- This topic has 5 replies, 2 voices, and was last updated 12 years, 9 months ago by
.
-
Posts
-
-
Hey Quentin,
It’s been awhile since I’ve posted. Morgan is working great but I’m having some issues calibrating for maximum dimensional accuracy. What arms do linkage_1 and linkage_2 correspond to? My arms aren’t exactly 150mm center to center. My psi arm is 148.3mm center to center and my theta arm is 149.2mm center to center. The arms measured there, are the ones attached to the shafts. Is there anywhere to adjust for variations in the distal linkages? I only see linkage_1 and linkage_2 in the firmware. If I print a 20x20x20 test cube it’s coming out approx 19.3×18.7×19.5 (need to adjust my z steps). But that x-y deviation I was trying to fix up with the arm measurements.
-
What happened to my post…
Here it is again:
Ideally you should print the arms on the same printer, in the same orientation and location (one by one) to ensure they are exactly the same length. That length you enter in the firmware.
Mixed lengths will make things a lot harder, and might make it near impossible to calibrate properly currently. The problem is that the uP needs to do the inverse kinematics calculations, and if it becomes too involved, speed will suffer greatly, and with it, quality.
Linkage_1 refers to the PSI inner (shaft mounted) arm, and Theta outer (attached to toolhead), and
Linkage_2 to Psi outer, and Theta inner. Oposites to form a parallelogram. -
Ahhhhh. Thanks as usual Quentin. Yeah this will be a problem as someone else printed them for me. I might just machine the arms to insure they are the correct lengths. Because all 4 of my arms are definitely all over the place. Ouch. I’ve gotten it fairly close with some guessing. I actually got a cube to print at 20×19.4(or so)x20. Which is good enough for random stuff to show off. But not quite there for anything requiring dimensional accuracy.
printing your calibration shadow.
On the spreadsheet there is a reference to M365? Scaling factor? How do I enter that exactly? Just give it a M365 X# Y#? Which is x and which is y? psi or theta?
Sorry for all the questions. And Thanks Again.
-
-
-
It can be a bit confusing…
M92 sets the steps per angle, in order to set theta and psi.
Do that first, and reprint SCARA-cal. Theta is x and Psi is Y according to M92.Remember to run the arm calibration again after changing steps/deg, of course omitting the section where you calculate the initial M92 values.
When Theta and Psi is close, X and Y should be as well.
If not M365 scaling will scale the actual X and Y. Remember that this will change your theta and psi, so it might take a bit of back and forth to get it done. I will try to work that math into the spreadsheet to make it more accurate in future – just need some time to hash it out.When your morgan is close, you can reprint your arms, one at a time. They should then be exactly the same length. Hang your first arms on the wall as a reminder of the struggle, and that there is proof that not all printers are made equal…
Q
-
- You must be logged in to reply to this topic.
