Need Help!- G540 and a Wildhorse Probe?

[Fastest1]

I am running a G540 and I do have the Wildhorse Probe which worked great on my Hobby Cnc Pro board. The function of the probe works fine as long as I have it wired backwards in the G540's input 4 (red to ground, white to pin 13) however the LED does not(direction of current is essential in an LED). If I swap the wires on the G540 the LED will light but no trigger is caused. I have been told this might be due to the optoisolation circuit of the G540. What is everyone else doing? Adding a resistor? If so what value? I want both the LED to light when triggered and for Mach to see the signal. Thanks

[morestuff]

This problem can also occur when using the Wildhorse probe
with most breakout boards, especially those that use current
from the input pin flowing through an opto isolator.

The reason there is a problem is that the current that
lights the LED in the probe is effectively a leakage current
that must flow through the input terminal of the G540 or
breakout board. An external resistor can help divert most
of the leakage current away from the breakout board input.

In the case of a PMDX-125 or PMDX-126, an external pull-up
resistor to +5 volts can solve the problem. We recommend a
470 ohm resistor. Next time we update our design, we will
make a provision on the board to add this resistor internally.

In the case of the Geckodrive G540, the internal supply is
12 volts. There are two problems posed by the G540. First,
the resistor values needed to work with a 12 volt supply
become impractical due to excessive current through the
probe and second, the G540 does not bring its internal 12 volt
supply out to an externally available terminal.

Your solution of wiring the probe backwards so that the LED
does not conduct is probably OK, but be aware the most
LED's specify a reverse breakdown voltage lower than 12 volts.
Opening the probe and clipping the wire between the LED and
the resistor inside will eliminate the leakage current that prevents
the probe from working with a G540.

Regards,
Steve Stallings
www.PMDX.com

[Fastest1]

Would using a 5v power supply for the probe allow me to work around that?

[morestuff]

Unfortunately no. The 12 volt supply inside the G540 is
directly connected to the LED part of the opto-coupler,
so converting the G540 circuit to 5 volt operation would
require major surgery on the G540. It would be better
to create an external 5 volt powered circuit to handle
the "leakage" current and provide a buffered signal to
the G540.

Regards,
Steve Stallings
www.PMDX.com

[john_100]

hi fastest1 ,

as I understand with the Hobby CNC board you had the probe
connected to one of the printerport input pins and a 10K pull up resistor to +5V

but the Gecko G540 input is the cathode of the opto isolators led
the anode being connected to +12V via a 2K2 resistor

the 2.5mA current through the probes led indicator is too high
and you don't see the probe switch open

any idea whats the current limit for the probe switch contacts
and indicator led ?

connecting a +12v supply via a 330 ohm resistor to the probe and gecko input
could be the simplest option , but will pass about 37mA through the probe contacts and about 10 mA through the led

for a lower voltage than 12V,
you need a buffer circuit with an open collector output
a NPN transistor and two resistors will do as in the circuit below
note the signal is inverted and you will need to invert it again in Mach3


John

[morestuff]

The second circuit in your sketch, using a transistor,
should work well. Remember to invert the sense of
the signal in the software configuration.

Remember that the G540 does not present its regulated
12 volts on any external terminal, so surgery on the
G540 or an external supply will be required.

The first circuit, I think, has problems. With about
10 mA of current flowing through the probe's LED
there would also be 10 mA through the 330 ohm
resistor. This will develop 3.3 volts across the 330
ohm resistor. This is enough voltage to make almost
1 mA of current flow through the LED inside the
opto-coupler in the G540. This is just enough to
be a likely cause of unreliable operation as the
opto may not turn off when it should.

You could make the 330 ohm resistor smaller in
value to compensate, but this would cause even
more current to flow through the probe contacts
when it was not triggered. I have not looked at
the 12 volt regulator inside the G540, but it may
get unhappy as the load current gets greater. If
the probe contacts can take the higher current
and you use an external 12 volt supply, it might
work.

Regards,
Steve Stallings
www.PMDX.com

[Fastest1]

It works so flawlessly. Sure wished there was a simple way. I dont need the LED verification and as a matter of fact I would never be able to stop the machine if I had noticed the LED didnt light upon contact. Most likely just clip the LED.