Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: Bhohbein on September 27, 2016, 01:36:10 PM

Title: Machine had a another brain fart.
Post by: Bhohbein on September 27, 2016, 01:36:10 PM
Hey guys. My machine had another brain fart yesturday. Its happened a few times in the last couple of weeks. Yeaturday i was running a program that i have run 40 or 50 times no problem. It was a circulair pattern 1.5" down it got about an inch down and took off in another direction.

Any ideas what the problem is and how to fix it

Program was made on fusion 360 cad cam.
Title: Re: Machine had a another brain fart.
Post by: Highspeed1964 on September 27, 2016, 03:08:40 PM
Check your wiring to the stepper drivers and in particular make sure the DIR line has not come loose.

Stephen "Highspeed" Kruse
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 27, 2016, 04:22:31 PM
Ok. Ill try that. 2 more things happened today. Hit auto zero and when the tool touched my zero pad it stoped for a sec pushed down and then retracked. Restarted mach 3 and worked fine.  Then doing a circuil pattern again frist trip around when fine then started to step over  but in an oval pattern. Does that make sence to anyone.
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 27, 2016, 04:57:48 PM
Checked my wires. Wiggled wires while moving all axis all worked fine. Checked plug connections did find a small amount of aluminum shavings in them. Dont think they were shorting out but could have been. Checked all connections at the stepper drivers. All were tight. The mother board mounting screws were loose. Thightened them. Rechecking now.
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 27, 2016, 05:57:40 PM
Ran the same program and worked fine. Ran it again right after and started stepping over again. I have my work zero in the center of my table and work out from there. Hit go to zero and its .150 -y. I didnt let the program keep running but last time it kept stepping over. Ill recheck the wires on y axis.   Any other ideas.
Title: Re: Machine had a another brain fart.
Post by: Highspeed1964 on September 28, 2016, 05:01:20 AM
Could be a motor going out. Or it could be noise interference. Also, what version of Mach3 are you running? I believe the last posted version (albeit some time ago since development on 3 has stopped) has some known issues and it is recommended to use the previous version.

Stephen "Highspeed" Kruse
Title: Re: Machine had a another brain fart.
Post by: Hood on September 28, 2016, 02:49:03 PM
When the machine took off in another direction as mentioned in your initial post. Did the tool path show the motion or did it show what was meant to happen?
Hood
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 28, 2016, 02:53:10 PM
My mach 3 version is r3.043.056. Rechecked the wires on y axis with a meter ohms reading good while wiggleing the wires. How would i check the motor to see if its going bad.  Should i switch to mach 4.
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 28, 2016, 02:54:31 PM
I did look on the screen tool path to see if it showed up. The pic was small but it didnt look like it. Tool path looked normal.
Title: Re: Machine had a another brain fart.
Post by: Hood on September 28, 2016, 03:18:14 PM
Ok, if using the parallel port and the toolpath didn't show the weird motion then it is not Mach that is commanding it and thus it must be an external issue.
Hood
Title: Re: Machine had a another brain fart.
Post by: Highspeed1964 on September 28, 2016, 03:25:06 PM
Good point, Hood.  If the screen is showing the correct tool path (you can check this also on the "Tool Path" screen which has a much larger image to view) then it is most likely a trouble with the machine itself and not Mach3.  So switching to Mach4 wouldn't help.  Mach3 is still a good viable program to use.  I have not seen a need to change to 4 myself so let's figure out what is going wrong instead of adding another variable to the mix.

Since it sounds like your Y axis is the one having trouble based on what I'm reading in this thread, could you try swapping the X and Y motors?  If you do that and the symptom changes to the X axis then it's the motor.  If it stays with the Y axis then I suspect electronic noise is the culprit.  Proper grounding cable/wire routing is the best way to deal with that and there are whole discussions around eliminating noise.  I am much less than an expert in this area so I would differ to others to chime in on that.  So try the motor swap if it's not too difficult on your machine and let us know what you find from that.

The one thing that bothers me is that Auto Zero issue you had.  This still sounds like a DIR line issue, but with a different axis.  You may have to adjust the Step Pulse setting in the Motor Tuning screen.  What it sounds like now that I think about it is that the step pulses may be starting too soon after setting the DIR line high/low (depending on direction) and the stepper driver does not have a chance to sense the direction change before starting the motor spinning.  Most stepper drivers have some minimum required delay between changing the DIR signal and receiving the first pulse on the STEP line and if the first pulse occurs too early it could spin the motor in the wrong direction briefly.  This would account for the .150 difference in your Y zero operation and would also account for the brief push down before retraction on your Auto Zero operation.

Keep us posted on your findings,

Stephen "Highspeed" Kruse

EDIT:  While posting this, I see Hood also replied.  Hood, what are your thoughts on that DIR/STEP timing I discussed above?
Title: Re: Machine had a another brain fart.
Post by: Hood on September 28, 2016, 03:43:01 PM
It is a long time since I used steppers but I do know some drives require a wider pulse and also require a certain active state for the step, whether that is Hi or Lo depends on the drives.
Hood
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 28, 2016, 03:48:15 PM
What about changing the step pulse and dir pulse setting in motor tuning. I dropped the velocity down from 147 to 100 and changed step and dir pulse from 0 to 3. Im going to try that before i go switching motors out.
Title: Re: Machine had a another brain fart.
Post by: Highspeed1964 on September 28, 2016, 03:49:39 PM
Sounds like a good plan to me.
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 28, 2016, 04:11:20 PM
Ok reren the program. Was doing fine then it tried to cut a big chunk of the corner off rechecked my zero and x is 1.4796 off and y is -.5190. I checked the toolpath screen and there is a white line where the too went through thw material. Ill try and post pics
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 28, 2016, 04:18:06 PM
My phone wont let me post a pic.
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 28, 2016, 05:08:50 PM
Could this problem have been caused by a bad program. Ive been useing a usb stick to transferprograms to the machine and ive been changing the programs alittle and saveing over the old ones with the same name. I resaved the program under a new name on the usb stick and ran it 3 time with no problem yet. Is it possible there was an error in the programing
Title: Re: Machine had a another brain fart.
Post by: Hood on September 28, 2016, 09:31:40 PM
It is possible that it was some sort of corruption.
Are you copying the programme from the stick to the machines hard drive or are you running from the stick? If the latter then that could be your problem as it is not recommended to run code direct from a stick.

Hood
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 29, 2016, 12:20:55 PM
I save the program to the stick then load it to mach3 directly from the stick. I would run the program once and was usualy fine. The problems would usually start after it was finished the program and would regenerate for a new piece of material
Title: Re: Machine had a another brain fart.
Post by: RICH on September 29, 2016, 04:48:44 PM
Are you using one of those Chinese 6030 controllers?

rich
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 29, 2016, 08:22:38 PM
It has gecko stepper drivers on it. Its on a shoptask mill and lathe converted to cnc.
Title: Re: Machine had a another brain fart.
Post by: RICH on September 30, 2016, 11:20:14 AM
Think your steppers are skipping.

The problem is just what is causing it.

A stepper can skip and continue with a small change in position OR just go totally out of control.
Usually you can hear the skip. If repeatable when running a program it could be easy. If random it can be a real PITA.

First know what you max velocity and acceleration values are and reduced them by 50% or more.
See if that helps, it could be velocity or accel and that is also related to the system components.

If not would need to ask for a lot of info to see if could pinpoint the problem.

I just went thru this with another user.

RICH
  
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on September 30, 2016, 02:57:03 PM
I cut the velocity down by a third  and still had a problem. Resaved the program under a diffrent name instead of saving over the old program and ran it 3 times with out a problem. Also made a new program today and has been fine. Hopefully there was just a glitch in the program.

Fingers crossed.
Title: Re: Machine had a another brain fart.
Post by: Bhohbein on October 13, 2016, 04:20:59 PM
No its not a chinese controler. I did not build it someone else did though. I have acabinet atached to the side of the machine that has the computer mother board   Power supplies and it has gecko6203 steeper drivers. I think i had a corrupted gcode file. Renamed the file and resaved it and has been running fine.  Hopefuly i have no more problems with it. Ill let u know if it happens again.
Title: Re: Machine had a another brain fart.
Post by: rcaffin on October 15, 2016, 04:47:32 AM
The new version runs, so it is not Mach and not the hardware now. It MIGHT have been a transient hardware problem, but unlikely. It MIGHT have been a faulty program, but also unlikely as as saving it under a different name made it work.
It MIGHT have been ... a bad block on the disk.
You renamed the file and saved it as something else: that would have written it to somewhere else on the disk. Can I suggest running CheckDisk or equivalent?

Cheers
Roger