The router has once again plunged into a workpiece randomly despite the fact I have run the g-code once before already. This has happened multiple times ruining pieces, breaking bits and burning out the router. Im using fusion 360 and the simulations work perfectly. Flattening seams to be the worst offender. Its making me so mad i’m thinking of selling my machine because it is so unpredictable.
Onefinity CNC seems to have a really hard time running flattening jobs. I can run a job multiple time on the same piece then I will randomly receive an error.
If I stop and restart a job for some reason it forgets all of the heights and plunges straight into the workpiece.
I have had this happen only once after months of no issues on the original software machine came loaded with(July 2022). I am using Vectric software.
I ran several xmas tree bowls last week I would flatten one side and then the other switch to a bowl bit and when I was staring the 4th one it did that exact thing at the start of the carve. Bowl bit was at origin and drove itself straight down through the 1.25" work piece and stopped the router. The only error I received was my hitting the E-Stop. Thankfully it was at the origin and only took the corner off, did not even hurt the bit. I was able to still use the block of walnut. I came online and did not find anyone else that had anything like that happen, until today.
This confirms for me it is “a thing”.
If it is fusion or vectric caused you would see the nose dive in the gcode created.
If it happens when flattening the boards it may be related to the cables and connections being strained and opened, causing disconnects (if like me you may not use the fully extent of the 1F unless flattening the wasteboard, if that is the flattening you refer to). If you search the forums there are posts on loose connections and symptoms.
If it is in the code it would have happened the first time I ran the job.
I have cut large pieces before using almost all of the cutting area with no problems. There is plenty of slack in the cables and the connections are snug.
I have however noticed my display screen will go blank and turn back on from time to time whilst running the machine. I’m wondering if there is interference happening within the cables causing errors to appear.
Yesterday (vectric) 3 toolpaths to one file it said tool change after first path (which they’re wasn’t) pressed continue. Z just started slowly descending all way down ???
I just set up my Onfinity and I have had the mystery z-axis drift three times out of about 10 jobs. I checked the gcode - it’s clean. Short programs and manually readable - there are no z numbers that can make it happen and the machine is not switching from absolute to relative or anything like that.
I also played with the cables in case the “up” signal was not arriving at the motor but the “down” signal was arriving OK. That would do it, but after wiggling the connections and the joystick to make the z move, there did not seem to be any issues.
Finally I caught it at it just after a job had finished. The machine homed back to 00 and the Z went to it’s parking position, and then slowly drifted down - the motor was turning. The screen was clearly showing that the machine understands what it’s doing and is tracking the reducing z value.
I grabbed the controller and cycled the z up and down a few times - fixed.
I wonder if the controller is somehow sending Z signals when it shouldn’t. I have not tried playing with the controller when a job is running to see if that moves or confuses the Z settings. I’ll run that experiment later on.
Aiph5u
(Aiph5u (not affiliated with Onefinity))
11
Hey Doug,
usually, one would say, since the cabling is the weak spot of the Onefinity Controller, most Z issues have this cause. Stepper motors need an idle current to hold their position when not turning. In the case that one of the two coil groups of a motor have a connector or cable problem, the motor holding torque could fail and the motor start to move. In this case you would disconnect the cable of motor 3 (the Z motor) on the controller and directly on the stepper and with a multimeter set to ohms measure each of the four wires between their ends for conductivity.
But you say that when the Z motor moves, the controller shows the Z axis movement on the display. That can only mean that it is the controller that executes movement commands, otherwise the values would not be displayed. Are you sure the axis values moved in correlation with the movement? Can you reproduce the behaviour?
I grabbed the controller and cycled the z up and down a few times - fixed
Could it be dust on the controller button? Was the rate of movement consistent with the setting (i.e. One of the four movement speed controls). I guess the best way to check might be to buy a new controller? Or take the controller off and use the onscreen buttons for a while. Testing a negative is tricky.
I simply used the joystick to raise and lower the Z a couple of times. The drift stopped after that.
My interpretation is that the controller was sending (or the received was receiving) supurious “down” signals. However, this does not explain what happened while the job was running - I would have expected the joystick input to be disabled while a program is operating. I’m going to run some experiments machining air and see what I can find.
It’s too bad that the controller does not have an “off” button because it would be good to simply make that part of the pre “go” checklist “turn off controller” and see if that stops the problem.
Have you checked to make sure your bit didn’t pull out of the router chuck? Climb milling can cause enough force to pull the bit out. Really gotta make sure the chuck and hit are clean of dirt and oils and make them tight.
The uncommanded Z drop happened again - this time it was clearly the joystick controller. The Z-axis was stationary, I pressed the “on” button on the controller and the z started a slow descent down into the spoilboard. The display was showing the Z position changing and the movement stopped when I wiggled the joystick. Looks like the controller powers up with some spurious “down” signals in the transmission. Best not to touch (move, turn-on or mess about with) the controller when the machine is running.
Sounds like you’re getting drift from the joystick - more likely a problem with the joystick than the controller having “spurious” commands. Try a new joystick or Google your joystick’s model and “control drift”.