Intro: Rapid Prototyping the Old Fashioned Way (Sheet Metal + Solder)
You want to make cool things? Cheap? Fast? You need some way to actually do the building. No point just coming up with ideas eh? This instructable presents a way to make small items faster than a 3D printer. It does take a little more effort though, and you can't produce quite as complex parts.
This piece measures 5cmx3.6cm. A ballpark 3D print time is about 3 hours. Using sheet steel I made it in just under an hour! It's also far more durable than a 3D print.
Last year was a bad year for me "because I need a 3D printer to make cool stuff." This year I realized my grandparents did cool stuff and they didn't have a 3D printer. After sitting down and thinking about it, I realized they made things out of two primary materials:
Surely I, in the modern era, could do the same?
With a soldering iron, and a empty can of tomatoes, I decided to figure out how this soldering business was done. Turns out, it's not that hard.
Time: 1-2 hours
If you like what you see here, please vote for me in the contest
Step 1: Tools and Materials
Chances are you have everything somewhere. If not the tools will only cost $20 from a hardware store, and you'll find plenty of other uses for them....
- Soldering Iron
- Brain (optional)
- Empty and washed tin can(or sheet metal)
- Glue stick
Stanly suggests using corned beef cans, as they don't have any corrugations, and have flat sides.
Step 2: Design and "unfold"
The first step is to take what you are going to make, and turn it into a 2D pattern. There are two ways:
Create a 3D model on a computer, and then use pepkura or blender3D's unfold plugin to make a template. I plan to make an instructable on this soon.
Old Fashioned Way:
Break out the pen and paper and engage your brain. Use some pythagoras, some geometry skills you thought you had forgotten from school and draw it.
Whichever way you chose, make a 2D template and test fold it to make sure it's about right. It's a bother to change things when you've already cut the metal....
- Don't bother with tabs
- Keep curves 'single' you can't do double/compond curves with sheet steel without a hammer
- Keep individual pieces easy to cut out
Step 3: Cut and Paste
Time to cut those templates out of metal. Open a can and flatten it as best you can. Then use a gluestick to stick the paper templates on. The glue won't stick well at all, but it only needs to stick long enough for you to cut the pieces out. Finally, using a pair of scissors, cut the templates out of the can.
- To open a can use a can opener on both ends and use a pair of tin-snips to cut down the seam
- Use a pair of scissors you don't mind damaging. They won't cut paper that well after cutting metal. I have a dedicated pair for thin sheet metal.
- Tin can sheet metal is thin, and while cutting it you can get small, sharp shards. These can be painful
- I do not currently have a good way to flatten the cans. Any ideas are greatly appreciated.
Step 4: Solder It Up
There are several major parts to making a joint:
1: Remove plastic
On both the inside and outside of the can there is a thin layer of plastic. On the inside it is white, on the outside it is clear. Using a knife scratch away the plastic from where you want the joint to be. Because of the surface tension of the solder, you don't have to get rid of all of it. So long as the solder has a place to get to the metal, it will lift the plastic and bond underneath
2: Tin the metal
Run a bead of solder along each side of the join separately. This means that we know the solder will stick everywhere, as well as knowing that the solder will be right in the joint
3: Hold pieces together and solder it
Hold the pieces together, using a clamp if necessary, and solder it! I find the easiest way is to put a blob of solder on the soldering iron and tack the two ends together. Then run the iron along in the corner. Make sure the shape of the solder joint is a nice curve. Blobs mean it hasn't made a good joint.
You can bridge gaps of up to 2-3mm with solder!
- You're working with a soldering iron. I can personally tell you they are quite hot.
Step 5: Repeat As Necessary (Some More Tips)
Yup, you can do them. Simply do them a centimeter or so at a time. Try to make sure the metal is bent into the right shape already so when you do the next cm, the previous lot doesn't flick out. A pair of pliers is really useful.
Because you tin each side of the joint all you need to do to make a good connection is heat it and hold the bits together. It's easiest to heat it by running the iron down the joint, but if you're trying to close a box you can do it from the outside! Solder doesn't tend to stick to the plastic so you can get it quite neat.
It's a tin can, there will be little gaps everywhere. My experience shows that this isn't a problem. You can still make good joints. They even make the structure stronger. That said, if you find a way to flatten out the can properly, let me know.
Bends and warps:
Because the can was originally round, the sheet wants to bend up, and this leads to warps in your constructions. I've found that so long as you have a closed/box section this isn't a problem. With just the three sides (first picture), it didn't want to sit nicely. Upon adding the tube-thing to join the sides it settled down. Did I mention how strong tin can is because of the corrugations? Yeah, it's really tough.
Step 6: Conclusions
Was this successful?
Yup, I discovered that I could make strong structures really quickly and really cheaply. All it took was a little thinking and a few burned fingers.Using this technique I made an air-tight pressure chamber for an electronic wind instrument, a prototype drive mechanism for a small robot, and I plan to use it for a lot more
But there are definitely areas that I need to think about more, mainly: How can I flatten a can better? Get rid of the corrugations and warps? This will mean that things look better, and the material is nicer to work with. Sure, I could just buy thin steel plate but that takes money.
If you have any good ideas for how to extend this method of rapid prototyping, let me know.