X axis looses steps in the exact same line of code each time.

Hi there,
Im loosing steps on my X axis at a certain line in the code. I'm running mach3 on windows 7 with a printer port to a breakout board then to individual cards. My stepper motor will make a buzzing noise just like if you were to try to run you axis into an end stop. it lasts for a second or two. It doesn't do it every time but almost every time. I made 7 pieces and it did it on 5 of them.
I have 1200 OZ motors, ball screws and linear rails. I removed the cars off of the rails and checked to make sure they were not clogged with sawdust the cars and the gantry head moves easily along the rails. the ball screw also rotates freely.
I'm cutting slots with a 3/4" bit in Ash I plunge down 3mm at 190mm per minute, then it travels to the end of the slots and plunges again.. travel speed is 1800mm per minute. My acceleration is set to 120mm per second in ports and pins. It happens right after a plunge as it starts to travel it makes it about a 1/4 inch then hangs up. there are 9 slots and it only does it in one slot at the same depth and place. I dialed down my acceleration and speed. same thing.
I run all shielded cable...…. power supply in in a separate box from the BOB and control cards...…

This is what I'm cutting the circled slot is the trouble spot.

https://www.cnczone.com/forums/attac...d=452638&stc=1


And here is the lines of code where it happens.
This is in mm

Z-12. F190.
X417.014 F2000.
Z-15. F190.
X436.014 F2000.
Z-18. F190.
X417.014 F2000.
Z-21. F190.
X436.014 F2000.
Z-24. F190.

Any suggestions would be great!

Glen

I've had the exact same thing happened with my first CNC machine (Mach3, parallel port, Chinese TB6560 driver/breakout board). I guess the "right" combination of X/Y/Z motor speeds caused some kind of a resonance. The solution was to replace the TB6560 board. The new board has been working fine ever since - 7 years and counting.

There's also a bug in Mach3 where in certain instances, it will accelerate far above the acceleration rate.
Try changing the F2000 to a lower feedrate and see if the stalling goes away?

Quote:

---

Originally Posted by CitizenOfDreams

I've had the exact same thing happened with my first CNC machine (Mach3, parallel port, Chinese TB6560 driver/breakout board). I guess the "right" combination of X/Y/Z motor speeds caused some kind of a resonance. The solution was to replace the TB6560 board. The new board has been working fine ever since - 7 years and counting.

---

Thanks for the info, I had planned on finally swapping the parallel board out today to a new USB board I’ve had sitting around for a year. Was thinking about it last night.


Glen

Quote:

---

Originally Posted by ger21

There's also a bug in Mach3 where in certain instances, it will accelerate far above the acceleration rate.
Try changing the F2000 to a lower feedrate and see if the stalling goes away?

---

Thanks Gerry, wasn’t aware of the bug. What I did do and it still failed was to override the feed speed in the Mach3 screen down to 1800mm per minute.Why don't they fix the bug????

Going to swap out the BOB to a new USB and see if it still occurs!
But before doing that I’m going to generate anew TAP. File in Fusion and just select that row of slots and repeatedly air cut that path and see if it happens again.


Glen

Quote:

---

Why don't they fix the bug????

---

Mach3 development stopped about 6-7 years ago.

So today I reran the problematic code and it missed steps in the first try.
So I went and redid the post processing and just slected the row of slotes that was problematic.
I ran that TAP. file over and over and over and it never missed a step!?!?!?! I even turned up my acceleration and it was still good.
Very confusing how the code will loose steps if I have all three rows of slots selected. But wont if I only select and runs the problematic row of slots.

Now I'm not sure if I should bother changing out my parallel port BOB for the USB one ????

Glen

What happens if you try a pocketing cut with a smaller tool?Just a suggestion to see if that one slot is at a machine location where there may be a problem.A change of BOB will be sending the same number of pulses as the parallel version,are they all being received by the stepper?I would be carefully looking at any joints and terminals on that axis just in case there is a hint of movement or a crack in a soldered joint.

Don't leave me wasting all my time,
Don't leave me helpless in this cold world standard deviation calculator

Quote:

---

Originally Posted by routalot

What happens if you try a pocketing cut with a smaller tool?Just a suggestion to see if that one slot is at a machine location where there may be a problem.A change of BOB will be sending the same number of pulses as the parallel version,are they all being received by the stepper?I would be carefully looking at any joints and terminals on that axis just in case there is a hint of movement or a crack in a soldered joint.

---

Thanks for your reply! If it was a loose wire or cracked solder joint, I think the problem would be more random .It occurs at the exact same line in the code each time it happens.

Quote:

---

Originally Posted by Biggyboy

So today I reran the problematic code and it missed steps in the first try.
So I went and redid the post processing and just slected the row of slotes that was problematic.
I ran that TAP. file over and over and over and it never missed a step!?!?!?! I even turned up my acceleration and it was still good.
Very confusing how the code will loose steps if I have all three rows of slots selected. But wont if I only select and runs the problematic row of slots.

Now I'm not sure if I should bother changing out my parallel port BOB for the USB one ????

Glen

---

That certainly sounds like a glitch in the CAM or post processor code. Pretty sure it is not a mechanical or electrical problem.

Quote:

---

Originally Posted by Jim Dawson

That certainly sounds like a glitch in the CAM or post processor code. Pretty sure it is not a mechanical or electrical problem.

---

Thanks for taking the time to reply! :-)
Did you have a chance to look at the TAP. file, where it acts up it looks good to me? What is your thought on the highlighted lines of code where it fails?

Z-12. F190.
X417.014 F2000.
Z-15. F190.

X436.014 F2000.

Z-18. F190.
X417.014 F2000.
Z-21. F190.
X436.014 F2000.
Z-24. F190.

Quote:

---

Originally Posted by Biggyboy

I ran that TAP. file over and over and over and it never missed a step!?!?!?! I even turned up my acceleration and it was still good.

---

...possible workaround if, its a bug. Program each row as a Subprogram to test.
DJ

I don't see anything wrong with the code. Maybe it's a Mach3 issue?

Can you post the code that runs OK so we can compare the two?

Quote:

---

Originally Posted by machinehop5

...possible workaround if, its a bug. Program each row as a Subprogram to test.
DJ

---

I tired breaking the path up into two paths. There are three rows its the middle row with 7 slots in it.
If I just run the middle row by itself its ok. If I have that center row path plus the other path that does the other two rows it looses steps on the center row exact same slot location and depth pass.

I ended running the center row path by itself as a TAP. file then closed that files down and loaded the other two rows path TAP. file and ran it its OK that way.

I'm thinking ger21 might be correct about the Mach3 bug that causes the motor acceleration to over shoot.

Glen

Quote:

---

Originally Posted by Jim Dawson

I don't see anything wrong with the code. Maybe it's a Mach3 issue?

Can you post the code that runs OK so we can compare the two?

---

Its the exact same string of code, only difference is I did not include the other two rows of slots in the tool path.
So If I run the tool path to only cut the center row its fine, only when my tool path includes all three rows, does the center row miss steps in the same location and depth.
ger21 says about a Mach3 bug that can cause the acceleration to over shoot.

(SLOT5)
M5
T17 M6
S16000 M3
G54
G0 X532.516 Y59.5
G43 Z15. H17
G1 Z5. F2000.
Z2.5 F230.
Z-3.
X589.816 F2000.
Z-6. F230.
X532.516 F2000.
Z-9. F230.
X589.816 F2000.
Z-12. F230.
X532.516 F2000.
Z-15. F230.
X589.816 F2000.
Z-18. F230.
X532.516 F2000.
Z-21. F230.
X589.816 F2000.
Z-24. F230.
X532.516 F2000.
Z5.
X465.014
Z2.5 F230.
Z-3.
X503.516 F2000.
Z-6. F230.
X465.014 F2000.
Z-9. F230.
X503.516 F2000.
Z-12. F230.
X465.014 F2000.
Z-15. F230.
X503.516 F2000.
Z-18. F230.
X465.014 F2000.
Z-21. F230.
X503.516 F2000.
Z-24. F230.
X465.014 F2000.
Z5.
X417.014
Z2.5 F230.
Z-3.
X436.014 F2000.
Z-6. F230.
X417.014 F2000.
Z-9. F230.
X436.014 F2000.
Z-12. F230.
X417.014 F2000.
Z-15. F230.
X436.014 F2000.
Z-18. F230.
X417.014 F2000.
Z-21. F230.
X436.014 F2000.
Z-24. F230.
X417.014 F2000.
Z5.
X222.064
Z2.5 F230.
Z-3.
X387.536 F2000.
Z-6. F230.
X222.064 F2000.
Z-9. F230.
X387.536 F2000.
Z-12. F230.
X222.064 F2000.
Z-15. F230.
X387.536 F2000.
Z-18. F230.
X222.064 F2000.
Z-21. F230.
X387.536 F2000.
Z-24. F230.
X222.064 F2000.
Z5.
X173.586
Z2.5 F230.
Z-3.
X192.586 F2000.
Z-6. F230.
X173.586 F2000.
Z-9. F230.
X192.586 F2000.
Z-12. F230.
X173.586 F2000.
Z-15. F230.
X192.586 F2000.
Z-18. F230.
X173.586 F2000.
Z-21. F230.
X192.586 F2000.
Z-24. F230.
X173.586 F2000.
Z5.
X106.084
Z2.5 F230.
Z-3.
X144.586 F2000.
Z-6. F230.
X106.084 F2000.
Z-9. F230.
X144.586 F2000.
Z-12. F230.
X106.084 F2000.
Z-15. F230.
X144.586 F2000.
Z-18. F230.
X106.084 F2000.
Z-21. F230.
X144.586 F2000.
Z-24. F230.
X106.084 F2000.
Z5.
X19.784
Z2.5 F230.
Z-3.
X77.084 F2000.
Z-6. F230.
X19.784 F2000.
Z-9. F230.
X77.084 F2000.
Z-12. F230.
X19.784 F2000.
Z-15. F230.
X77.084 F2000.
Z-18. F230.
X19.784 F2000.
Z-21. F230.
X77.084 F2000.
Z-24. F230.
X19.784 F2000.
Z15.
G28 G91 Z0.
G90

https://www.cnczone.com/forums/attac...d=452704&stc=1

...post us tomorrow. Get some rest is my advice. :cool: before getting madhatter

If it runs OK up to the offending line, then I would say it's a Mach3 problem. I don't see any difference in the code.