RepRap: Blog

have you ran into any troubles with lack of torque with your micro stepping at 1/16th everything that i have read says they loose torque and some finer repeatability in the movements if it is too fine.
 

No torque problems - quite the contrary. I have the current pots only at about 1/4 their maximum setting, and the motors seem rock steady.

The repeatability problem is inevitable, I think. It is dynamically balancing two magnetic fields using PWM, after all. But just half-step gives more precision than we need, so slight variations on 1/16 step should not be a problem. But the great thing about microstepping is how smoothly and quietly everything works.
 

Is that some new kind of new extruder. I have never seen an extruder with such a large gear wheel. Have I been hiding under a rock? Where can I find more details?
 

Keith,

It's the Geared Nema17 Extruder Driver.
 

Great work, Adrian! That goes a LONG way towards making Mendel simple enough for a much larger audience to assemble. :-)
 

If you're lacking interrupt speed, look at Triffid's firmware. His is much, much faster, and shouldn't have any problem with very (ridiculously) high step rates.
 

I wonder if using digital-to-analog converters (DACs) instead of pulse width modulation would improve things?

The silence is really excellent - and reducing vibrations has a whole lot of other benefits than just audible ones. It should reduce the likelihood of screws shaking loose, for example. =) If it's difficult to overcome repeatability, perhaps it would be worth investigating some kind of closed-loop position control.
 

I've been running those drivers at 1/8 step for a while. I can't see any noise improvement on 1/16 stepping.
 

Transistors dissipate the least power when turned off (because so little current flows) or fully open (because their resistance is at a minimum). Driving them analog means spending some time at a setting where maximum power is dissipated within the transistor, which means much heavier-duty power electronics are necessary.

If, on the other hand, the pulse frequency is high enough that the inductance of your stepper motors absorbs most of the variation, that would probably help solve any unusual feedback situations that might otherwise occur.
 

That's a great improvement!
Especially because this way the electronics may be far more cheaper than G3 and they can be found also more easily.
Is this boards combination going to be the next "official" electronics for the RepRap? (I would avoid to buy another "old" G3 kit :o)
Fabio
 

This comment has been removed by the author.
 

Others are experimenting with these:

TB6560

:-) You just buy an Arduino Mega, one of those, and three optoswitches, and you're good to go.

:-( Not open source.

Now I've proved the Pololu electronics work satisfactorily, I'll finalize the simple single-sided PCBs for them. Those should be solderable by an eight-year-old...
 

PS - the firmware in the repository is now fully working for this design.
 

This looks great! I designed a single-sided shield to fit four of the Pololu drivers on top of a Duemilanove:

http://www.cibomahto.com/2010/06/one-shield-to-fit-them-all-and-in-the-darkness-bind-them/

Which I'll retool to match your updated firmware :-)
 

I'm thinking of changing the pinouts, so give it a little while before doing any electronics based on the current ones...
 

Has anybody tried CiP (Circuits in Plastic)? Perhaps the whole thing could be reprapped.