Ohhh, 3200 pulses/rev is a lot different! Most stepper motors have 200 steps per revolution, so 3200/200= 16 microstepping. The SLB is set up for 32 microstepping by default so that means you’ll get even more detail on the SLB but you’re right there is a point you get to where the level of detail coming from the motors has diminishing returns. With 32 microstepping on a 8mm lead, lead screw, that means that there will be 800 steps/mm of machine movement, meaning that the theoretical movement accuracy of the LongMill becomes 1/800= 0.00125mm or 0.0000492" which is just an unbelievably small amount for most hobby applications and is likely not able to be hit by any hobby machine realistically because of other sources of error from the machine mechanics.
I probably knew some of that when I originally was building my machine about three years ago. It vaguely rings a bell. Thank you for the refresher course.
As rigid as I have tried to make my build, I am nowhere close to that level of detail, as you said. It will be interesting to see if my aluminum jobs end up smoother. I don’t think I will see too much difference in things milled from wood and epoxy.
Thanks for the conversation. I am sure there will be more to come when I am back.
I have returned from my trip, and have been able to start to address the issue of machine calibration.
The settings for $100, $101 and $102 set to =200 was way off for my steppers and 1605 ball screws. All three of the axis on my machine use 1605 ball screws. So, the values for $100, $101, and $102 should be very close, but I will test for each axis.
After several trial and error values for $100, I have narrowed it to 321.4 using 100mm as a reference distance. I will be testing a larger distance to check the precision before I cut the first job using the SLB.
The issue I ran into setting these to this higher value is the stepper would fail at the default jog speed of 3000mm/min. Through trial and error I was able to run my tests at 2600mm/min. So, I change Max speed for X to $120 = 2600.
So, not the zippy 5000mm/min. of the default setting for the SLB. I am also concerned if I will be losing steps under load when actually cutting material.
Can Acceleration be lowered to compensate for this?
I love the way the SLB operates, and all the advantages it offers over the Arduino controller I had been using, but this is at the heart of my initial concern of switching to the SLB. Going from external stepper drivers that allowed for more current to the on-board drivers that are close, but less than the current I was using for my steppers.
One issue of concern, while I conduct calibration on the YY axis, has come about. Although the two steppers on this axis are from the same batch, over time, one may have become weaker than the other. While testing accuracy at 500mm, I have twice forgotten to reset the jog speed down to 2000mm/min. This has resulted in the left motor failing while the right turns. Obviously a scary situation that requires me to stop the trial immediately. I think I can operate within a tolerance to avoid this while milling, but I am not super confident given the fact that this has occurred.
@stoddad to answer your question, if the motors were stalling near the beginning of the jogging motion, then it’s more likely originating from acceleration settings that were too high instead of max speed. In that case, if you reduce acceleration then you should be able to bring max speed back up higher. I’m also doing some testing right now on an area of the SLB I realized I could teak to potentially get more power output from the drivers to the motors but that testing is still a WIP
For your Y-axia binding setup, I’d recommend also looking at acceleration and max speed so you can ensure that you don’t run into edge cases where one gets stuck up and the other doesn’t. In future, there are other things the SLB should theoretically be capable of that are yet untested which may also be able to help keep track of things like this
I had lowered the acceleration, so it is good to have that idea reaffirmed. I have not raised the default speed to test it, but I will do that. I am currently resurfacing my waste board. Since this pushes the X and Ys to there limits area-wise, and I am using a 1" surfacing bit that can hit some drag when there is a high spot, it might be good for testing speed.
As I get closer to normal operation with my machine, I am generally happy with switching from my previous control system. I like the features and easy of use of gSender and the SLB. I look forward to what you discover, as you develop both.
I did run into a strange thing while zeroing the axis and setting up for surfacing. I was try to return my X to Zero, so I clicked the Go To (blue box) next to Zero X in gSender (or at least that is what I was trying to do). My Z axis started climbing (quickly). I pushed the E-stop, and have been manually jogging back to Zero from then on out of concern. I didn’t try testing this again. Should that button take the axis back to Zero, or does it have another purpose?
If you go to Settings → Safety → Safe Height and that value is something that isn’t zero, then it will cause those buttons to move in Z before moving in X or Y so that’s probably what you encountered. If you set to 0 it should eliminate that type of movement
Great report back, glad to hear you’re enjoying your transition over!
