Dual Y Axis Steppers

I'm in the process of building a reprap, and am just about ready to hook up the electronics to test out the movements. The printer uses 5 NEMA17 stepper motors, 1 for x axis, 1 for z axis, one for extruder, and 2 for the y axis. I've found in other forum posts how people connect two motors in parallel for the use on the same axis (usually z), but I want to run each Y motor from its own driver. Is there simply a configuration setting in the firmware that needs enabled (Plan on using Marlin) or is it a little more in depth than that? Also, is there a way of doing it so that each motor can home individually (to prevent binding - the Y motors move the whole X carriage from both sides and if they got out of sync it would jam up) and then the move synchronously throughout the print?

Thanks

Shapeoko Marlin Dual Y Link

Some light reading...

Thanks after reading through that and some other forums I found that in Marlin's 'Configuration_adv.h' file line 154 there is an option to do this(uncomment the line to enable):

//#define Y_DUAL_STEPPER_DRIVERS

Now I just need to look into the dual homing.

EDIT:

If I were to wire the endstops together, then it would only read as being tripped when both are tripped at the same time. If I slowed down the homing speed, would the stepper that hits its endstop first stall until the other motor hits its endstop or would the stepper even at low speed have enough torque to slip the belt? Something to experiment with I guess

Edited 2 time(s). Last edit at 06/09/2015 03:33PM by 1nxtmonster.

You could wire them in series for NC, so that which ever endstop would hit would stop it.

I'm not even sure if Marlin supports 2 min endstops anyways?

Edited 1 time(s). Last edit at 06/09/2015 06:34PM by madmike8.

The reason I want both endstops to be activated is so that printer is 'squared off'. If one side is further than the other then the carriages won't be perpendicular to each other and prints will go all screwy. By having two endstops, one for each side, I can assure that the carriages will be perpendicular. Marlin doesn't support two minimum endstops (or at least i couldn't find anything to say otherwise - if someone knows how please chime in) so by wiring together the two endstops it wont be tripped unless both are. The issue I think i'll encounter with this is getting both endstops to trip because if there is a misalignment, one end will hit first and that stepper will keep going, likely slipping a shaft or belt, until the other end hits its endstop as well.

Edited 1 time(s). Last edit at 06/09/2015 09:20PM by 1nxtmonster.

I think you're over-thinking this.

If your Y-axis motors are out of line, then *once* when you home the Y axis one or other will skip a few steps. Thereafter they will be in line.

If they don't stay in line, then you have some other problem that needs addressing (friction in the movement, low current to the motors, whatever) that needs to be addressed. End-stops don't stop the carriages getting out of line... they only detect it. And then only if you home the carriage. If you have some motion issue, then after homing the carriage and re-aligning, the next movement will put things out of line again.

The friction is fine and it shouldn't misalign on it's own once aligned. But my moving the bed by hand could misalign it some. I agree with you that when it homes it will fix the alignment, but what I'm concerned about is how it will fix it. Would the stepper miss a step before belts slip and parts come loose?

If the belt slips, that would be perfect, because after it slipped, it would be in the right position. Unless it also slips when the bed isn't against the endstop.

Whether the stepper misses a step before the belts slip depends on the belts. If they're GT2 timing belts, then it will take a lot to skip a tooth. Parts would have to be weak to break before it skips a step. You can turn down the current to the motors so that you can guarantee that it will skip before anything breaks.

Quote
madmike8
You could wire them in series for NC, so that which ever endstop would hit would stop it.

Are you sure about the serial wiring?
I thought it through over and over again, and came to the conclusion, they had to be wired in parallel to build an "or"-function.
-Olaf

My first Reprap that I built about 6 years ago used 2 stepper motors for the Z axis. ---- [www.reprap.org]
It used a Arduino Mega but not a Ramps board.


Each of the two Z stepper motors have their own LED home sensors as well as separate stepper driver boards. An additional PCB is used to allow these two Z stepper motors to auto synchronize when being sent to their home position. Basically the stepper motor driver boards (either one ) don’t receive step pulses from the Arduino mega processor if that stepper motor LED home sensor is blocked when the motor is being driven in a backwards direction towards the home sensor. The Z axis home signal is only sent to the processor when both Z axis LED home sensors are blocked. This hardware feature also stops the processor ramming the extruder heater barrel into the bed plate should the processor ignore the LED sensor signal for any reason. The auto synchronize PCB circuit that I designed for the Z axis is designed for the LED home sensor signals to be 0 volts (e.g. driven to 0 volts) when the sensor is blocked e.g. axis is at home position.

This setup has worked well for me,,,, no issues with moving Z axis motors about as they both get homed before printing ......

Would be good if we could have the option to have the Marlin firmware do the same thing for any axis without having to build extra hardware apart from a second home sensor input to the ramps board and then have the firmware control both stepper drivers.


I was going to look into this but ended up designing a CoreXY printer that only uses single motors for the axis's............................[forums.reprap.org]
However I can still see a need even with CoreXY to have 2 motors for the Z axis as this helps with better print platform movement.


Cheers Bruce.

Quote
o_lampe

Quote
madmike8
You could wire them in series for NC, so that which ever endstop would hit would stop it.

Are you sure about the serial wiring?
I thought it through over and over again, and came to the conclusion, they had to be wired in parallel to build an "or"-function.
-Olaf

Not for normally closed (NC, as he mentioned) endstops. You would want parallel for normally open, though . . .

- Tim

Quote
tadawson

Quote
o_lampe

Quote
madmike8
You could wire them in series for NC, so that which ever endstop would hit would stop it.

Are you sure about the serial wiring?
I thought it through over and over again, and came to the conclusion, they had to be wired in parallel to build an "or"-function.
-Olaf

Not for normally closed (NC, as he mentioned) endstops. You would want parallel for normally open, though . . .

- Tim

OK, I expected NO switches, because IMHO that´s the common case.
-Olaf

Quote
o_lampe
OK, I expected NO switches, because IMHO that´s the common case.

If NO is common, so is NC. NC are generally recommended because if a switch should fail open, or a wire becomes disconnected, it will be seen as tripped and movement halted. A NO switch will appear the same whether it's connected or not, creating the potential for bad things to happen.

I´m only familiar with the Prusa world and have never seen a NC microswitch or optical switch for them. But I understand, that NC switches are saver.

THX
-Olaf