Axis goes one direction (Prusa, Marlin, Ramps 1.4)

Not the conventional one direction issue.

I have a stepper (in this case the Z axis) that from time to time, goes 'down' when you send a 'down' command, and also 'down' when you send it an 'up'. This is happening "fairly often", although I have just installed new electronic so I haven't used it enough times to get to do a proper print, just calibration (so twice). Also, I have not noticed it swap between directions once on, but haven't used it long enough.

First time I "fixed" it was when I was first calibrating and updated my z steps per mm from the default (of Marlin-latest as of a week ago) to what I actually have for a Prusa from an old Gen 6 *2, so initially I figured it was that, but couldn't think how.

When this happens, and I test the wiring, from the stepper 'test points' (top of the driver shield) to the other end of the wiring looks fine, can give that a wiggle and doesn't seem to be dodgy. This latest time I swapped the Y axis cable and it goes back and forth ok, then swap back to Z and is ok again. That doesn't quite tell me if it is a problem with the board or with the wiring, but does suggest that it is intermittent and related to hardware.

As originally, this happened 'out of the box' I also changed Arduino bases (from the 1280 which came with it to a 2560 I happened to have). At that time nothing changed, until I updated the firmware for z step per mm, which appears to be a complete coincidence.

Interested to know how to track it down, and fix the issue. Figure this could potentially blow a stepper.

If it is of value, I am using:
Ramps 1.4 (new)
Marlin "1.0.0 RC2" (as per the VERSION_STRING),
Pronterface and Slic3r (both latest as of a week ago).
Endstops: Mechanical.
Rest: Standard 'almost original' Prusa
Z axis wired in parallel (so resistance is halved)

Thanks.

Seems like the direction signal has a bad connection somewhere between the ATMEGA pin and the Allegro pin.

---

[www.hydraraptor.blogspot.com]

It could be the direction signal, but it could also be an intermittent connection on one of the four wires leading to the stepper, or the corresponding pins on the shield. Using only one phase of the stepper will cause motion direction to become unpredictable/reversible.

Thanks guys. The direction signal seems most logical, but I haven't been able to track that further.

When I check the connectivity from the relevant pin (D48) on the bottom of the MEGA board to the DIR pin on the top of the z stepstick, and also the same from the stepstick output to each of the motor wire, they are ok. (doesn't say anything if its intermittent).

I have pulled apart all the boards and examined the pins on the bottom of the RAMPS board, and while the solder varies, the contacts look ok. The only thing I couldn't see is where the DIR pin goes once on the stepstick stepper board (behind the heatsink). Also triple check my cables, and while I have cleaned, reseated terminators and whatever else, they are the same which I have been using for a year now on another electronics.

Putting it back together, I have swapped the X and Z steppers, and will see if it happens again.

So for now all the axis' move ok. Will update if it happens again. If the X starts playing up, then it might be easier just to put that stepper aside and move on.

At worst (next to having inconsistent electronics) I have learnt a bit about how RAMPS boards work.

Ok, happened again. This time though, it was only going upwards, then started demonstrating more obvious 1 phase unpredictable direction - so starting one way straight way swapping direction, and then consistently doing that for up and down.

Checked again and connectivity looked 'ok'. My multimeter is pretty basic and holding it still on the pins is not easy, so can be hard to say. Sometimes I think I moved my hand slightly. Also probably not a good contact for the current which is applied, but ok for testing connectivity.

Then I had a little think, in the context of one phase, the place where this would be the worst, would be at the connector to the Z output.

To avoid unnecessary changes on an otherwise serviceable printer, I had kept, along with the other axis, as the Molex KK type from the Gen 6. X and Y fit ok, but Z is a bit of a squeeze with the two connectors. Having a look at the connector, the terminator crimps are a bit sloppy and twist around a bit in the Molex housing - it is possibly twisting and making a poor contact with the header.

Anyways replaced that section of loom with a nice new set of wires terminated with the black flat type (don't know what they are called - but are common/standard). These 'hug' the pin.

Will see how that goes. Is running ok now, but I had that last night too...

Having swapped stepper drivers around and actually changed something I am a bit more confident that that is it.

If that is the case, I am also not sure how I didn't blow the stepper.

Peter

my printer had this same thing and when i switched around all of opto stops i found that i had a bad opto. thats what was making it go one way. because my printer thought it was always being blocked

If you have optos, make sure there are no intense light sources shining down on your printer (theSun, hallogen lights, etc).

thanks for all the responses.

Have changes all the connectors for the steppers, and it seems to be running ok now. Have run it up a half dozen times and can do prints which last longer than an hour without issue, so I think that was it.

Cheers

Hey I'm having the same Issue how did you fix it?? I have ramps 1.4 and latest version of marlin firmware using pronterface. I can only get my axis to go 1 direction when I have the endstopp plugged in. When I unplug the endstop they move in both directions.

Invert your endstop logic.