I’m trying to figure the best way to handle some very large jobs that I have coming up.
What i have noticed is that the machine doesn’t necessarily keep the zero positions EXACTLY if it’s been powered off.
I had a long job the other day where i broke a bit, so i told the machine to stop and shutdown (sending the spindle to the back right corner) and then powered it off, waited a few days for a bit to show up. Powered it back up, installed new bit and resumed the 3d carving and it was off 1/16th of an inch, which in this case ruined the job.
The material was never moved, so the only different is in the homing procedure relative to the zero position from the power off to power on.
Just looking for some input/thoughts on the matter. Right now my only solution is to leave the machine powered on.
It’s always a good idea to make a Zero-point mark(with a V-bit is even better) to double check against in situations like this. Or make a corner the Zero point.
The hall effect probes usually provide more accurate repeatability than 1/16" though. So you may want to check that your machine is good and square & that the motion mechanics aren’t binding anywhere.
I don’t have a SPRO5, but on my SPROXXL, I have no issues pausing a job & then leaving the machine to sit idle while still powered on until I get back. Never had an issue with my 3XXL either. That way, the zero point pretty much can’t change & I can easily resume later. The power draw of the stepper motors & controller while idle is under 40W.
But I don’t know if that holds true for the SPRO5.
I am curious to know from Carbide if this is even possible. I have used the Pro 5 about every day for a year using the same Zero for often weeks at a time as I use a fixture and material of the same thickness for most of my jobs. I shut the machine off nightly and haven’t run across this issue.
I notice this too. Turning off the machine means the exact zero point is inaccurate, for every job I’ve had this happen too. Maybe it’s the stepper motors moving a hair after poweroff, maybe there’s dust on the sensors. I don’t know what causes it. There was another thread recently talking about it.
Heat kills electronics. The stepper motors hold position when not in use AND powered up. The body helps to wick off the heat, but it is still warm. The controller also has heat that would remain there while we are away. If I have a mean time to failure (did I say that right) I don’t want to reduce my usable time by leaving it on.
Excessive temperature kills electronics. I’ve seen no indication on my machines that idle power consumption results in anything approaching excessive temperatures - I’ve checked with my thermal imager.
If I want to reliably maintain my zero point on a project run but need to pause it for some hours for whatever reason, I have no problem or issue doing so.
I have had the same problem on my S5P. Yesterday, I resumed a job that was glued to the work surface and the Z was .03” off and the Y was off around .002. I restarted the job, as I had only drilled 4 holes, and noticed that it was drilling my holes in a somewhat oval shape now. I had also shut down my machine and parked it for the weekend. Not the first time I’ve noticed this. I make sure I rezero with every start up of the machine now.
Not sure, guessing just being 1/16th wasn’t a lot of room for error. I was running 40/20 on the speeds, i slowed it down after that to 30/15 and didn’t break anything. I was cutting Ash as well, which is pretty hard.
I didn’t see it break, i was elsewhere in the shop. Came to check on it and it was just cutting air, so i was like darn!
Here is a test you can try. Design a 3" x 3" square pocket and set it to go down 0.5" and mill that out. Power cycle and rehome. Without changing Zero cut the same pocket but go down to 0.25". If the two pockets line up homing is working properly. If not then something is wrong.
I’ll start by saying that I’m not sure this accounts for all of that 1/16" it was off but it is something to think of.
Homing switches are not fully repeatable. Inductive sensors have a range that they will activate in and a repeat accuracy spec. I don’t know what the spec is on the sensors used by Carbide 3D. It is usually presented as a percentage of the activation distance. As an example, here is the spec for some sensors I use on another much larger machine:
If a repeat accuracy spec is listed, it is usually 5% or so. On an 8mm distance sensor that would be about 0.4mm or about 1/64". I have seen sensors with as high as a 20% rating. This is over simplified. There are lots of other variables that get into electronics design, sensor positioning and material science.
The other thing I have seen in my machines is the temperature when the homing sequence is done. I can easily measure a change in homing location when it is cold (40F) vs warm (80F).
Sure enough, your suspicion could be repeated by setting the zero shutting the machine off and back on, and checking the zero again. Have you done that and isolated the actual problem?
Well, i don’t know that is really a problem per-say, it could well be operating within spec. Sounds like a number of people have the issue. We’re talking about a $4000 CNC, not a $15,000 one, so maybe it just is what it is.
If i get time, i’ll try a few tests with powering on/off, but I don’t usually have time for that, i’m usually already 2-3 weeks behind on customer orders, never get the time to test things out.
