My limit switches are frequently off by roughly 0.01" or more, forcing me to throw out about half of my projects. Any ideas on how I could increase accuracy here? Are there high end limit switches I can buy? (currently using a carbide3D set that came with my HDZ.
Thank you,
-Nicholas
Are you using push button or inductive switches?
Which direction or along what axis are they off?
One thing which I’ve found helps is to pull the gantry against the front end plate before powering up — this ensures that the gantry is consistently aligned with the frame.
Could you provide an example project (preferably small and simple) which shows this? Provide the .c2d file, generated G-Code, step-by-step notes on how you are securing your stock and setting zero relative to it and managing all tool changes and at what point re-homing is necessary, and photos showing each step of how you’re working and we’ll do our best to puzzle things out with you.
Inductive
Usually Y, sometimes X
I’ll try pulling the gantry forward
I might make a simple project just to show off this issue, Ill let you know if I do.
What’s your homing feed rate? Did you increase it? I’ve had a problem with repeatability after increasing homing speed, dropped it back to 1000mm/min and the problems disappeared.
Does it matter which end to use as a reference? Front? Back? (SO Pro)
You have to use the front so that you won’t get hung up on the homing switches.
It’s also important that the Y-axis belt tension is even for both belts.
video of issue
drifted was not the right word to use, the change in y position happens during the homing process, not while milling
Is the change in Y consistent across the X axis movement or does it vary?
consistent
post must be at least 20 characters
OK,
That’s good then.
One problem I found is that if you didn’t pull the X beam into the endstops to make everything square for startup on the job where you cut the jig you’re properly stuffed as ensuring a square startup for jobs on a non-square jig doesn’t work too well…
I am still on the mechanicals so no useful input on the inductive switches from me, sorry.
I apologize if I am missing the issue, but one thing I noticed is your use of a MDF jig.
I use a jig like yours to complete milling on the second side of an initial cut, when I first setup this jig I was using MDF. Then in CAD, once I flip the parts the machine should know where the parts were, but the MDF was expanding/contracting meaning I was not getting the cuts that I wanted. Could this be your issue? The MDF jig you have at the bottom of your stack may not be perfectly square due to moisture making the MDF expand/contract?
To solve this, I am using a jig that was cut out of aluminum. I was getting cuts that were off by 0.1 mm to 1 mm, due to my MDF jig expanding/contracting.
I know the issue is with the switches because the holes for the screws will be in the right place, then without turning the machine off, the Y axis of the next operation will sometimes be off ±0.0135
Are the screws in the right place? See my comment above, you are using MDF figs/fences with expand contract.
I doubt that the fixturing is the issue because if I turn the machine off and on /re home the thing, it will have the right positioning (or with a similar range error ±)
Check for loose stepper motor pulley set screws.
While reading the carbide motion changelog, I noticed there was an issue with engrave toolpaths that only occurred when the XY zero was not the southwest corner, fixed in 529. I don’t have engrave paths, but it might be a similar issue, in retrospect I did not have these issues until changing to a northwest zero. I have also been using an older (than 529) version of carbide create/motion.
To address other possible issues, I am installing a Saunders baseplate and fully tuning my machine this week.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.