Prusa Mendel Update #5: It’s alive & printing!

After a night of fixing various problems, I finally have my Prusa printing objects that actually resemble objects!

It’s no where near calibrated nor perfect but it’s working reliably now that I have the J-Head hotend installed along with Greg’s ‘Guidler’ extruder along with the motor strain relief bracket.

I had to fix a number of alignment issues with my belts, adjust a pulley, reseat and level the print bed to get to this point.

This is a quick printing wire clip which turned out really well considering I haven’t really calibrated anything yet beyond levelling the print bed:

The one on the left is literally the first thing I tried printing (before I fixed a number of issues) and the one on the right is pretty close to what it was supposed to be. It’s a little smooshed as I grabbed it while it was still warm.

Printing in PLA is quite different than the ABS I’m used to…it’s very oozy. I’m going to need to install a fan to cool the print faster.

It’s SOOOO much quieter than my MakerBot!

I’m excited to finally have things up and running. Now I just need to start calibrating the printer.

UPDATE: Received a fresh roll of transparent green PLA and added a temporary light to the extruder. I’d originally tried printing on glass with a cold bed, then a heated bed and then settled on cold blue painters tape.

The first few layers always seem to have issues sticking. Perhaps my Z height needs to be lower (squishier) or it’s a temperature thing. The green PLA seems to be more cotton candy like with stringy micro filaments that get caught under the nozzle and wreak havoc.

The cube on the left was literally the first thing I printed when I installed the new spool of green PLA and turned out nearly perfect. Glassy bottom, not too smooshed 1st layer and the top is nice and smooth. The middle one was a combination of things including a filament feed issue (spool not rolling freely) and the last one was the last thing I printed yesterday on blue painter’s tape. It’s decent but not quite as nice as the 1st cube.

I was tired of printing cubes so I thought I’d try something relatively fast to print, and was a trickier item to print due to the shape of the text. I’ve printed tons of these geeks on my MakerBot so I know how they are supposed to turn out (in ABS at least):

Unfortunately it didn’t finish printing. Just as well too since there were problems with it due to the errant blobs of PLA causing the hotend to skip and the alignment was off on the text layers as it went higher.

It stopped printing because my laptop keeps cutting out (like it’s completely powered off with no warning, kernal panic, just off) randomly while printing (after 7-10 minutes of printing), killing the print. I’m guessing it’s a USB communication error (tried both ports & my USB cable already has ferrite cores on each end of it) or line noise from the wiring. It was much worse when I tried to tidy my wires up by bundling them together but once I undid that, it got better (longer prints) but it still happened with this print. My desktop doesn’t seem to have the same problem though but that’s not nearly as convenient to use as a controller for the printer.

Maybe it’s time to get the SD Ramps board installed so I can print without needing it connected beyond starting the print.

Lastly, I found these RepRap coins which are a nice alternative to the calibration cubes. They print fast, are a little tricky and print well:

UPDATE #2: Last night I finally soldered the pins to the SD Ramps board I had ordered with my electronics originally and the headers to the RAMPS (along with the important D1 diode under the X/Y drivers to allow for computerless operation) in the hopes that this would solve my issues (or make it moot) with my laptop not staying connected:

Julian’s post was very helpful in navigating my exact configuration. This involved disconnecting everything and removing the RAMPS 1.4 board to solder onto it which gave me the opportunity to detangle some of the wires which may have been causing my line noise issues. I also labeled each wire before removing it to make it easier to reconnect.

I had originally planned on wiring the SD Ramps instead of the normal header way but ended up going with the headers as it was easier and a more robust connection that I had originally thought. To my great surprise, upon plugging the SD Ramps into the arduino mega, it worked! I then reconnected the RAMPS and all the wiring and reflashed the arduino with Sprinter with added SD support (by uncommenting one line of the configuration code) and hoped for the best. Booted everything up and again, to my great surprise, Pronterface recognized there was a SD card connected and allowed me to load up some of the gcoded files I already had on it. The hot end heated up and away it went printing. I then disconnected the printer in Pronterface and unplugged the USB cord and the printer kept going. A few minutes later I had my first untethered calibration cube printed…and it was nearly the best one yet! Yay!

I got a little cocky and then reconnected everything again and tried to print the geek keychain. It works flawlessly until about 5 minutes into the print and then some blobby PLA caused it to skip again and I ended up reconnecting to the printer to stop it – unfortunately there is no other way to stop the printer when it’s going by itself other than to kill the power.

Stay tuned as I attempt to solve this problem…my guess is that my extruder calculations in the Sprinter firmware needs to be recalibrated. Brad’s also suggested that I switch to Marlin which I may also do in the coming days.

UPDATE #3: Once again, simply showing a photo of a failed print to my pals in the reprap group helped troubleshoot my printing issues. My set screw on my x-axis needed to be tightened and bumping up the temperture to 190C for the green PLA I was using seemed to have fixed things:

Still not 100% but pretty damn close to the happy place I was hoping to get to!