a day drawing up an optoisolation scheme for the parallel
> > port. I know the inputs go between +5V and 0V but, when you

set an

> > output to low, is it connected to ground or does it go hi-Z?
> >
> > Silly me, I was thinking in terms of sourcing current from the port
> > (OUTput...) and getting fancy with transistors. Just as I finished
> > the drawing (As usual!) I realized that its TTL and TTL is much
> > better at sinking. If it goes to ground then a couple resistors and
> > 1N4148's are all I need to drive the opto-leds.
>
> Is this for input or output? The Gecko's already have opto's on the
> input so they will work with direct current sinks or using a non-
> inverting buffer chip (i.e. 74LS541). Driving opto's is nothing more
> than driving an LED lamp. They are current devices so you have

to

> pass the rated current (10 - 15ma on most). Since, like you stated,
> TTL/NMOS parallel port chips will normally sink more current the

opto

> is normally given +5 from the computer through a current limit
> resistor on the anode and the signal pin is connected to the

cathode

> so that a low turns it on. Most of the motor drive units either have
> TTL inputs referenced to same ground as parallel port (Rutex,
> Xylotex) or have their own opto inputs, so building opto isolation

on

> the step and direction pins is often redundant. If you need to
> trigger larger external loads, a SSR (Solid state relay) has it's own
> opto input and will work directly off of the parallel port using a
> current sinking setup (low = on)
>
> If you need opto inputs to the parallel port then you have a
> different configuration where the opto input is driven from the
> signal (limit switches, table sensors, etc) with an isolated voltage
> source and the opto transistor is used to pull the parallel port line
> low through a collector resistor.
>
> IMO, unless you work REAL cheap you would be ahead to buy

one of the

> available IO breakout boards like the Campbell Designs model

that has

> all of the buffers, opto's and isolated power supplies. Consider

you

> already have a full day of design invested and probably at least 2

or

> 3 more plus that much more to build, troubleshoot, modify, test and
> finalize a design you will have worked for about $2.50/hr (less if
> you have to buy parts). If you are retired or consider your free
> time to be of no value then I guess the tradeoff is justified.
>
> I am an EE with enough experience and spare parts to build any

kind

> of IO interface I want. I bought a Campbell board.