I’m currently using mechanical micro-switches with arms for homing.
(and an inductive switch for the rotational A axis )
When I have to pause a cut to switch everything off and continue the next day (after homing), I still notice a small step between today’s and yesterday’s cuts.
What could I look at, that is more precise then cheap micro-switches?
It’s your !
I’ve read that levers reduce accuracy and precision.
I don’t have switches. So when I stop I issue the command G0 X0 Y0 Z0 Then when I start up again I just say I’m at X0 Y0 Z0. It seems to work to me.
+Paul it’s simply archaic, obviously you don’t need any precision
@E_eF_CNC_Falegname whether I need precision or I don’t I have it just the same. My machine never loses steps.
@Paul_Frederick Does it looses 1/8th Micro-Steps when powered off in earthquake territory with regular micro-tremors?
@Marcus_Wolschon we had an earthquake here once. I thought I was cracking up. Because I was sitting in my garage and I saw the concrete slab rise up towards me in a wave. I was watching it thinking that’s not normal. I really thought I was losing my mind. Because my next thought was, so this is what it feels like to be crazy? As far as I was concerned I was seeing things. But as it turns out it really was an earthquake. It was the same one that broke the top of the Washington Monument off. So it really did happen.
So you are saying your homing is changing each time you rehome it? Micro switches shoujnldnt change - they are really simple. Maybe they have built up crud and are shorting before contact?
Anyway - use an optical sensor. Light doesn’t change, so any difference in movement must be down to your machine, not calibration.
These are really simple and will cleanly trigger consistently with an object passing through the sensor.
You will need to ensure swarf doesn’t fly through, as that could trigger it.
If thats an issue, there are inductive switches, but much more of a pita to install.
I live in NZ and we have earthquakes daily - never affected any machine I’ve worked on.
I conducted some experiments and it seems that the axis jump quite a bit when I power the the stepper-drivers back on the next day. Also the mount for the X endstop flexes a bit as the axis doesn’t stop quickly enough after making contact.
An optical sensor probably makes most sense. I’ll see about designing a mount for it.
I just made a contact plate for tool-length (need that anyway) and am using that together with a dial gouge as a temporary workaround.for Z and X.