Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Hood on February 06, 2006, 03:15:46 PM
-
Hi, I have a very strange one here, I have been using the PhotVcarve program to do a few engravings, I did a picture of a Buzzard and it came out good but decided to do a finer version and that is where my problem made itself known as you can see from the pic below. I thought possibly the code output from PhotoVcarve was wrong but looking at it there is no problem. I ran the same code today and exactly the same problem happened at exactly the same place, I looked at my ballscrew on the Y axis and it had some chips on it so I stripped and cleaned and put it back. I then ran the code again and exactly the same again. I rewound and ran back over three times and each time I could see the Y stopping and then carrying on again. I then single blocked and it ran OK then again ran at normal speed and again the fault.
Anyone have any thoughts on this, especially as it is exactly the same spot every time, even if I move the X0 Y0 so that its a different part of the table/ballscrew that is being used when the error happens.
Hood
(http://mysite.wanadoo-members.co.uk/hoodsmachiningpages/BuzzardFine1.jpg)
-
What are you running the feed over ride at??
-
The feedrate is 1000mm/min and no override.
Hood
-
Is this a stepper machine? if so try turning down your acceleration
-
Yes its a stepper and will try turning down the accel tomorrow, but really if that is the problem should it not show up before that, it is 30793 lines of code into the program and always on that line.
Hood
-
It would be some VERY odd move that is right on the edge of the machines ability to do the the move... I have had this happen to me in the past with a stepper mill...
-
I have the Y axis rapids at 2M/min and X 4M/min and they have been run faster, so 1M/min should be within its capabilities. Also the angle is exactly the same for all the lines so its not as if its changing course thats causing the problem. This is strange, think tomorrow I will turn down the accel like you suggest, then if no change will hook up another computer and also might try swapping the X and Y geckos. If you have any other suggestions of things to swap/replace/try please shout out.
Thanks
Hood
-
Hi:
Check the backlash is not on as well, Its not likely given the symptom but I though Id check..
Art
-
OK, it shouldn't be on as I have never set it up but brought the XML home with me so I will go load it on this computer and check.
Thanks
Hood
-
Just checked, backlash is off, but thanks for the suggestion.
Hood
-
Well heres an update, went to the workshop this morning and tried to run the program from the previous Z2 move before the problem and it ran OK. Then decided to run the whole program again without making any changes and at exactly the same point it again happened, had taken the latest version of Mach with me so installed that and things got worse, the first attempt at referencing the Axis only the Z moved, next attempt the Z and Y and the third attempt all moved. Then on go to zero they all moved but seemed slow, so I referenced again and when I then attempted to zero again the X axis stalled. I could then repeat this every single time I tried after that, also I noticed that the rapids were slow with a G0 move, only 1.5M/min, I tried G1 F4000 and it too was limited to 1500mm/min. I checked to make sure that I had no rapid override set and I didn't, so I thought it must be the parallel port on the motherboard or something . I decided to install 1.84 version and this time all went OK , rapids were back to normal, no stalling on X or Y so decided to run the code again and this time it completed without error, in total I ran the code three times and it has worked out like it should on all three occasions.
So in summary not sure what the problem was/is, possibly a corrupt Mach, possibly something else on the computer interfering with Mach or possibly bad memory or motherboard. I think I will get another motherboard in any case just to be on the safe side. If anyone has any thoughts I would appreciate it and also thanks to Barker and Art for the help.
Hood