I'm looking for suggestions of something cheap/household-common that already has a motorized moving head (with about 1 mm resolution), or alternately a motorized moving surface, which I could re-use -- just so that I don't have to set up the entire mechanical portion (rails, stepper motors, etc.) for my project from scratch!
The goal is this: I am tinkering with a project that involves several hundred points where I need to drill holes into an object. The holes will be located exactly 1 mm apart.
I want to automate this process using my Arduino to control the timing, iteration, etc.
This could be accomplished by either of the following
--Option A: Move object by 1 mm, Drill hole, Retract drill, Move object by 1 mm, and repeat.
--Option B: Keep object stationary, Move drill by 1 mm, Drill hole, Retract drill, Move drill by 1 mm, and repeat.
Do you have any ideas for some household or purchaseable item that already exists which I could hack or re-purpose into what I want above? Anything below US $300 would work.
PS: My first thought was Disk drives, but those would be too small.
Not exactly "household", but if you can get hold of a microscope stage mechanism, and drive it with a stepper, you could probably get sub-millimeter movement.
cr0sh:
Not exactly "household", but if you can get hold of a microscope stage mechanism, and drive it with a stepper, you could probably get sub-millimeter movement.
@cr0sh: Seems very fitting but they are extremely expensive; haven't found one even remotely close to $300
My XY table made with Home Depot ball-bearing kitchen drawer slides, two dollar steppers, and threaded rod. Made for about twenty dollars. Now admittedly this is a total hackjob, but the design works. Those steppers are actually geared down to over 2000 steps per rotation, with one rotation moving the carriage a tiny amount.
Make a slightly beefier version, less epoxy and better steppers, and you will be in business. 1mm resolution shouldn't be TOO hard to achieve. This has been sitting on a shelf waiting to add a third axis and be a CNC router for a few months..the whole project is supposed to be making a CNC router for under $100...
Terry- Yep, it's those AirConditioner steppers. Love them, so much torque it's crazy. I've not ordered from your store, but have browsed a number of times and seen you carry them along with the ULN2003A darlington driver boards for a couple of bucks. Great item, everyone should have a handful of them. I have also mounted wheels on two and a caster for a third wheel and made a bot with them, and if you remember my Frankenbotic Thinganator (the wall plotter made of salvaged stepper motors and fishing line) I went back and replaced the original steppers with those little beasts, but I haven't redone the code yet. Suffice to say, the resolution on the Thinganator is going through the roof... the video below is the original Thinganator, somewhere around here I have a couple of photos of it writing "Arduino Rocks" in a couple of fonts. Taking out radial distortion was very hard and done with recursive approximation.. the new version will implement some newer, faster code which is based on Heron's extrapolation of Pythagorean theory, calculation of the height of a triangle (y coordinate) by the length of the sides/tensors (my fishing lines form two sides and the beam is the third) and then calculating the x coordinate from the y. It actually made sense once I started thinking of it as a tensor/triangle problem-- the inverted triangle, no matter the angles, is composed of two right triangles of equal height, gravity being the third tensor. MUCH more accurate and fast than recursive approximation. I've also been thinking about a tweak to Bresenham's line algorithm (which I already redid for this as unidirectional while Bresenham's is arbitrary) that ought to make location calculation a bit easier too, but that's for another time...
As for coupling on the XY table, that's a 1" coupling nut, filled in with epoxy and jammed onto the motor halfmoon shaft, then caked in more (stick this time, not liquid) epoxy and secured with drywall screws. If my projects don't give actual engineers the heebie jeebies, then they obviously haven't looked too closely.
I'll MacGuyver anything, badly. If I thought it were even remotely possible, I'd probably somehow make my own transistors out of sand with a bic lighter, just to be a moron about it In reality, I have more than once thought about throwing you guys here on the site a curveball and make something robotic but instead of transistors, how about vaccum tube-- or better yet, home-made triodes. Those can be made fairly easily.... just need a torch, some nitrogen, and some funky bits of wire, as far as I can tell...
I'll be definitely be showing the Frankenbotic Thinganator clip next week when MaryAlice and I do a 4-day Arduino workshop for 15 HighSchool kids!
I love the whole thing of making stuff out of things that were originally made for a different purpose. Oh, that's now formally called "RePurposing"..
Like your Stepper Motor - Leadscrew Coupling. Years ago I was fortunate to meet Buckminister Fuller, and he said something that I have applied ever since to Electronics and Computers: "Dare To Be Naive !"
It definitely applies to Mechanics. In your case 8)
These days you are a hero for dumpster diving for parts. It wasn't that long ago that being a tech packrat got those "hoarder" looks from folks.
Aha, here's one, but before the code to take out that radial distortion. The source "image" is g-code (which is remarkably like Postscript) produced as a CNC toolpath by DeskEngrave. I simply screen out the z axis.. with only two motors, there's no way to lift the pen.
I make a habit of doing things that aren't supposed to be done. Eventually, I have the idea of scaling this (as it really only depends upon the string length) and putting together a more interesting "print head"- I want to attach a solenoid or cam... and use cans of spraypaint. Solenoid pushes spray button, or turning cam does. The Airwick Air Fresheners use the cam method, and the hardware looks like it would be easily harvested....
You could graffiti entire sides of buildings.... with no gantry (being a suspension plotter), scaling is just putting the motors further apart and more string, and a few lines of code... fact is, it isn't outlandish to think that with a little better construction, something like this could be used to do a number of very useful, difficult to accomplish tasks..
When I get around to it, I am also planning to add a third line/axis. The code is going to be tricky, but adding a third dimension to this arrangement... now, forget ink or paint. Extruder nozzle is what you want. Room-sized, virtually any sized, 3d printing. Anyone want to fund that? I have too many toys going already, but that one is in the "hmm, it really wouldn't take much" phase.. Tensor math can be confusing, I did well in geometry and trig, but I have forgotten more than I ever knew in the thirty years since my last geometry class
@focalist: Motivating to read up on your project. I think I'll try that from-scratch approach -- I'm very new to the whole linear-rails-stepper-motors world, but if you were able to do it with a few tens of dollars, I suppose I have nothing to lose by giving it a shot.
For a crude first-draft, would you say this covers the shopping list?:
--slides (e.g., kitchen-drawer-type)
--threaded rods
--one stepper motor for each threaded-rod/direction
--nails/screws to set everything stable
Despite mine and Terry's comments, those little steppers are not really suited to the job very well. They are made to turn vanes on air conditioners and stuff like that and are not that robustly built. They are made amazingly well for two dollars.. but they are two dollar motors!
Happily, steppers are not really expensive and you don't need anything with immense torque or speed. You should be able to get away with any "standard" stepper. Here's one that was one of the first hits on an ebay search..
That's five NEMA17's for $43, pretty good deal. You are just looking for something similar to these, exact ones don't really matter all that much.
The drawer slides really do make amazingly stable rails for what they are. Use two per axis, spaced as far apart as possible, that'll keep the wiggle to a minimum.
The key for you is going to be keeping that wiggle factor to a minimum. As for movement, if I remember right, the 1/4" rod is 12 thread per inch as a standard.. which means it will take 12 turns to move the carriage one inch. Those ones above are 1.8 degrees per step, 200 steps per revolution.. so that's 2400 steps per inch. WAY smaller than a millimeter. The thing is, the construction method and "play" obviously makes a difference. Even if you get this down to only 100 steps per inch of movement, 1/24th of the stepper/rail movement accuracy, you should be reasonably good to go.
Buckets of epoxy, sticks of it, and have hotmelt handy too. You are going to want it!
Very cool with the DIY nut.. two things come to mind. One is that you could probably well-oil the rod and then form the "nut" out of Hotmelt glue, doing essentially the same thing as HDPE (might even be HDPE for all I know) except not having to do all that clamping and such. Epoxy grabs pretty tight and is more brittle usually, but "pouring" hotmelt into an impromptu mold would be simple.
The other is to mention to the original poster that since his build is rather small- larger than a disk drive size but not big ole CNC router either- he could very likely harvest the carriage assembly from a printer or scanner. They tend to be really quite good.. attach a drive screw and you are good to go.
Which brings me to the last idea: many printers and scanners have both a very solid rail system and a drive method which is often a stepper motor. It may be worth it to just hack the whole assembly, and use the existing stepper motor. Printers tend to belt-driven, so the delivered torque might not be as high as a leadscrew-- but he may not any real torque for what he is doing..
focalist:
Very cool with the DIY nut.. two things come to mind. One is that you could probably well-oil the rod and then form the "nut" out of Hotmelt glue, doing essentially the same thing as HDPE (might even be HDPE for all I know) except not having to do all that clamping and such. Epoxy grabs pretty tight and is more brittle usually, but "pouring" hotmelt into an impromptu mold would be simple.
You could also use that "shapelock"/"polymorph" plastic (goes by various brand names - technically called Polycaprolactone: Polycaprolactone - Wikipedia)...
It'd probably be tougher than most standard hotmelt glues (although some of those industrial glues can be pretty tough - but need a much higher heat to melt); it's almost like nylon when cooled.
Putty epoxy on the oiled/greased threaded rod might also be an option. Also, two regular nuts on the rod put together to remove the slop (but not be tight on the rod) epoxyed/glued together night be another option to remove nut play.
focalist:
The other is to mention to the original poster that since his build is rather small- larger than a disk drive size but not big ole CNC router either- he could very likely harvest the carriage assembly from a printer or scanner. They tend to be really quite good.. attach a drive screw and you are good to go.
Which brings me to the last idea: many printers and scanners have both a very solid rail system and a drive method which is often a stepper motor. It may be worth it to just hack the whole assembly, and use the existing stepper motor. Printers tend to belt-driven, so the delivered torque might not be as high as a leadscrew-- but he may not any real torque for what he is doing..
Oh man -- this is a useful idea; I do have an inkjet lying around, and this should cut down the time to final build. I was going to go shopping at Home Depot this weekend for the earlier idea of getting the individual rails but postponed it; glad I didn't go (although would probably have been educational to build the thing from scratch). @focalist: Really appreciate all your input; thanks again.
Tell you what- here's something you can do in return. I'm building and trying to have promoted a safety valve system for residential gas installations- long story short, I lost my parents in a gas explosion as a teenager, and am trying to make those tragedies a thing of the past.
In reality, it's not money which will make or break this thing- it's public knowledge and support. The technology behind it is quite simple and established- I am just tying a few things together. What the project really needs is for people to understand what it is and what the impact the device can have. It needs for people to say "hey, why isn't that in my house already?". It needs to attract both the media as well as industry attention to succeed. That's people, not money. Most engineers recognize that the device itself is pretty simple, almost common sense. I believe that due to a number of influences, the technology costs and ability to muster support are now within reach, implementing a device to save the millions of lost dollars and the lives lost every year due to preventable tragedies is a possibility I have dreamed of for thirty years- and you can help, if only by looking at it and thinking about what it can prevent, and maybe telling others who might have the power to help do what is required to do what I am attempting.
What I'm asking is that if you can't contribute any other way, take a look at it, tell people you know about it. This will take groundswell support to move an entire industry. The device works, I am trying to upscale it and to bring a couple of people to bear that have some of the legal and industry contacts required to make this device a reality.
Will do what I can on my social networks to help spread it!
It's a much-needed project, and was inspiring to read the background on what first motivated you to do this.
Also, have you considered submitting a summary to some tech blogs/blog authors? Early PR can help generate a lot pledges for crowd-sourcing projects, and for a project like this which practically everyone can relate to, people becoming aware of it on popular blogs might make all the difference.
Been doing it, no bites yet. I've been interviewed by the patch.com affiliate, and I am meeting with the editor of a local paper this week. What would be really ideal would be something like mainstream media, of course.
It may very well be that if shared around, it might end up in front of the right person, someone with either the media or industry connections to move the mountain.. I am hoping that I can crowdsource the right connections, because it's the connections and media support that I need.
with no gantry (being a suspension plotter), scaling is just putting the motors further apart and more string, and a few lines of code... fact is, it isn't outlandish to think that with a little better construction, something like this could be used to do a number of very useful, difficult to accomplish tasks..
