Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: SimonD on November 24, 2007, 06:48:26 PM
-
Guys,
All was working well on my 3 axis mill until I decided to try and be clever and add a 4th axis.
I have it all working again but for some reason I seem to be picking up noise on my limit switch signals.
I only have limit switches on the X,Y & Z axis.
I have connected them to +5v via a resistor as shown in the MACH3 manual.
They are using shielded cables.
I have a capacitor across each input.
I have had to raise the debounce level to 2500 in the config page to even run a program and it still keeps tripping the limits every now and again.
I noticed that others have reported problems with the flexible keyboards and will try a different keyboard but would like to find a way to measure the noise if I can.
I don't have a scope to test the wires with, is there any other way I can track down/measure the problem?
Thanks
Simon
-
Noise is hard to track down. Is it always the same switch? Do you have low quality switches? Is everything grounded well?
-
I think it is always the same switch but cannot be sure. I am using reasonable quality microswitches. I think everything is grounded well. I have grounded the limit cable drain wires & motor cable drain wires at the controller end only. They show a dead short with a multimeter.
-
I think I have found the problem...
Whilst "improving" the electronics I decided to switch to a metal enclosure. In doing so the parallel port connector outside is now grounded to the case of the enclosure. This seems to be causing the problem! I guess all grounds are not the same or I have inadvertently introduced a ground loop!
-
Keep us posted. I always watch the posts about noise. It always ends up being grounding, bad connections or faulty switches.
-
I cut out a bigger hole in the case and mounted a plastic plate on the case with the LPT connector connected onto the plastic which keeps it completely isolated from the case ground and have been able to change the debouce level back to zero from 3000.
It has been running for several hours now without tripping the limit switches once!
I wish I understood electronics a bit better! I am not sure I understand the ground problems properly!
e.g.
I have a homemade relay board which is powered by the 24v power supply and takes 2 signals from the controller to change the relay to power the spindle and coolant. Now, Do I connect the other half of the coil to the -ve terminal of the power supply or the ground of the power supply or the ground of the controller or the signal ground of the controller?
I currently have it connected to the -ve terminal of the power supply and it works although the LED glows dimly all the time.
If you follow the electricity flow it looks right from the power supplies point of view but wrong from the signal point of view as the signal goes from the pin on the LPT port to -ve on the power supply. (If you see what i mean)
I have attached a pic of the relay board, perhaps it is wrong?
-
+24 from your power supply to the 24v of the board. The negative of the supply goes to ground on the board. The LPT goes to signal.
-
Thanks for the reply... That is how I currently have it wired up.
But presumably there is some kind of problem though as the LED's glow dimly all the time and then go bright when switched?
-
Try to pull the signal terminal low to shut of the transistor and see if the LED goes out. Or hook it to the LPT and see how if it works that way. The circuit looks correct so it should work.
-
Try to disconect the zenar diode. The one between the ground and signal. I wonder if that is causing a voltage leak keeping the LED on. The zenar doesn't need to be there.
-
Thanks I'll have a go at those suggestions tomight and report back...
The reason I put the zener in was as an extra safety measure as described on this page...
http://www.epanorama.net/circuits/parallel_output.html#relaycontrol