Hello finally got some life in the machine. All built on 1500 c beams. 1605 ball screws. Open Builds box and high torque motors. see video. What exactly is this? Pushing motor too hard. Maybe some alignment issues? Running at 2000 for speed. 150 for acceleration. Amperage is turned up to max. Motors are all cool to touch. 24 volt power supply. Can run the y at 2500mm and 150 with no issue. Standard setting on box. I can drop to 1500 and 120 and have no issues. Just testing so want to see what it can do. Thoughts on the video? I’ll add without motor hooked up runs well with a drill attached or by hand.
Only thing I can think of is linear rail alignment issues- they need to be parallel along the whole length to around 0.01mm or less to run smoothly- by dial indicator, not by carriage movement. They don't come as nice straight lines either, a lot of the time, kinda wiggly and you have to force them into shape. This is why they're supposed to be mounted against a machined shoulder, but that's hard to do with V-Slot. I'd check the alignment of the screw as well, relative to the rails. Everything else looks good, it doesn't seem to stop upon acceleration, but the little extra "shimmy" at the end is throwing me- it's acting like it's an alignment issue but performing like it's an acceleration issue (which at 150mm/s^2, it's definitely not, you should be able to up acceleration to at least 2000, maybe more depending on the motors). Does this happen everywhere along the length of the axis, or just in this spot? Try turning the acceleration up a lot, to at least 1000, see what happens once it gets moving at top speed more. If it immediately cuts out (like this but in fast motion), maybe your motors are underpowered. I'd want at least 2-3Nm on this axis (280-420oz-in). It shouldn't be doing this with ANY motor completely unloaded like this though. There's gotta be something else wrong.
my reaction is 'sticky spot'. that is what my machine did at first, Vslots and belts and I tuned X and Y to 7500mm/min but was only moving on the near side of X, when it went to the far side of the rail it got stuck due to a tiny bulge in the rail and a protruding screw head. Slowing down to 3500mm/min solved it entirely. note that tuning max speed and acceleration is always a tradeoff. You can get a very high max speed with low acceleration. You can get huge acceleration but at lower max speed. (huge acceleration will jerk the machine which can cause the bit to hit the work in unintended ways) I set acceleration so it get up to speed in what feels like 1/4 second, then I adjust for max speed, when I find where it stalls I adjust down by 20% and then adjust acceleration up till it stalls, then drop it 20%. this gives me a pretty good balance and reliability. Reliable running is worth more than ultimate max speed (-:
Yeah, almost seems like there's too much force and it's causing a racking jam- not necessarily between the rails though it's possible, more between the screw and the rails, maybe? It would explain why, once the force is reduced under deceleration at the end, the carriage moves again. Usually means something's loose (or flexing). May or may not be something you can do much about, but 150mm/s^2 is very low acceleration to be having problems with. I'm not sure I see how with steppers (or most other motors, thinking about it) there should be much meaningful relationship between acceleration and top speed- acceleration is limited primarily by inertial coupling, how much your machine weight and friction- specifically under unbalanced forces like here- is trying to slow down the motor's commutation. Top speed under zero load should be where the motor's torque curve meets the minimum torque required to turn the screw under balanced-forces weight and friction of the machine, which should be roughly the same at both 50mm/min and 20,000mm/min. Not until you start getting into air resistance (which some machines out there do, as I recall!) should the top speed become limited by variable forces. Under normal conditions, it should be very close to the absolute maximum commutation speed of the motor based on stator inductance (and thus field collapse duration). I believe you could do your "stall - 20%" (which is pretty much what I do as well) consecutively on speed and acceleration and come up with essentially the same numbers in the end. This is very true though- grbl has no acceleration planning or even S-curve lookup maps. Purely linear acceleration, due to the limitations of the ATMega328p. Means initial jerk (second deriviative of speed, acceleration over time) is very high, which motion control components don't much like. Have you seen this? I've never seen a tool enter a workpiece at anything other than a constant speed as intended, this would be pretty concerning.
yes. Fusion360 adaptive toolpath with small retract height during those short rapids as it repositions for a cut. Z rapids up, XY moves, Z rapids down and stops so suddenly that the gantry flexes down and the tool leaves a divot at the restart point. Yes, there were some loose wheels but with those adjusted it can still happen.
Oh, that makes sense. I don't have anything flexible enough for that to happen yet, I'll have to keep an eye out.
Thanks guys, got some time tonight to troubleshoot. I will say I can run the axis with a drill at high speed with not issue. And with the screw not attached it glides pretty darn well. That said, really no precision tools at my disposal. I was able to run my Y up to about 3750 before it started getting angry. Pretty much duplicated what you see in the video. The one thing I'm second guessing is the location I used for the screw block. It is not in the center of my plate. It's right on the side (for easy access). I had to enlarge the screw holes too to help with a slight alignment issue. I'm thinking it might be pushing , or pulling it into racking? Thoughts on location? It's a pretty robust machine, or the best I feel I could put together with my tool set.
