Some lessons from my experiments with 3D printing
3D printing is a lot of trial and error, I think anyone who gets into the hobby is mentally prepared for that at minimum or most likely just finds it to be fun. But there’s no sense in torturing yourself to re-learn some of the same lessons that I have (painstakingly) figured out after about ~3 weeks of effort. Or trawling through zillions of pages of only semi-relevant Reddit discourse to find a few nuggets of good information. Here are a few of the things I figured out after toying with this.
Glue sticks and painter’s tape - a must. If you are struggling to get your first prints to stick to the print bed this is an easy way of getting to the point where you can troubleshoot some of the deeper issues that may be plaguing your prints - basically glue them on until you can figure out what the issue is. (You will almost certainly have this problem with the cheap printers that are popular for home use.)
Not everything is calibrated correctly - don’t be afraid to raise temperatures. You may read somewhere that the ideal temperature for PLA is 195, 202, etc. I never had much luck using those low ideal/”default” temperatures and I suspect that is because the temperatures are not calibrated that well or consistent across the entire hardware box. Feel free to jump up to 210, 215 for the nozzle and 60-65 for the print bed to get stuff to actually work. It has to sort of melt into the print bed or you’re not going to have any luck because the print will just slide and drag under the nozzle. Oh and double check Celcius/Fahrenheit, on everything. Celcius seems to be used the most often but that doesn’t mean everything you encounter will be in Celcius.
Rafts/supports. Cura has some pretty handy tools that can automatically generate this stuff. You don’t have to jump all the way back to a CAD program, or download new .stl files, if the supports are missing or not the way you want them. And the preview mode (before saving to the micro-SD card) will show you a rough idea of what the supports will print like, so you can decide if you think it’s enough. The goal is to hold the print down to the bed - at any cost. At first while you are still getting used to doing all of this tweaking and getting things set up correctly you may find yourself printing supports that take 30-45 mins before it even runs the item, and that’s OK. If it’s what it takes to save the rest of a 2-hour print job, you kind of just have to eat that time and plastic.
Leveling - the print bed probably has to be closer to the nozzle than you will be comfortable with right away. I still feel weird with the bed jammed right up to the nozzle, kind of at almost the full extent of where it can even go in the limits of the hardware. But it seems like that has resulted in the most success. Don’t be too risk aversive with this, if you get a bit of grinding or damage to the plate just stop and dial it back a bit and it will be fine!
People download a lot more .STL files than they actually use. If you go on popular sites like Thingiverse, Cults3D, etc. you will see lots of designs that have a really high view/download count and then no actual makes posted. Don’t start with those. You probably won’t have much luck unless maybe you already do advanced manufacturing at your job. It seems like people are overoptimistic about how much they can or will print and download a lot more models than they ultimately run in a real printer and that can skew the metrics - 2,000 downloads is a lot, but if it has 0 prints, you may not be the lucky first. Start with the stuff that has 20,000 downloads and 30 successful prints because you can more reliably troubleshoot issues if you know for sure that the file is not the problem.
The file is probably not the problem on popular designs. Check your settings in Cura before it generates the gcode… success or failure is mostly determined by about the first 12% of the print, which mostly comes from the profile settings. Raft/skirt/brim, infill, supports, temperatures, print speed, all that stuff. Oh and you can save these profiles and tweak them. So if you do prints that have some similarities, you can share the profiles across multiple prints with just minor tweaks. But be sure to constantly adapt to the design you are doing or the likely result is stringy plastic blobs on the print bed.
When in doubt, slow down. Yeah there are printers that can do 100mm/s but unless you have one of those, slow down! There is no shame in printing at 11mm/s or 20mm/s if that is what it takes to get a print job that actually runs to completion. The default settings are often going to be set at 45 or 50 mm/s even on very cheap printers just meant for home use, but that is too fast for what the hardware is actually capable of doing IMHO. Slower printing than the default will usually result in a lot less mess and fewer failed prints.
