Morgan 3D printer adventure!
September 12, 2013
Thanks to Ossipee and his jedi mind tricks I have started working on another RepRap. My first RepRap was a Rostock which was on the cutting edge at the time and now I'm building another rather immature model. I have Ossipee to thank for the ongoing Rostock adventure as well!
Since the Morgan is so new, documentation is still in a state of flux with quite a few people documenting various virtual and real-world builds. While this is good it also has a way of muddying the already murky RepRap waters...
The creators site is here: http://reprap.harleystudio.co.za/ and there is a lot of info along with a message board there. The creator, Quinton Harley is a frequent contributor to the forums.
Even though the Morgan is a close cousin to the Rostock you'd be hard-pressed to see any family resemblence. In fact, the similarities are in the inverse kinematics used to run Morgan by positioning the SCARA arms in a similar fasion to Johann Rocholls delta code which allows the processor to do most of the heavy lifting.
While the Morgan assembly is pretty straight forward it also has some hardware quirks which I haven't fully explored yet. The BOM calls for .9 degree stepper motors and non-typical drivers for actuating the arms plus it uses Hall effect sensors with magnets for the end stops. The rest of the build seems pretty straight forward with RAMPS 1.4 and regular 1.8 degree steppers for the Z axis and extruder. I'll update this info as I get closer to that portion of the build.
The first phase began with Printing the parts:
As you can see, Morgans parts are rather big! I had to do quite a bit of fidgeting to make them fit the build area of the Rostock.
After printing all the parts I turned to the frame assembly:
There are some templates posted on the Morgan forums that are supposed to be printed out in 4 pieces and taped together. I couldn't get it to work but some have had good luck with it. I finally just took dimentions from the printed parts and sawed some wood \o/ I really love the economy of design that uses the semicircle "waste" as the top plate!
Another quirk of the Morgan is the use of copper pipe for a lot of the construction. It calls for metric variants which aren't available in the US so I used 9/16" for the 22mm and 1/2" for the 15mm. K120 modified the files for SAE tube, which is posted on the Morgan forum' but I had already printed the metric parts.
I used a homemade reamer to open the parts up to size:
Q.Harley actually recommends reaming the parts using a Uni-Bit because they need to fit well.
The basic frame:
I used 1" ID thick walled PVC for the supports. The Metric parts fit fine after I ground the "locking ring" off the end caps. The PVC tubes are installed under tension. The outer tubes actually curve around the 2 inside pipes so when the assembly is unscrewed they spring apart. This is part of the original design and it's surprisingly rigid!
The mechanicals of Morgan aren't really as intuitive as those of the Rostock or cartesian bots. The main functional unit of Morgan which drives the arms consists of 2 concentric copper tubes which have arms on one end and the drive wheels on the other.
The center of the assembly consists of an 8mm threaded rod with a 15mm copper tube providing torsional support:
The printed drive wheel has a hex for the ID which nuts fit into and the end of the 1/2" copper has a 5/16" nut pressed into it.The 5/16" threaded rod is at the center of the assembly in above pic. I used 5/16" threaded rod instead of the 8mm because it is readily available in the states and so far it seems like it will work fine. Q.Harley recommends wrapping the threaded rod with teflon tape where the bearings ride on it.
The flanged bearing on the left goes into a bracket on the bottom platform and the bearing on the right fits into the other printed drive wheel with the 9/16" copper pipe like this:
Please don't use these pics as assembly instructions because they were taken while mocking up the parts!!
While I'm often not very adventurous about modifying designs that have been proven to work I decided to go out on a limb here because the original Morgan plans call for a metric masonry bit for the Z lead screw. I had trouble finding one locally and I'm wary of ordering the wrong part so I'm going with a "threadless ball screw" design I found on thingiverse:
Hopefully it will work but if not I'll probably just go with threaded rod. It's immediately apparent that some pretty hard rod is needed for this rig because the bearings have already scored up the un-hardened 8mm rod I'm testing with.
Here's a pic of the Z axis stuff with the concentric copper pipe assembly:
The top bracket is to the left, mounted to the top plate. Near the top of the copper tube is the sleeve that a 6805 bearing rides on and fits into the top bracket. The BOM calls for only 1- 6805 bearing but the machine requires 2. One goes on the 9/16" copper pipe and the other one mates to the tool holder for the hot end.