Here’s a couple of prints that I ran after completing calibration. The butterfly warpped a bit, causing things to shift and during the print. Time for a headed build platform some more skeinforge tweaking.
So I spent a good portion of this past weekend calibrating my MakerBot. I might have been a bit obsessive, but now I have a much greater understanding of my ‘bot and skeinforge. After settling on a version of skeinforge to use ( I choose version 0007 ), I began running a bunch of test prints. These were done using a 0.5mm nozzle which extrudes unstretched ABS at approximately a 0.56mm diameter.
Using this value, I set skeinforge’s diameter over thickness ratio to 1.62. (note: 4/21/2010: I now think that this ratio shouldn’t be on the unstretched extrusion diameter. Instead I think it should be based on the stretched, but unsquished extrusion diameter which will be based on flow and feed rate )
The other choices I needed to make before calibration was the print layer thickness as well as feed and flow rates. From looking at a chart I found that showed some theoretical values for extrusion ( note: I believe this was for a rapman machine ), I decided on a 0.35mm layer thickness and a 0.63mm perimeter thickness. To achieve this I printed out a single wall calibration piece tuned my feed and a flow rates a bit to 24.0 mm/s and 240 pwm.
I then set out to adjust the perimeter width over thickness and infill width over thickness ratios. I was looking to get a relatively smooth, solid surface. I wanted to remove the gaps between the infill and avoid ridges. After trying various values haphazardly, I decided a more systematic approach was needed to narrow in on the correct settings. I started by annotating every print and only changing one variable at a time. Doing so allowed me to see subtle differences between each setting until I was happy with the output. I settled on a ratio of 1.64 for both.
Things i learned:
- Getting the raft to stick. I was able to get my raft to stick consistently while printing with little manual intervention. I accomplished this by adjusting raft’s ‘time before raft setting’. Having the extruder moving in the horizontal plane as it starts extruding the raft seemed to significantly help.
- 2 layer objects are not a good test. You really want 3 or more layer. This is because the raft thickness has significant impact on the first layer and can cause great variation in 2 layer prints even with the same settings. ( the more squished the raft, the better the print, but also caused extruder jams. I’m looking forward to raft-less printing )
- Sometimes it’s the software. I was having a problem getting my infill to touch all 4 perimeter walls. I kept changing the infill overlap ratio, but never achieved success. Turns out there was a infill bug in version 0007 of skein forge. Once I fixed the bug, I had my infill consistently touching the perimeter.
- Complex objects such as the butterfly need a slower extrusion or higher feed rate. Too much filament was being deposited on the perimeter as the extruder zig-zagged over short distances. This caused problems and blobs to form on subsequent layers.
My take on the ratios. Diameter over thickness is a constant based on nozzle size, feed rate, and flow rate. This value should be used as a minimum for the other two ratios as you can’t really extrude thinner than this value while squishing it on a previous layer ( unless you are really stretching out the material ).
Next up, upgrade to a new version of skeinforge and make sure I can get the same output.
I also need to assemble my heated build platform as warping sucks and causes lots o’ problems.
I’ve had moderate success printing some test objects and getting the raft to stick. I could always tell if it was going to work by whether or not the first segment adhered. If it didn’t the part that was sticking up would usually get knocked around by the print nozzle and eventual cause problems with the print. I recently switched to using blue painters tape on the acrylic build platform. This seems to have a bit better adhesion.
The big winner for getting the raft to stick a comment in Configuring Skeinforge wiki article. Manually adjusting the z-axis by pulling on the belt solved my raft issues. Just don’t squish the raft to much or you run the risk of causing too much back pressure in the extruder and causing melting ABS to find another way out.
In the future, I’m going to be looking into a heated build platform with the hopes of going raft-less.
After a bit of snowboarding, I’m back and ready to complete the Plastruder MK4. Here’s some photos of the extruder assembly and first extrusion using white ABS. I also choose to mount the extruder circuit board off to the side with only 2 screws. This allows my to view the movement of the filament while still allowing clearance for the z-axis.
My MakerBot #545 of Batch 10 showed up in early Feb. After oiling the lasercut plywood with some Watco Danish oil, assembly began. Here are some pictures.
So I’ve been working in 3D printing a few years now with FigurePrints, but my experience has mostly been one the software side of things as the actual 3D printers reside in Vancouver, B.C.
Not any more! Now I’ll actually be getting my hands dirty. I decided to purchase a MakerBot CNC back in December. I won’t be doing anything FigurePrints related on it, but I’m sure I’ll have days of fun with it.
