RepRap Prusa i2

From BloomingLabs
Revision as of 20:12, 21 March 2013 by Josh (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

I started my building of a RepRap after much research of which machine to build. About 6 months ago, I finally decided to build a Prusa i2. The design was stable (for an open source project), and many people knew the failure modes in which the machine could be subjected to. When I started, I wanted the knowledge and understanding of building this for myself. Many a times, when people buy 3d printer kits, things go wrong, and not always how to fix them.

I started, with the first question: which printer?

  • Prusa i2
  • MendelMax - $150 more than Prusa i2
  • Rostock - Delta printer, at the time had unstable firmware
  • Wallace - I wanted a bigger build envelope

There were a few more, but what stuck to my budget was the trusty design by Josef Prusa. I am revisiting in building a Rostick, as the firmware for handling a non-Cartesian printer is much more stable.

The next question: Metric or SAE? To me, this wasn't a question: metric is extensively used in science, programming and open source projects. The SAE version has problems too, regarding the Z axis. I will touch on this later. The easy answer was Metric.

Now, when you buy parts for a RepRap, the frame pieces have been printed by another printer. So, the innovative way to work with this is to head to eBay or to irc.freenode.net #reprap and buy a kit for the printer you're building... But not that simple! Fun questions arose when asked about the extruder and the LM8UU's... huh?

The LM8UU is a linear bearing that interfaces the smooth rod stock (for movement across the axes). As with any open source project, improvements are constantly being made. This improvement was switching from printed bearings made of plastic, to proper ball bearing enclosed cylinders of steel. The different printed part is a part that holds the LM8UU and uses a ziptie to hold the bearing in place. They added about $10 to the cost, yet provide much better and smoother motion.

The other choice I made was about the extruder, or what we call the Cold End. There's quite a few different types. Many of them vary from the standard called the Wade Extruder. It contains a frame that holds 2 608 skate bearings, a hobbed bolt for grabbing filament, a spring tensioner system for holding filament, a stepper motor, and reduction gears to drive the hobbed bolt. The kit I bought had Greg Frost's Wade Extruder.

After ordering plastic parts, I then went to order my rod stock I needed. I had no easy way to cut them, so I invested in a $7 hacksaw and sized and cut them by hand. The thread pitch on M8 thread rod is 1.25mm, so I could calculate exactly where I needed to cut.

I ordered my electronics package from eBay user RepRapDiscount. She's one of the best in the world for a full electronics package (aside from steppers). Her package also included a heated bed place, which is essential for working with ABS, nylon, and polycarbonate.

I also ordered my hotend (where the plastic is heated and comes out) from IRC user thejollygrimreap. His hotend, unlike many on the market, is made completely of metal. While no plastic needs super hot temperatures, this hotend can reach up to 575C. Most other hotends contain PFTE, which is also known as Teflon. Teflon turns to a liquid-glass state around 250C and causes a printhead jam and failure.

I attended the MRRF and got my last few parts on my printer assembled. I spend quite a while calibrating the E Steps, which translates motor motion to exact measurements. So if I print a 25mm square, it actually prints 25mm by 25mm. I got calibration done. And I start trying to print, and nothing...

  • 3/20

I end up aimlessly trying to figure out what's wrong. I pack away from MRRF, and head back to Bloomington. I get on IRC, and JollyGrimReap is there. He helps me by discussing PID tuning. PID controls are for adjusting when to turn on and off to maintain a constant heat. It stands for proportional, integral, derivative. I tune my PID, with a GCode M303 S150 and off it goes. In console, after 10-ish minutes, it spits back the numbers for my hotend, which are drastically different than the "standard". I put them in, and... and.. HOTEND JAM. What the hell?! Back to IRC. While chatting, I'm removing the jam manually with toothpicks and gently using a screwdriver bit to pick out bits. I had raised the temp to make the print easier, but succeeded in burning the plastic in the hotend. Bummer.

Back in IRC land, I explain my new predicament. The answer was announced by multiple people: I need a fan on the upper half of my hotend. What's happening is that heat isnt staying down in the tip area, but creeping up and getting just under melting point. When it does this, it solidified into a plug. So, I head off to the Bloominglabs junk room and find a spiffy old CPU fan and attach it to my hotend with zipties. I hook it up to 12v main, so it stays on all the time. I have to recalibrate the PID this time, with a fan present. I start.. and first layer is a SUCCESS!........wha?

And then, I notice that it quit printing.. It turns out, when JollyGrimReap made the hotends, he used high heat threadtape so the threads wouldnt move under tension and movement. When I burnt out the plastic, I also burnt the threadtape (I thought it was more plastic bits). I failed to realize its purpose. When printing, the hotend threads turned ever-so-slightly, and caused a bigger liquid chamber, causing print error.

Printer printing first layer.

Next - I need to buy some high heat thread tape, clean out the end again (this time NOT burnt in), and properly install it in the threading. I'm so close.

Up next: This printhead was equipped with a .5 mm tip, using a standard M6 thread attachment. I have ordered a .35mm and a .4mm tip to allow finer printing. I have also bought from an aluminum caster a Cast Wade Extruder. You can see that here

Personal tools