Wally Z Axis Motor Brake

[see3d]

A bit of a brainstorm hit me tonight for how to keep the Z axis from going into free fall if the power is removed. A stepper motor will become a generator if spun without applying power to it. A shorted coil on a generator causes it to brake hard. So, what if the Z axis was attached to the controller through a 4PDT 24V DC relay. If the power is removed, the coils would be disconnected from the controller and shorted out. That should keep the Z from falling quickly. The non powered state of the relay is of course the one that shorts out the coil. It might even work with just a single coil shorted with a 2PDT relay (much more common). It should be easy to experiment by just disconnecting the Z motor and shorting the wires and see how it falls. I would try it, but no Wally here to test.

[Nicholas Seward]

Nice! Easily done.

Won't I need a few more components to turn the two coil voltages into something I can feed the relay. How do you plan on energizing the cool?

This works on CoreXZ and others.

[see3d]

Here is a hand drawn sketch of what I mean. The 24V relay coil just goes to the power supply, or it can go to a controller driver circuit if you want it programmable to take the brake off. Hmmm, I don't see a way to upload a screenshot. I will email it to you.

[Nicholas Seward]

Hmm. Wiring to the PS will take care of some of the problems. Getting at the driver's supply would be better but not as easy. I think we can do this with a circuit inline with the 4 coil wires.

[see3d]

Yes, if you put 8 diodes from the drivers to the relay coil (double bridge rectifier), then the relay would be energized if the motor was being powered. A small inline circuit board between the driver and motor could take care of it. It could all be done electronically instead of electro-mechanically, but it is easier to understand and test with a relay.

[Nicholas Seward]

Okay, I was wanting to throw something together and test this but I am dumb in the ways of electronics. Steppers that we use can be anywhere from 2.5V to 4.5V. How would I do this? Would I have the double bridge rectifier piped through a voltage regulator to a 3V 4PDT?

[jasonharper]

I've heard that a sudden disconnect of a powered stepper motor risks destroying the driver (well, the cheap ones used for RepRaps at least, presumably commercial stepper drivers are more robust). There might be problems if the relay drops out before the driver power supply decays to zero. A big electrolytic capacitor across the relay coil, after the diodes, would delay the turn-off a bit.

[Cozmicray]

I guess it is a problem?

I don't have a slab of basault on my bed

From the past
"Cozmicray I don't think the bed dropping (whack or Twang) is very good?

Nick There is no reason this should ever happen unless you have a malfunction. In the post code in slic3r enter a "G1 X0 Y0 Z0" and you will gracefully return to home and you are ready for another print. You could also redesign the z motion to have some kind of spring assist for the z movement. For me the whack doesn't happen anymore and when it happens from a malfunction it isn't destructive."

viewtopic.php?f=7&t=28&p=760&hilit=bed+drop#p757

[Nicholas Seward]

@jasonharper: Smart precaution

@Cozmicray: This goes in the wouldn't it be nice category. I don't think the dropping really hurts anything but it isn't something I would want to have happen on a production bot. I have quite a few other bot ideas that would benefit from having a brake function when the power is cut. For instance, I want to make a 1-Arm SCARA that will be oriented in such a way that the arm has to fight gravity.

[see3d]

Several comments:
@jasonharper: A large cap across the relay coils is not a good idea with this diode circuit, as it would suck all the power away from the stepper and perhaps even destroy the driver circuits.

If pulling the plug to the stepper motors on the controller will kill the controller drivers, something is wrong with the design.

I would be concerned with the sparking on the relay contacts reducing their life if it were not for the fact that this should be a more rare event.

@Nicholas: I would experiment with something simple first, like shorting one or two unconnected motor coils and letting the bed drop to see what happens in each case.

Relays usually have a pull-in current and a release current. There is a built-in hysteresis to disconnecting. That makes them more tolerant of dropping out due to driving voltage variations. Low average voltages into the relay coil due to low voltage motors makes it more difficult.

The driver may output up to the power supply voltage during initial steps (it is actually a current regulator, not a voltage regulator). This could create too large a variation for the relay coil. A small resistor in series with the coil and a small cap in parallel with the coil would smooth out the voltage/current variations it sees. The coil voltage/resistance/current specs have to be derived from what the minimum average voltage seen at the coil from this diode circuit. It is best done with an experiment and an oscilloscope. You can substitute a larger resistor for the relay coil for the initial experiment. For instance motor coils to 4 diodes to a 100 Ohm resistor to a 1K Resistor with a 1 uF 50V cap across it, to 4 diodes to the motor coils. Measure the differential voltage across the 1K resistor with the steppers micro-stepping slowly, and at high speed moves. That should give a good idea of what the relay coil would see. The look for what might be available as a coil resistance/current in a relay.

If the voltage variations are too high, and the average voltage is too low for a practical relay coil, then I would look at bringing out a power supply and ground wire to the relay board. Then the relay coil can be driven from a transistor that can be gated by the low voltage signal derived from the motor coils.