Stepper Voltage Question

I recently purchased the Pololu A4983 Stepper Carrier [www.pololu.com] to drive my reprap setup. this requires an external power source. and i wanted to make sure I was calculating the voltage properly.

my steppers are [www.alltronics.com]
2 A/Phase
1.2 Ohm/Phase

So I am not sure how to make the math work. If want to run this at 1A I should use a 12V power source, correct? If I get a supply that is rated at 2A is that enough for 3 steppers of do I need 3-6A?

i am starting to pay more and more attention to this forum i need to learn allot of stuff like this too, i am still a newbe when it comes to electronics.

---

[mike-mack.blogspot.com]

Yes, 12V. The chip adjusts the current. Remeber yu will be running other things off f 12V for extrusion. In general, you will want to get an ATX power supply for your system. 200W - 300W should do. Go higher if you find a deal. It helps reliability.

---

B^2 : [replibot.blogspot.com]

~~ We Are The Factory ~~

Current (I) in Amps = Voltage (V) in Volts divided by Resistance (r) in Ohms

I=V/r V=I*r r=V/I

To drive one Amp (I=1) through 1.2 Ohms (r=1.2) you need 1.2 Volts (I*r)

A 12V supply can drive as much as 10 amps through 1.2 Ohms (I=12/1.2) but he driver chip provides current limiting so it's OK to use the higher than necessary voltage.

I was not sure about the units. So, I am off by a factor of 10.

The steppers are rated as 2A 24V, if these are not scaled by a factor of 10 then isn't it impossible to achieve 24V since that would be 20A.

This is actually why I asked the question since the math is not making sense to me either. The only explanation was everything was off by 10x.

Motors are funny things when it comes to Ohm's law. The current you derive from measuring the resistance of the coils is the "stall current" which is the maximum current the motor can ever draw. If the motor is held motionless, it will actually draw this current, and quickly burn up. When the motor spins freely, the situation is different - the induction in the coils causes the current to ramp up slowly, and so it switches ('cause the motor moved) before it gets to a super high current.

The only reliable way to determine the draw of a motor is to read the plate. It'll tell you the ratings (amps, Watts or HP) for the motor. Don't hold the motor harder than its HP rating or it'll bog down, draw too much current, and melt the insulation on the coils.

Of course, none of this applies here.

Stepper motors are a different beast, and the chip uses PWM to control the current going to the coils to the maximum set by some sense resistor, in your case the little pot on the breakout board. You can adjust the current going to the motors by changing the pot, to little and the stepper doesn't move, doesn't move under load, or skips steps under load. Too much and the motors get too hot and, again, melt the insulation on the coils.

I'm sure there's a test point on the boards, you should (in theory) be able to adjust the pot while testing the voltage, until the voltage reflects the amps you want. Not sure how that works exactly, but there's got to be directions somewhere. The data sheet for that chip, if nowhere else...

---

--
I'm building it with Baling Wire

I think I understand all of that, and I am able to trim the current using the pot on the board down to ~1A, but should I be trying that at 1.2V or 12V?

mfsamuel, as jgilmore said, ohms law only tells you stall current. When the motor is moving, it requires more voltage for the same current due to its own back-emf and inductance. At some rotation rate which you could calculate with the full motor datasheet, it will take 12v to push an amp into the motor. At some higher speed, 12v will no longer be enough to overcome the friction and other losses (assuming an amp was enough), at which point it will go no faster, and start skipping steps. Don't think that 1.2v is enough

Torque is proportional to current, so all good stepper controllers ignore voltage and control current tightly. In fact, it's necessary for proper microstepping!

Run them at 12v, and let the driver sort out the current for you.

ps: I use the A4983 drivers, and I found I can't get more than 0.5A without a heatsink. See my latest blog post for the arrangement I ended up using, which seems to be most effective. The motors get hot while the drivers stay cool now

[wooden-mendel.blogspot.com]

great thank you Triffid_Hunter.

i have read your blog, and (i think) your posts on the pololu forum and am fully prepared to heatsink these babies.

I have some harddrive coolers that should work for this, but the header pins and resistors seem to sit higher than the chip. A copper shim with thermal paste to get the surface flush with the other heatsink may work, but the thermal conductivity of that may be poor.

I can always fish tank these babies in mineral oil if all else fails. <-- I may actually do this

far more important than thermal paste is lapping your heatsink (and shim) mirror smooth.. get wet and dry sandpaper in 320, 800, 1200, 2400 grit. start with the 320, put it on a piece of glass (for flatness), add a bit of water and start sanding your heatsink in a circular motion. keep going until you can pick up the glass with the surface tension alone and the surface looks uniform, then switch up to the next higher grit. after you finish with the 2400 grit, the surface should literally look like a mirror.

You may be appalled at how far from flat your heatsink starts- you can see the curvature easily when you start off as the lowest sections will shine first

When this is done properly, thermal paste only makes the tiniest difference


I think a copper riser is a great way of clearing the taller components. It has higher thermal conductivity than aluminium, so if you use a long piece that's as wide or wider than the board you won't gain much thermal resistance. Remember to lap this piece too- both sides for best effect.

The Gecko Controler pages have sime very informative and accesible docs, admitedly biased towards their product. But the fundamentals are sound.

[www.geckodrive.com]

---

Necessity hopefully becomes the absentee parent of successfully invented children.

thanks everyone, now I just have to wire them up properly.