CoreXY Z-Axis steps/mm question
-
Hi guys,
I'm building my first ever custom printer, following the 300x300 BoM for the D-Bot but with 3x lead screws driven by one belt for the Z-Axis. Previously had a Wanhao i3, so not new to printing but new to building/programming from scratch.
Really happy with how straight forward the Duet-Wifi has been to use so far and how well documented everything is! I seem to have everything working, the only thing I'm still tweaking is the steps/mm value for Z.
My setup is as follows:
1x NEMA17 motor with a 20 tooth pulley (1.8 degrees)
1x GT2 timing belt
3x 8mm (2mm pitch) lead screws with 40 tooth pulleys
1x Bolt with bearings for additonal tensioningSo, I've used the Prusa Calculator for both X and Y and it has worked. However the Z axis doesn't seem to want to play ball. I'm putting in the following:
1.8 degrees, 1/16 micro steps, 2mm pitch, 2:1 ratio. This gives me 800 steps/mm. However this value in the firmware thinks it has only traveled around 22cm by the time it has reached the end of the 34cm of travel.
By using the current steps/mm and the measured distance against the actual distance I have tweaked the steps to about 612. Am I missing something as to why the calculator is not providing me an accurate result?
-
Chances are you might have a multi start lead screw. While it is a 2mm pitch, the lead will then be a multiple of that. E.g. 4-start lead screw with 2mm pitch = 8mm lead (8mm travel per turn). In that case 8mm would be the number to put in the calculator, not 2.
-
Chances are you might have a multi start lead screw. While it is a 2mm pitch, the lead will then be a multiple of that. E.g. 4-start lead screw with 2mm pitch = 8mm lead (8mm travel per turn). In that case 8mm would be the number to put in the calculator, not 2.
This looks like a great starting point, thank-you very much!
-
Newbie disclaimer, probably not the right way but worked for me…after running the calculator, my z movements were still way to low. I basically used the same method as extruder calibration and kept changing moment values until bed moved the same amount as i was telling it to. took a few tries but it appears to have resolved my issue. I started with small movements about 20mm once i got that right, bumped up to 200mm z+ and z- and again at 300mm bigger movement should multiply any inaccuracy so i think it did a good job of getting movement dialed in.
-
Newbie disclaimer, probably not the right way but worked for me…after running the calculator, my z movements were still way to low. I basically used the same method as extruder calibration and kept changing moment values until bed moved the same amount as i was telling it to. took a few tries but it appears to have resolved my issue. I started with small movements about 20mm once i got that right, bumped up to 200mm z+ and z- and again at 300mm bigger movement should multiply any inaccuracy so i think it did a good job of getting movement dialed in.
Thanks for the input, this is where I am at the moment. Marking off against the frame where I'm starting and ending.. noting the distance and recalculating. Then repeat!
-
As pointed out above, you most likely have a multi start screw. Many people get confused by "pitch" and "lead". So to clarify, the pitch is the distance between adjacent thread peaks (or troughs) while the "lead" is how far a nut will travel along the screw in 1 revolution. On most nuts and bolts, it just happens that the pitch and lead are the same because they are single start threads. However, if you have say a 2 start thread, there are two helical grooves set 180 degrees apart. This means that the distance between each peak will still be 2mm but the distance between one peak and the next on the same helix will be 4mm. A 4 start thread has four helical grooves set 90 degrees apart so typically the pitch will be 2mm but the lead will be 8mm.
So with lead screws, it is the lead that is important, not the pitch. I did a bit of a write up on my blog which has some pictures describing the thread pattern of multi start screws which may help your understanding https://somei3deas.wordpress.com/2017/01/31/z-axis-lead-screws/
To calculate your steps per mm is not that difficult. Using a 1.8 degree motor means that it will take 200 full steps to rotate one revolution. As you are using a 20 tooth pulley on the motor and a 40 tooth pulley on the screw, it will take 2 revolutions of the motor to rotate the screws 1 revolution so 400 full steps. Now all you have to do is divide that by the lead to see how many steps it will take to move 1 mm. So if it is indeed a single start thread with 2mm lead the steps per mm will be 400/2 = 200. Due to reasons that I won't go into, 3d printers always need the steps per mm to be one 16th microsteps rather than full steps so you multiply the 200 by 16 and get 3,200. If your lead is 4mm (as it would be with a 2 start screw), then your steps per mm will be half that so 1,600 and if it's a four start screw, the lead will be 8mm so the steps per mm will be 800. So with 2:1 gearing it's 400/screw lead in mm * 16.
-
Thank-youi for such a thorough and thought out response. Looking at my rod it looks like it's a 3 start (is that a thing or is it always an even number?).
Howver taking this into account when using the calculator still give me innacurate results. Using a dial gauge indicator I seem to have it accurately moving 1mm at 655 steps/mm. But this gives me a leadscrew pitch of around 2.4427. Can this be right?
Granted, it is currently moving a mm when I tell it to but the figures that get me there don't seem to make sense and I'd like to understand how I'm getting there.
I did notice that the motor is using the default current setting of 800mA while the motor is rated up to 2A. The motor doesn't seem to struggle raising and lowering or holding the bed with the current settings, but would increasing this affect the steps/mm it requires?
-
No the current doesn't make a difference there's broadly either enough for it to move a step (or a series of microsteps) or not. If not increase it but don't go higher than 80% to prevent heat build up. Sounds as if you're giving it enough.
Maybe your leadscrew is imperial? If it moves 1mm when you ask it to then you're there.
-
As DJ says, it could well be an imperial lead screw which will definitely give you odd numbers. If it was 2.54 that would make sense as it would equate to 1/10th inch but 2.4427 doesn't equate to anything I can think of off hand. I have never heard of a 3 start screw. They are usually 2 or 4 but I guess 3 isn't impossible.
-
It could well be that my gauge isn't as accurate as I'd hoped. Will have another play this evening with 2.54 as the pitch to see what results that gives me.
