parts with arcs grow

[balbs]

Having some issues with my DIY water jet that really doesn't make sense and I am curious if you guys can give me any feedback or anything I can try. Currently the steps per in are near perfect to within a few thou. If I cut a square the part dimensions are perfect. However I am cutting a bracket for a customer that is roughly 12.75" by about 3" at its widest and 2" wide at its narrowest. The part has large radius on every edge and bolt holes in the middle. Also the part grows about .125" on the long side and a little on the short sides. I am using sheet cam for my gcode and worked with Les to test a few things and everything came out perfect on the gcode side. Tried abs and inc arcs as well as arc line segments with no change. I have a test part that is correct dimensions that was cut on a pro water jet. if I run the gantry again the end stop and move it the exact length of the part I can put the part in between the carriage and the end-stop and its perfect.
Things I have also tried:
Switching to metric units
I've used both mach3 and uccnc (both sae and metric on both)
abs vs incremental arcs ( I think i did it properly)
line segment arcs
MP1000 THC post (same on I use on my plasma without any issues)
replaced gear rack and gears with different brand in case there was issues

Gantry runs very smoothly, 46v, mx4660, uc400eth to a laptop. Same laptop make and model I use for the plasma without issues

I have a uc400eth on my plasma table that uses the exact same laptop as I am using on this and it has no issues. I might try and swap the uc400 but will only be able to test with mach3.

Help!

[cncdrive]

It has to be something with your machine mechanics, because:

- The UCCNC has path tolerance control and it will never make larger path error than what you setup.
- It is impossible that 2 softwares UCCNC and Mach3 have the same bug. (Or if mach3 does not do it then it is an issue with your tolerance setting in the UCCNC, the tolerance is setup too high for your requirements.)
- Nobody ever reported such issue.

Is there racks and pinions drive on your machine?
What I suspect is elasticness of the racks and pinions on your machine and since circles change the movement vector continously it is the most visible when circles are cut.
Can you try to cut slower to see if the issue remains or lowers?

[balbs]

I have replaced the gear rack to ensure there was no issue like a defective gear rack. I have tried multiple methods for calculating the steps per in and when I move 40" it moves exactly 40". If I cut a square part the part comes out correct in size, its only when its a more complex shape with multiple arcs that it has an issue. I am not necessarily saying its a bug but its more an issue that I need help with. The machine uses rack and pinion drive with minimal if any backlash registered. I have cut ranging from 5ipm up to about 30 with no change in part dimensions. I have also confirmed that the machine returns back to the exact same spot when the job is done. Is there any way to force reflash the firmware on my uc400? Any ideas for other things I can test? I am desperate at this point for a solution, it doesn't make sense.

[cncdrive]

The UC400ETH can't have a firmware issue, or if it does then it would automatically correct it. It never runs with broken firmware!

I think it is not backlash, but elasticness of the driving mechanism, that kind of issue shows itself on complex shapes when the cutting direction and so the speed of the axis is continously changing.

[balbs]

The carriages are the shopdroid 4x heavy duty carriages on 2x4 steel tubing. the y axis is dual carriage dual motor that are digitally slaved separate drivers. The gear rack is VHB taped directly to the steel beams. gantry is 5x10' squared to within 1/32"

[cncdrive]

How is the pinion pushed into the rack? Is it stiff enough? By elasticness I mean that it moves in and out as the force changes with the accelerations creating tolerance problems like larger circles and arcs.
It is a usual problem with rack and pinions drive, the solution is mostly to use 2 pinions instead of 1.
And this problem occurs more with heavy gantries, because the forces on the pinion are then larger on accelerations and deccelerations due to the higher mass the rack and pinion have to move.

[balbs]

Its spring loaded with a fairly heavy duty spring. I suppose I could add another spring just to rule that out. The carriage is fairly light, its a waterjet so the cut head doesn't weigh that much and there aren't any resistance forced in the x or y when cutting. I have run the programs with a sharpie as well as air cutting to confirm that any potential x and y forces weren't somehow causing the issue.

[cncdrive]

Yes, the spring will be the issue, beleive me.

[balbs]

So do you recommend adding an extra spring to increase the pressure? I can test that tomorrow.

[cncdrive]

Well, I would recommend to rebuild the driving mechanism, e.g. with 2 pinions, but it's not my machine, so I'm just advicing not recommending.
If you could fully fix the spring and test it without spring with the pinion fully fixed down on the rack would be the best. (E.g. on a short distance with some circles)
Unfortunately those forces on the pinions are usually much larger than people think, you think hey this spring is so stiff I can barely move it with my hands, but you only have human power and that is a machine where much larger forces can occur than what you can do with your hands and what you think of.
And the issue with too high spring force is that it will kill the pinion and maybe the rack too, which one is softer.
Rack and pinions drive is therefor not very good for precise works. 2 pinions can inprove it though. And also helical racks are better than straight tooth ones.
I'm talking from experiences, we building machines about 15 years or so.