Need Help!- Mill stops for no reason

[bucky898]

I have two IH CNC mills. Both will stop for no reason while running a Gcode program. The mill just stops as if you had pressed the E-stop button. Mach3's red "Reset" button is flashing. No error messages are shown. It's not a problem in the programs because sometimes they run to completion with no error. I searched the web and found that perhaps I should turn off "Use Watchdogs" in the general config. I did this but no improvement. Any ideas?

[bucky898]

Followup to my previous post - there is a status message and it says "Limit switch triggered". So I suppose there may be a limit switch malfunctioning. Nothing is actually touching any of the switches.

[beege]

Some machines won't move if the way lube is low.

[ger21]

You're getting noise on the switch lines, triggering Mach3. One workaround is to increase the switch debounce in the general config screen until the problem goes away. Try starting at 2000-3000.

[MichaelHenry]

Had the same problem with a different brand mill and the solution was to install one of those ferrite donut thingys on the parallel port cable.

Mike

[bucky898]

Originally Posted by ger21 You're getting noise on the switch lines, triggering Mach3. One workaround is to increase the switch debounce in the general config screen until the problem goes away. Try starting at 2000-3000.

Thanks - I'll give this a try.

[Cruiser]

There is always a reason ! I have fought this same thing for long time and finally won. There are several ways the noise can get generated. To solve this I will assume that you have a Bboard & aux 5v/10v supply.
take a wire from the aux power supply, negitive to chassis, one for 5v & one for other. this may well heal your noisy wound. There are also similar points on Bboards, spindle speed boards etc which will kill noise. All you are doing is connecting all your grounds to chassis. Only do one at a time for start, I did every point I could find on each system and now it is quiet as baby. I tried ferrites and I tried debounce settings, The star grounding is what it took to quiet mine.
Just for clarity ! This is an OUTPUT neg ! NOT INPUT !

[bucky898]

Thanks. I will get out my oscilloscope and start checking for noise on the lines.

[bucky898]

I installed an opto-isolated breakout board and this fixed the problem. The mills were supplied with a non-isolated board, model C13 from CNC4PC.COM. This board has no active components at all, just traces leading to screw terminals.

I had called Tommy at IH and asked for his advice. He said to disconnect the grounded shield wires that were around the cables going to the limit switches. This didn't help. I then looked at the +5 and +12 power lines with a scope and they were extremely noisy. IH had wired everything with only one ground: servo inputs, limit switches, servo power supply, electronic spindle control. I concluded the only thing that would help would be an isolated breakout board.

I suppose IH supplied the non-isolated board as a cost-cutting measure but I think it was a very poor decision - I don't see how any Mach3 mill could run reliably without one.

[Cruiser]

QUOTE
"I suppose IH supplied the non-isolated board as a cost-cutting measure but I think it was a very poor decision - I don't see how any Mach3 mill could run reliably without one."

Think about it a little differant ! The opto isolator will isolate ! It will not cure the noise or better the stray influence of voltages where they don't belong. The opto is a good thing, but It does not get rid of the potential for causing problems. The best way of dealing with it is to ground out the negitive potential so it can not influence anything. I may not be explaining it correctly, and I'm sure someone with more knowlege will chime in. I didn't realize how significant the grounds were till I started setting them to the chassis. The cost cutting board you mention is just a board used for making of connections to the parrallel ports. It is not any more than what is neaded. The grounds are not any part of the circuitry proper. They are only a way of grounding the stray potential which is percieved and described as noise. It is the true fix !

[Runner4404spd]

i agree with cruiser, the optoisolated board really shouldn't make any difference in your problem. you either need more grounding or more shielding or some filter caps. i had to build a filter cap circuit to bring the limit switches into the control box because they carried alot of noise.

[SWC]

I have been having the same problem. Tommy told me to disconnect the ground and that did not help. I am not sure as to what to do next. I have been trying to get ahold of tommy for 2 weeks now. I am not familiar with the above possible fixes. Can anybody elaborate for me in laymans terms. The machine also leaks oil and grease like crazy from the z axis. Tommy says do not worry bout I am getting tired of ruining parts and clothes. Any suggestions? the machine only has 10-15 hours of run time on it. Thanks in advance