I recently ran into an issue that’s occurred a few times where when using the keyboard to move the bed left (-x) the bed over-travels and hits a hard stop… Thunk.
This is after the system was properly referenced. (Hard Stop check box is clicked in settings).
So I initially thought there was an issue with the limit switches until I realized that the Stop on the bed on the right side can never make contact with the switch with the bed all the way left. Why is it there?
Secondly, manually pressing on the limit switch both X and Y does nothing, It doesn’t stop movement when referencing, it doesn’t stop movement when jogging, it doesn’t stop movement when commanding a position.
This 770Mx is new with about 30 hours on it so far and I was going to check on the gibs since I have chatter at 12OClock and 6 O’clock when cutting circular features under load. However this over travel and stop issue seems more pressing at the moment.
Any insight would be appreciated. The image below shows how with the bed all the way left the limit switch is nowhere close to being activated by the block.
Do we know where to find the status of the limit switches in pathpilot now? Mine are missing from the status page as the troubleshooting and tutorials show it.
The limit switches are non functional on the MX machines. They’re a bit like a vestigial tail, remnants from the lower spec M machines that use steppers instead of servos. You can remove them without causing any issues as I and many others have.
Without the limit switches how does the servo / bed know where the end of travel is? This morning when I powered up the machine the bed was roughly centered and when I referenced the x axis it moved about 2 inches then stopped acting like it had successfully referenced. At least pathpilot said it was referenced.
Pressing the reference button again and it then moved full travel and successfully referenced. I’m suspecting there is something else going on it… If the servo can think it’s at the end when its not I can see how it might not think it’s at the end of travel sometimes and causing the issue originally posted about.
I checked the electrical connections and they’re all secure clean and free of oil, etc.
The mx machines reference by what’s called hard stop referencing, meaning that table is moved until it bumps into a rubber bumper on the ballscrew. When the motor needs more than X amount of power to push the table it assumes that it has bumped into the stop. The other end of travel is handled in software.
It sounds like your table isn’t moving as freely as it should which is throwing off the referencing sequence. The scenario you originally posted about is happening when the reference gets thrown off then you run into the end of physical travel on the other side.
I took the covers off today to check the lost motion on the x/y axis/gibs and everything was well coated with oil… I have the automatic oiler on this.
The more I think about when it fails to reference it’s always in the morning when I turn on the machine right after the automatic oiler runs for it’s 12 seconds after a night of the power being off.
I’m not sure why the X axis would “stick” in this case but it sounds like that is the most likely cause. I’ll keep an eye on it. I don’t have the oiler running when the system is off, but I could keep that powered if needed so it runs during the night once.
I’ve got a new (November) 770M and also experience occasional tripping limits while jogging. It happens enough to be annoying and I’ve been wondering if there is something messed.
After reading this, it is almost certainly because I have manual oiler and my machine is used lightly, sometimes with days between usage. I’ve gotten better at hitting the manual oil plunger every time I start using it. I’m thinking that maybe my start up should include a full traverse jog and two pulls of the oiler.
Also, the comment about chatter when making round features - I had a project that included machining round bosses that were about .5 inch diameter. I was surprised to find that they were out of round (diameter varried) by a couple of thou depending on the axis measured.
I tried a really slow feed and got the same results, so I concluded that I shouldn’t expect circular features of small diameters to end up better than a few thou.
This is a very new machine with a few (<10) hrs of actually machining on it. I would hope that it doesn’t need gib adjustment from the factory.
I had this problem and I think it was after I moved the end stops. I remember moving the limit switch stops, (trying to figure out how to put the ETS on the front of the table at table height to free u some space). But I remember the table didn’t reference correctly after that. I think it does use one of the hard stops on the x axis to reference but the other stop (on the right) is actually used.
For roundness I’m machining a part that’s .558" in diameter from 1215 steel using a 5/16" coated carbide endmill from Haas tooling. Part is finished with a “spring” pass after a finish pass.
I don’t have backlash compensation set… I’m not sure that’s helpful here… or is it?
It looks like it’s “roundness” is running between .558" and .555"
The smaller dimension lines up with 12/6 o’clock when cutting the circluar feature and the .558 which is what I’m calling for is at 9/3 o’clock.
@David_Belser the M machines reference by bumping into the limit switch. If you are bumping into the end of travel then you might be losing steps. Usually that only happens when you run the machine too hard into the material or when you crash it.
This has happened occasionally while jogging, no load. Just setup, moving to a probe position, raising Z for tool change or workpiece change.
I’d guess it’s happened maybe 3 or 4 times in Z and once in X. But that it likely because I’m in the habit of running Z to the upper stop when I’m “in” the machine. Gets it out of the way of the work and easier for manual tool change with power drawbar.
The fact that It’s happened more in Z is because that’s the axis I’m always running up to its stop. It’s annoying because if I trip a limit, don’t trust the counts anymore and have to reference and set the workspace zero again.
I’m familiar with lost counts, my prior mill was stepper only.
