Thursday, 24 July 2008

A bit of a contraption

As I was adding the diagonal tie bars my wife said "it's becoming a bit of a contraption". I am not sure if that is a good thing or a bad thing!

My alternative z-axis using four tin can stepper motors works reasonable well. I am running them from 36V with a constant current chopper drive. They are all wired in parallel and the total current is set to 1.25A. They have 22Ω coils so that corresponds to about 7V. I am guessing they are rated for 12V, but if you run steppers at their maximum rating they get very hot, especially when mounted on plastic rather than a metal chassis. Under running them, as I am, they only get to about 40°C, which should be fine even for PCL and PLA brackets. Total power used by the axis is about 9W.

The standard Darwin z-axis can carry a small child: reprap-prints-child. Not having any small children, I tested it with a vice that weighs 3.3Kg. As it would take 300 hours to print anything that size so I think it is a reasonable worst case test.

The pull-in step rate (i.e. the maximum rate that the motor will start at with no acceleration) is about 400 steps/second. It runs reliably at 320 steps/s, which is 8.33 mm/s. If I understand the settings page on the wiki then this is more than 10 times faster than people are running the belt drive version. Still much slower than HydraRaptor's z-axis though.

Here is a video of it in action: -

Alternative RepRap Darwin Z-axis from Nop Head on Vimeo

As you can hear, it doesn't make a lot of noise, something that is a big improvement on HydraRaptor, which has a very noisy z-axis.

The total travel is 230mm, which is also a bit better than the standard Darwin I think, but you have to subtract the length of the extruder barrel to get the maximum work height.


  1. That's awesome! Is it cheaper to go with the four-motor approach, or at least comparable? Much of my frustration with the RepRap so far has to do with the timing belt and trying to drive the four posts with one stepper motor (not enough torque). I like you idea a lot.

  2. Cool! You got it working with those big tin can steppers! Congrats! :-D

  3. Andy,
    I can't find a retail source for these. The nearest I found is RS do one that is unipolar, has no pin through the shaft and no mounting lugs for £12.48, RS are never cheap. So not really an economical success.

    The smaller motors I tried first are cheap enough but it would mean using smaller threaded rods. Something I may come back to after I get the machine running.

  4. You have to watch out with those smaller rods. They're floppier and tend to buckle in vertical orientation under load with anything less than about 1/4 inch diameter. That's the biggest issue I'm having with the linear steppers.

    Mind, if you were to mount the steppers on the topside rather than the bottom that problem would go away.

  5. Hi Forrest,

    Roughly how much would your Chinese supplier charge for four tin can steppers powerful enough to turn these rods?

  6. Apologies if this has already been covered, but - a customer came to me to buy some obsolete computer printers, because he wanted the stepper motors for moving the printhead side-to-side and for the paper advance. Inkjet, laser and dot-matrix (except VERY early models) were apparently suitable, common resolution would be 600 dots-per-inch. Any use ?
    And let me say what a pleasure to follow this project, congrats to everybody concerned !

  7. thetrevor,
    The motors that I have found in printers are smaller than the ones I ended up using, but bigger than the tiny ones I tried first. They are probably about right for a z-axis but finding four identical would be the problem as you tend to only get two in a printer and often not the same.

  8. Nophead,
    I recycle probably 50-100 printers a month, it wouldn't be a problem to get four identical motors. Might take a (short) while to get suitable matches, but it's feasible, certainly. What are the characteristics I should look for if I find myself surrounded by motors ?

  9. thetrevor,
    The be compatible with the RepRap electronics they need to have four or six wires and take less than 2A and need no more than 12V.

    For the z-axis they obviously need enough torque to turn the rods but I don't have a figure for that and you probably don't have a figure for the motor so it's trial and error really.

    For the z-axis a large step angle an low inductance are advantages for speed.