Now that I’ve had my 3D printer for 2 1/2 weeks, I thought I’d share a few things I’ve learned (in no particular order):
I wish I’d paid more attention in geometry! I jumped into the deep end and started designing my own stuff. That requires some math & geometry (neither are particularly strong suits for me). Now, don’t let that scare you – it’s not that hard, and today Mr. Google is there to help.
Frankly, it’s been about what I expected. The printer itself has just worked. I haven’t had any issues where I wondered whether or not the printer was doing what it’s supposed to do. On the other hand, there is lots of futzing to do – lots of little lessons to find (lots of reading on 3D / CAD forums); some obscure tips & tricks. Fortunately, I came across Lon, who I consider a mentor, who has been INCREDIBLY helpful. He has several years of this under his belt and shares things that help, when I need to know them. If you’re thinking about getting one of these printers, and all you’re going to do is print objects from Thingiverse, you may not have as many of these trials & tribulations.
It’s fun to gauge people’s perceptions of 3D printing. As expected, most don’t think it’s in the home user’s domain yet. Others, who are kind of like me, are at the tipping point – about ready to make a decision to get one. The absolute best way to get someone’s imagination flowing is to hand them an object printed on a 3D printer. Amazement always follows.
It’s not as expensive as you think! I was conversing with a friend who lives across the country and who follows me on FaceBook. She said something like ‘…but it’s so expensive!’ (and she was thinking like $10k kind of expensive!). No – it’s about 10% of that! Still, not cheap – but what did a good laser printer cost 10 years ago? What does a good, fast computer cost today? They’e in the realm of reality if you do your homework. But, you can also pay lots more than I did for one. Shop wisely.
I’ve lost lots of sleep. I want to perfect the things I’m designing, and I want them to be right – right now. Between drawing the parts up in CAD and printing them, it takes time and commitment. For better or worse, I’m the type that digs in and won’t let go until I’ve got something solved… and that takes time. For the past two weeks, I’ve spent about 2 hours in the morning before work and 4 or 5 hours in the evenings on 3D printing related things. And, when printing overnight (which is about nightly), I’ll get up in the wee hours and kick off another copy or another design. I’m at the age where I need to get up a couple of times a night, so that just works out if timed right :).
The printing itself is slow. Just now, I printed a small piece that’s a proof of concept for a larger model (one functional part that will be replicated out in a pattern). It took 9 minutes to print. But, in final form, the whole piece will take 3.5 hours to print. Some things I’ve made take over 7 hours to print. You’ll need to schedule your printing / designing time wisely.
For the above reasons, I need to think things through before printing. “Measure twice; print once“. I’ll learn that someday. I’m not the most patient person in the world, and instead of just printing & seeing how it comes out, I need to be a little more thoughtful about whether I’ve double checked all of the small details, because they matter.
It helps to be handy, but you don’t have to be a master craftsman either. Pretty simple stuff, really – like going to Home Depot for plexiglass & putting holes in it (used to make side covers to control the temperature); mixing up acetone & ABS to make ABS slurry; measuring with calipers.
Organization is hugely important! This applies to physical stuff (components, pieces & parts) but maybe more importantly on the computer. I’ve generated hundreds of files between the CAD drawings and the sliced model files; many of them are different revisions of the same pieces. Having a structure & file naming scheme is important, otherwise I’d be lost trying to remember which one worked the way I intended it.
While it’s not quite ready for every household, I think it’s far enough along for more than people realize. They just don’t know it yet… Soon. Very soon.
I now have a full size, working copy of the core shaker. I tweaked the design last night based on the ideas of a friend (the sides or ‘fence’ is permanently attached to the top tray instead of the original design, which was a box that both trays sat in.
I tested it, and it works! There are still a few kinks to work out (the biggest one is that the cores are catching on the edges of the jackets). But at least I know I won’t be manually placing a core into each jacket when my final set of swaging does arrives!
Kind of like building blocks, I’m taking what I learn and then adding to it. In this case, the original design for this core swager was 100 holes. Because of the warping problem, I divided the part into quarters and printed them separately. I still had some warping, but after printing the first few my ‘mentor’ suggested something that helped a ton! Since I knew the basic concept worked, and while I worked on solving the warping / shrinking problem, I printed enough pieces to make a whole top piece. Next step – combine them.
When I brought up glue, my ‘mentor’ seemed to scoff (if you can infer emotion in an e-mail without emoticons). Not glue, ABS slurry (pieces of ABS melted in acetone). Easy enough and I certainly have enough scrap ABS built up! While I don’t know the strength of the joint yet (it’s still drying), this is a common practice for 3D printers.
The other obvious takeaway here is that 3D printing isn’t limited to objects that are the size of the build plate. Pieces can be printed and bonded as necessary to create a much larger object. Even a car! Check this out.
The pieces that make up this top plate of the core shaker are a hodgepodge of proofs of concept; the amount of warping on the bottom varies as I worked out the kinks. The white one? Yeah, that’s what happens when you’re tired & trying to get a print started before going to bed. I told it to use the wrong extruder (…that was loaded with white ABS instead of black). Oh well – great visual reminder for me to check all of the details twice (or, from an earlier post, ‘measure twice, print once’. Yes, sometimes I’m a slow learner.
After several setbacks, I have a working proof of concept for a core shaker, adapted from a design from a member over on CastBoolits! It’s not perfect, but it works. But, there are still several problems to solve:
I’m still fighting some lifting / warping on the bottom, even though this print is much smaller. However, even though the bottom is a little bowed, it will still work. It’s not as pretty as it could be though.
My hole sizes may be a little large. I’m finding that with ABS, holes (in particular) tend to shrink from the size they’re drawn at. So, after my first prints, I had to enlarge the holes but I didn’t know for sure how large to go. After a few trials, I went with a 7.4mm (.291″) on the drawing. That translated to a 6.6mm (.259″) hole in the plastic. That works for the jackets, but the cores tend to hang up on the edges of the jackets as they’re going in the holes. This may also be solved by making one of the trays a bit higher (so the core is at a steeper angle falling into the hole and less prone to catch on the lip of the jacket). Or, a combination of both.
Combining pieces together. As drawn, I included a small cut / relief on the base edges thinking they’d align well and be a place for the glue to go. But, with the warping, those edges may need to be at the top, and I’ll glue these together sitting face down. These pieces will make a good test for that.
Layer alignment. Right now, nothing is implemented to align the top & bottom layers. I’m thinking of putting little nubs on the bottom layer that would align with small divots on the top layer.
The fence. This part (should) be simple :). This should just amount to a box. But, given the warping / lifting, I anticipate the size of the completed box is going to cause some problems.
So, in case I jumped into the weeds and the concept of this ‘core shaker’ doesn’t make sense, here’s a brief explanation. The cores need to go into the jackets; doing it manually takes time. After stacking the black tray on the white tray & surrounding both with a fence (the index cards for now), empty jackets are poured onto the top of the black tray. Shaking the whole thing around causes the jackets to tip & fall heavy-end first into the chamfered holes. Then, cores are poured on top of the black plate and shaking again causes the cores to drop down the holes and into the jackets. Lift off the top plate and you have them all sitting up, ready to be picked up individually for the core seating die.
This one is a bit of a challenge… After learning more about how to control some of the variables, I’m trying again. All it needs to do is stick to the plate & not warp!
Update: Unfortunately, after approximately 6 hours of printing, I came to the realization that this print was doomed. Although not as severe as before, two of the corners started to lift & warp. Because of the tolerances required of this part, this would impair its function.
Back to the drawing board – the next approach will be to design this part in separate pieces, which I’ll glue together after they’re printed. So, the entire assembly will require 9 pieces – 4 for each section of the two trays and 1 for the case that will hold the two trays.
Again, a simple design to solve a basic problem. I have dies that I use often enough that I want them handy, but I’d rather not dig through a bunch of boxes. I don’t use these on a progressive press, so they don’t sit in a shell plate. So, I made up a die holder:
Handy for either the top of the bench / desk, or in a drawer. Problem solved!
After a week of travel, I came home to a new shipment of plastic, including a roll of lime green ABS. Lime green just so happens to be my wife’s favorite color, so the next thing out of the printer was a pencil cup for her desk. She’ll wake up to see it this morning. I think she’s having some trouble seeing the practical application for this printer, so perhaps this will help :).
Pencil cup; downloaded from Thingiverse & printed in green
This was treated in acetone vapor for the shiny, faux injection molded look.
My world includes a full-time career that has its own demands for my time. Notably, it pays the bills :). This week, I have limited time and access to my tools… so take this opportunity to read some of the material I’ve updated in the ‘Swaging‘ section of ‘Making Your Own Bullets‘.
I’ve added quite a bit of content to the ‘Swaging Equipment‘, ‘Cleaning Brass for Jackets‘ and ‘Annealing Jackets‘ subsections. In these sections, I’m laying out the choices I’ve made as a new bullet swager, and I’m starting to detail the processes I’m using. If swaging interests you, this is a great starting point to begin understanding what you’d need to do to create your own projectiles.
Another milestone today. Overnight, my ‘catch tray’ finished printing. In the thread over at Castboolits, a member that I’d already printed & boxed up an RCBS decap tray for expressed interest in my design. So, since I’m still cautiously jumping into this, he is my first official beta tester.
For those not familiar with swaging bullets, here’s a video showing cores being swaged to weight. They’re falling out into my new catch tray.
The lead cores you see were originally stick-on wheel weights (almost pure lead). They were melted and cast into rough cores using a purpose-built mold. These cores are what’s inside the jacket of a bullet. The rough cores are swaged to a specific weight (the weight of the cores + the weight of the jacket = the finished weight of the bullet). Consistency in weight is very important – these will become 55 grain FMJ bullets when they’re done.
When the cores are swaged, they are compressed under tremendous pressure, forcing the extra lead out through a hole in the side of the die (and into the small cup at the top of the press in the video). Then, when the handle of the press is lowered, the core is ejected and it drops into the catch tray.