Hi all,

I'm facing a limit switch problem, too:
Firstly the machine works pretty well with NC linit switches at X++/X-- in series and using the X-- as Xhome, too. Same setup for all axis, despite of Zhome is = Z++. There is also an E-Stop button (NC) connected, what directly stops the axis.

The odd behavior is this: Driving an axis manually into a limit switch then the movement immediatly stops (as required). BUT: after a delay of 1...4 seconds it happens that the Spindle/Coolant output relais will be activated and the spindle starts!

The program is in e-stop mode then, I have to press the RESET-button first and then I can manualy move the axis from the limit switch (auto override is enabled) and then I can press the spindle button. This must be pressed twice to stop the spindle, as MACH recognises the respective output to be low (in Diagnostic screen or in run screen by the yellow frame of the spindle button). So first press will switch on (the already running spindle), second press will turn it of.

What I yust tested was this: Terminating the MACH3 program and then press a limit switch by hand. Ooops, this also activated the spindle output, without Mach involved!

Any idea how I can corner this bug? Who/what controls the printer port beside of Mach or when Mach has been terminated?

(ok, I should involve the output relais into the hard-wired E-stop loop, but that's an other issue, I know)

Thanks for support!

Holger
 

Some breakout boards have a charge pump, if you have Mach open it sends a pulse out to the board, when in reset it stops the pulse and the breakout board will disable all I/O.
Hood

I have a similar occurrence. When I home my X axis, the mist coolant relay clicks on then off. Not really a problem, but I just wondered, why?   

you haven't said what your config is for your e-stop and relay signals but if I was to take a stab at it it I'd say it was dodgy sig gnds.
Something I learned recently is that not all parallel port and cable manufacturers properly tie all sig gnd pins to proper sig gnd. Some only bother with those they know will be used by a printer. Check pins 18-25 are all bussed together and if not do so. Otherwise some of the sig gnd pins you've used may actually be floating. Worked for me 

credit where credit's due http://www.cnczone.com/forums/showpost.php?p=356625&postcount=9

As Stirling has said - you haven't said how it all works.  My M3 M4 M7 and M8 outputs from the computer all drive relays (driven by a Darlington array) They require an active low signal to drive them. I have noticed, in certain circumstances my Omron inverter says both M3 and M4 are on simulataneously (this is not allowed and the Omron flashes up a fault). This means that there is no output from the computer, i.e. the computer is at Ov.  The relays and Darlington array are powered from the stepper driver board supply - via a 5v regulator, so if the computer is at 0v, both lines appear active to the Darlington and it turns both M3 and M4 relays on simultaneously. (It is probably turning on the M7 and M8 relays as well, but I haven't connected up my coolant yet )

 The thing to do is check the signals from the computer, first at your breakoutboard board, and see if what is happening is being signalled by the computer, or if it is a mismatch of power supplies to your relay drivers. My breakout board is not powered, so it accurately reflects the output from the printer port. If yours is powered, you may want to disconnect the cable and try checking the port directly at the plug.

Some breakout boards have a disable pin as well, and if you have enble/disabled this from the computer, it may, when on E-stop, disable all outputs from the breakout board. This is fine, but if your relays are active low and powered from another source, this could be your problem.