Unfortunately it is not often enough that one get great service in South Africa, especially not from the technology sector.
Today I was surprised…
Black box industries not only undercut the pricing on their DRV8825 based “Coolsteps 2” stepper drivers, but modified them for fast decay for me, and supplied heat sinks as well for the same low price!
You can contact them at info and the domain in the link above. At time of writing the page is not up yet, but I am sure they will find a good reason to do so soon!
Additionally, they are bringing out an updated version of the driver on my request – there will be a solder jumper on the bottom of the board that will activate fast decay with only a tiny drop of solder.
The boards below has been modified… Spot the fix (if you can)!
The amazing DRV8825 driver carrier has a similar flaw, strange jumping of steps in 32th micro-stepping mode.
After doing a bit of research, it seems that we need to put the chip in FAST decay mode, by pulling the DECAY line of the chip to 5V. Luckily the DRV has exposed pins, so I carefully soldered the DECAY pin (19) to the M2 pin on the board (will be 5V when 32th uStepping is enabled.)
Calibration of your 3d printer is sometimes even more important than building it, and being SCARA, Morgan is not very forgiving in this regard. You have to get it right in order to get something resembling the stl you fed to your favorite gcode slicer…
I don’t like working too hard though, and thus I spent even more time writing up a calibration routine that will make it child’s-play… Well, almost.
Here is the wheel I printed for a small tricycle. The orginal calbration routine zeroed the arm angles to a line drawn on the top platform. While it works well enough, it caused the wheel to be slightly oval. While this does not subtract from it use, it is not quite how it is supposed to be.
In order to fix that, I had to change the way the calibration is done in order to focus on the angle difference in stead of the actual angles. Psi is now calibrated to Theta.
I am going to make some special phantoms in order to make the calibration process a bit smoother. After playing a lot with the steps per angle, I found a way to calculate the steps. The steps combined with the scaling factors makes it possible to get a perfect circle.
After a very long wait, my SCARA RepRAP is finally printing. I printed 4 calibration cubes and a piece of a tricycle wheel (in the wrong colour…)
In the video you can see the new hall-affect end stops, my “Mostly wooden Gregs’ Wades” extruder, the mounted hotbed all working together to print a tricycle wheel.
What you cannot see is the new additions to the firmware that sports:
Quick calibration routines (in Mcodes) Calibration of the arms can be done in as little as 2min without having to calculate the relative arm positions for the arms – for any size machine…
Axis scaling: it is not possible to easily calibrate the x and y print sizes by steps alone, and therefore the machine has internal and Mcode adjustable scaling factors to ensure your prints come out the right size
You can check my Youtube channel (quentinharley) out for some footage…
I have a couple of upgrades to the frame coming, focusing on the z-axis stability without having to add another motor. There might also be an upgrade to the arm configuration in the pipeline soon. I also thought about a practical way to install more than one hotend on the arm – bringing multi-head printing to SCARA.
This weekend will see the pre-flight test and some documentation before the big test… How well does it print!
This week I had limited time, but I completed the firmware for the test, including a very nice electronic calibration section that will allow you to calibrate a Morgan SCARA in around 2minutes. No hardware adjustment needed after the build…
Last night I attempted to write an Auto homing routine, but it seems there are more important things to look at first. The current branch of Marlin that I am on at the moment seem to have lost the ability to check the endstops.
I looked into it briefly, but after comparing my branch with original Marlin, I see that it is time to consolidate some changes. I will have to fork Marlin itself and import the kinematics into it in such a way as to have it functional on cartesian systems as well. This is the only hope we have to get it incorporated into Marlin properly.
The frame is finally starting to take shape. The interesting thing is, that even after all the planning that went into the build, it is still changing subtly as I go along.
In the picture you can see the double arm SCARA, the belt pulleys and the sliders for the z-axis. Still to come:
The build plate, top mounted z-drive, the motor mountings for the two arm motors, and the Bowden extruder.
The movement is smooth, and it looks very promising indeed!
The boards are ready and I am picking it up tomorrow evening after work.
Electronics wise my next task will be to inspect the boards, and get all the required components. My Allegro chips should arrive next week Thursday. If all goes well, next weekend I will be making my RAMPS!
I have also put together an experimental hinged XY carraige, at least in OpenSCAD, and will hopefully get one of our local 3D prototyping houses to make me a couple for testing.
I know I am being very cagy about the new concept, but it will be well worth it in the end!
I have sent off the electronics to the circuit board people, and put in an order for some A4988 chips. Had to get 25, so I’ll have enough to help out a couple of local Morgan builders eventually! In order to save some time and effort, I decided to go with the tried and tested RAMPS 1.4 platform, especially since I already have an Arduino Mega board.
The main driver for eventually going with RAMPS is the fact that the flavour of the day in terms of RepRAP firmware changed about as many times as I changed my mind on the best mechanical platform for Morgan. As much as I would like to create and maintain my own firmware, it would be unwise to exclude Morgan from the rapid development happening in that field at the moment. This will allow me to make minor tweaks to the firmware in a patch form, to add functionality to the system.
The build on a mechanical front is finally also starting up. The RepStrap will be mainly built from PVC piping, in order to start testing the concept. It will be one of the goals to use as little as possible non-printable components in the build.