First 3D printer project upgraded with DuetWifi
-
This is my project. I built it from parts an instructions. It was first built with RAMPS 1.4 electronics from some Chinese firm (whom I wouldn't recommend, really). In the pic, it's printing a Dr Who chess set with some generic PETG filament that… actually I wouldn't particularly recommend, either. This re-build, however, I'm pretty proud of. I'll continue the description south of the pic.
I built the original RepRap Prusia Mendel fairly straight on. Generally, I feel it's best to do the given recipe before trying to riff on it. In the end, I didn't connect the heated bed I had due to a lack of clear instructions and I ended up testing it successfully with some PLA.
When I tried, PETG, however, I wasn't having as much success. I had been using painter's tape. I had also bought purple gluestick and purple hair spray, but I detest gunk, and hadn't used them. I tried all kinds of slic3r modifications to make things better, but in-the-end, it wasn't working well. I could also accept that this paricular PETG filament is persnickety… and I have some HD glass to try later.
My first upgrade decision was to get the heated bed working and to try "LokBuild". In reading about the LokBuild, most said that a heated bed was key. Along the way to this purchase, I discovered the BuildTak system. I wasn't that interested in the surface (there were comments for and against it for PETG) ... but I was very interested in magnetic build plate upgrade. In speaking to the people at BulidTak, they recomended I not put their magnetic system directly on top of the PCB bed heater --- too much flex.
So ... I determined to upgrade my header from a Mk2 to a Mk3 Aluminium. Now my bed was going to be a board (thought about upgrading it, but couldn't come up with a reason) with screws and springs with a Mk3 Aluminium heat plate followed by the BuidTak magnetic base, a BuildTak flex plate and then LokBuild suck to the top of that. I did buy an "extra" flexplate from BuildTak to try out their surface, too. It sounds like it's great for many other plastics ... so I'll have that to try later.
After looking at all this, I was also worried about the PETG I had and the extrusion temperatures I was facing. I had a J-head direct from Hotends.com ... and if you're not going above ~230C, I can recommend it. It worked very well and gave no problems. I, however, was interested in 240C and beyond ... and frankly, although the "limit" of the J-head is 247C, even 240C scares me. It was probably my own fault, but on warm-up my hot-end was overshooting by 3-ish-C ... and this was all a bit too close for comfort.
Everything I wanted seemed to point to the e3d v6 or e3d lite. I could have gone with the lite, but the price for the full e3d v6 is very reasonable... so I picked that one. it was quite a bit of work to jam in, and in-the-end I had to angle the cooling fan a bit... but it's in.
I thought about what I wanted at this point. The whole printer cost about $150 in parts + $75 in printed parts originally. These upgrades alone came to about $200. In retrospect, however, good investment. Before the upgrades, PLA performance was very good... but PETG performance was not.
Lastly, I thought long and hard about my electronics. The RAMPS 1.4 hardware works and his amazingly cheap. The stuff I got from China wasn't a sparkling example of RAMPS 1.4, but it was cheap and it did work. If I was going to upgrade my electronics, I had my eye on the Duet ... but the price ... the price.
The last straw was deciding to go with the IR bed leveller. It hasn't arrived yet, ironically, and I also might not need it as a matter of course, but I wanted it and I was convinced that the Duet hardware would run it better than the RAMPS could.
But, TBH, having installed the DuetWifi (and not the IR leveller), I am very impressed. The motors run cooler. The motors run quieter. I miss some of the cool features of Octoprint (power calculations, etc), but the quietness alone is compensation.
In-the-end, I'm happy. the built volume is 200x200x100 (which is smaller than spec for a few reasons). There's also (as I posted elsewhere) a 30x50x* volume that the head cannot go. I'd like a way to express that in the firmware.
I plan to next experiment with the snappy next.. but that's only one of a host of projects. I may build a printer with more Z if it bothers me.
