Z Axis not moving during print

Hey,

I'm new to 3D printing and I've recently finished building my printer (Cherry 3D printer) and have been trying to fix it's problems, this one I could not figure out, so maybe you guys can help me out!


The problem is the Z axis does not move up after the printing starts. It DOES however move during homing and in manual control. This results in the everything getting printed on first layer...


I am using firmware Repetier 0.92 btw. And the driver currents are adjusted correctly. (0.12 mA, they are small steppers!)
The endstops are working correctly, they show low when endstops are not triggered and high when they are.

Z-axis Max Feedrate: 0.3 mm/s
Z-axis acceleration 5 mm/s^2
Z-axis travel acceleration 5 mm/s^2


Also, if and whenever I press "Emergency Stop" button to stop everything and then try to move the Z axis, it won't move.
I'll have to wait a few seconds for the non-existent Z movement (as if it's actually moving when it's not) to be finished and then the next Z movement commands work.

I've attached my EEPROM.

Edited 1 time(s). Last edit at 04/08/2016 04:06AM by White.Renard.

I am thinking:-
Lack of current from the PSU to allow the steppers to operate at the same time as the hot end = unlikely.
Somewhere the z axis direction is inverted = when printing the steppers are trying to lower the hotend and it just sits on the z end stop.
The Z offset in your slicer is incorrect (Has a large positive number in it). = The slicer thinks the bed is a lot lower than it is, tries to move lower but can't because its siting on the end stop. Can extrude the first layer because of the physical gap between the zstop and the bed. The next layer it has calculated is still lower than the bed, still cant go lower as siting on the zstop.

I checked the Z offset in slicer, it was set to 0.
I also tried triggering Z endstop manually during homing precedure to prevent it from actually going to Zero location. It started printing in mid air and the Z axis still didn't move, so if they were inverted they should have moved down towards the bed.
Btw, the Z location on Repetier Host does show the movement, it went from 0 to 1 before I stopped the print.

This is the last GCode for Z axis (Layer 3) that I found in the log:

G1 Z1 F3000*5

I entered it manually after stopping the print and it actually did move the Z axis. So why doesn't it do that during the print?

That F3000*5 is too fast for your printer - 250mm/s. The speeds must be set too high in the slicer, I'm guessing something clsoer to 5mm/s is more reasonable for Z. The firmware should have limited the speed to the maximum that your printer is physically capable of, so the fact that it still didn't work also suggests that the firmware isn't configured correctly. The acceleration values in the firmware also need to be set to reasonable values for your hardware, so it's worth checking those too.

I changed every speed in slic3r settings to 9. (They were all above 9 before)
The G code for Z axis now shows F540. I think the Travel value in Speed for non-print moves section is where the Z speed is taken from. Since I changed it to 1, the gcode shows F60! But that still didn't solve the issue. I can't set it below 1 either and it affects other axis too!

Weird thing is, no matter how much I set the Z feedrate (or travel feedrate) in EEPROM, Z axis moves exactly same speed! Nothing I tried could change Z axis speed. It does work for other axis though! I feel like I'm missing something obvious!

I restarted Repetier Host and my printer as well and tried again, no difference.

Even with custom G code like G1 Z1 F18, F3000 etc... it still moves the same speed.
In the logs, when I send a G code (or manual control), the F value is multiplied by a number, what does that mean? The number changes every time.

You're right, and I was wrong - there doesn't seem to be a separate setting for Z speed in the slicer, although there is one in the host software for controlling the manual moves. I guess that must have been what I was thinking of. So I guess we need to figure out what's going on with your firmware.

Edit: I took a look at your eeprom settings and I don't see anything that looks like a problem. Like you said, you have max Z feedrate set at 0.3. I'm baffled why it would work during manual moves but not during the print.

Edited 1 time(s). Last edit at 04/08/2016 09:59AM by JamesK.

Yeah, It baffles me too! I can't think of anything anymore...
Thanks for responding here btw

I figured out why changing Z feedrate didn't affect the speed. I had to change Z feedrate in Printer settings, I had it set to 18 mm/m, so it didn't go beyond 0.3mm/s in EEPROM.
But the main problem still presists...

Trying to think how to narrow the problem down - I could imagine three possibilities:

- Mega2560 not sending the correct signals to the stepper driver

- Stepper driver not generating current to the motors

- Motors not responding to current

Do you have access to an oscilloscope? That would be the most direct way to visualise what's going on. You might be able to pick up signs of life with a meter, particularly one with freq and duty cycle modes, and failing that you could lash up something with an led to try and probe what's going on.

Could you run your current calculation past me again, I'm thinking that was probably a typo in your first post? I looked up the motor and found a value of 200 Ohms for the dc resistance. At 5V that would give 25mA and with two motors in parallel (I think for your Z axis?) that suggests you'd need the driver set to provide around 50mA.

I don't have access to an Oscilloscope unfortunately... They're pricey!
And I don't think my multimeter has frequency or duty cycle modes. How would I do it with a an LED?


Well I mesearued the Vref on the drivers and for X and Y axis, they are 0.06 and for Z axis I've set it to 0.11. The motors don't get hot at all, just a little warm.
These motors are hacked to be bipolar, using this guide: http://www.electronicsmayhem.com/?p=13
The resistance is doubled after doing this.
Both of my Z motors are connected directly to Ramps, this would be parallel, right?

Quote
White.Renard
I don't have access to an Oscilloscope unfortunately... They're pricey!
And I don't think my multimeter has frequency or duty cycle modes. How would I do it with a an LED?

For checking voltage output from an IO pin you could just use a LED in series with a 1k resistor - that would draw about 4mA and should be visible, and the IO pins should supply that easily. Connect the negative side of the probe to the supply 0 and solder some sort of mostly insulated probe pin to the positive side of the LED (a piece of solid core wire would do). Then carefully hold the pin to the enable, dir and step pins in turn and see if there are signs of life. You can compare with the x or y axis to check what it looks like when things are working, but I would expect the enable pin to be on all the time, the dir pin may or may not be on depending on which way the motor turns and the step pin should flash (but probably too fast to see, may just look dimmer than solid on) when moving.

What voltage power supply are you using? You can probably use the same 1k resistor for a 12v setup, the led will just be brighter. To check for output from the stepper driver you would connect the probe across each of the motor output pairs. I don't know which will be positive and which negative, so you'd either have to try it both ways or solder a second led back to back with the first one so that one or the other always lights up.

Be careful not to accidentally disconnect the motor, the stepper drivers can be damaged if disconnected while powered up. It may be tricky to find a good place to connect the probe to to test this, it's not something I've tried. You don't want to short anything out either.

Quote

Well I mesearued the Vref on the drivers and for X and Y axis, they are 0.06 and for Z axis I've set it to 0.11. The motors don't get hot at all, just a little warm.
These motors are hacked to be bipolar, using this guide: http://www.electronicsmayhem.com/?p=13
The resistance is doubled after doing this.
Both of my Z motors are connected directly to Ramps, this would be parallel, right?

Yes, the Z driver on the ramps has the two sets of output connectors in parallel, so you set the Vref to deliver 2x the current of a single motor. If the motors are only just warm you can probably afford to increase the Vref a little. The motors will survive being quite warm, but not burning hot. 40C or so should be ok.

Edited 1 time(s). Last edit at 04/08/2016 07:04PM by JamesK.

Quote
JamesK
For checking voltage output from an IO pin you could just use a LED in series with a 1k resistor - that would draw about 4mA and should be visible, and the IO pins should supply that easily. Connect the negative side of the probe to the supply 0 and solder some sort of mostly insulated probe pin to the positive side of the LED (a piece of solid core wire would do). Then carefully hold the pin to the enable, dir and step pins in turn and see if there are signs of life. You can compare with the x or y axis to check what it looks like when things are working, but I would expect the enable pin to be on all the time, the dir pin may or may not be on depending on which way the motor turns and the step pin should flash (but probably too fast to see, may just look dimmer than solid on) when moving.

What voltage power supply are you using? You can probably use the same 1k resistor for a 12v setup, the led will just be brighter. To check for output from the stepper driver you would connect the probe across each of the motor output pairs. I don't know which will be positive and which negative, so you'd either have to try it both ways or solder a second led back to back with the first one so that one or the other always lights up.

Be careful not to accidentally disconnect the motor, the stepper drivers can be damaged if disconnected while powered up. It may be tricky to find a good place to connect the probe to to test this, it's not something I've tried. You don't want to short anything out either.

I'm sorry but I'm not sure what you're referring to when you say "Supply 0"? I know where the enb, dir... pins are but not this.
The probe is ready, I just couldn't figure out what you meant.

And yes, I'm using 12v power supply.

Quote
JamesK
Yes, the Z driver on the ramps has the two sets of output connectors in parallel, so you set the Vref to deliver 2x the current of a single motor. If the motors are only just warm you can probably afford to increase the Vref a little. The motors will survive being quite warm, but not burning hot. 40C or so should be ok.

I can try to increase it a little bit but a bit more current than what I've adjusted right now makes the motors whine when they move, it's really annoying.
Unless I increase it a lot then the whine goes away but they get super hot!

Edited 1 time(s). Last edit at 04/09/2016 06:50AM by White.Renard.

Sorry, I should have been more clear. I just meant the 0V or sometimes marked as -V side of the supply. You can connect to it anywhere that's convenient, but it's often easiest if there's a spare connector at the power supply to wire the probe in there so you don't have to try and hold two connections in place by hand. You want to be able to concentrate on the + side of the probe and not worry about the other end.

Yes, whining stepper motors tends to be normal. It can be quite amusing if you have any musical tendencies - I often hear melodies as my printer does it's thing

What sort of stepper drivers are you using?

Quote
JamesK
Sorry, I should have been more clear. I just meant the 0V or sometimes marked as -V side of the supply. You can connect to it anywhere that's convenient, but it's often easiest if there's a spare connector at the power supply to wire the probe in there so you don't have to try and hold two connections in place by hand. You want to be able to concentrate on the + side of the probe and not worry about the other end.

Yes, whining stepper motors tends to be normal. It can be quite amusing if you have any musical tendencies - I often hear melodies as my printer does it's thing

What sort of stepper drivers are you using?

Aah, ok, got the probe working now!
I turned printer on, connected Repetier Host, tried Enb, Dir and Step pins. At this point only Enb pin turns on the led. So I tried moving the axis a bit then probing again.
This time, Enb pin doesn't turn on the led, but Dir and Step do. led is pretty much completely lit (no blinking) on Dir pin. On Step pin, it's just like Dir only it's dimmer this time. All 3 axis return the same result.

I'm using A4988 drivers. (with heatsinks)

Awesome, so we can rule out a problem with the signals between the mega2560 and the stepper drivers. Or maybe I'm getting ahead of myself - did you see the same pattern of behaviour both when the Z axis was moving and when it was supposed to move but didn't? I know that second bit will be hard because the problem Z movements are so short.

Now to check for output from the driver to the motor. My first thought was to connect the probe between pairs of output pins from the driver, but I guess you could keep the -ve side of the probe connected to 0V and work your way along the 4 pins in turn. All you're looking for is some sign of changing output when the Z axis should be moving and isn't, so any one pin is probably enough.

Alright, I just tried doing this again, but during the actual printing. After hotend heated and printer started working, I held the probe on Dir pin and waited for first layer to finish, the led turned on fully lit when Z was supposed to move, but no actual movement. I did the same thing for Step pin on 3rd layer and it also lit, just dimmer. I tried Enb pin while waiting for 2nd layer to finish, didn't turn the led on.

Btw I'm running the Z motors on Half step, the others are 1/16, if that makes any difference.

OK, that all sounds good for signals getting through. You took the 1/2 step setting into account when you calculated the steps/mm for Z? As long as it matches up ok 1/2 step should be fine, if a bit noisy.

Quote
JamesK
Now to check for output from the driver to the motor. My first thought was to connect the probe between pairs of output pins from the driver, but I guess you could keep the -ve side of the probe connected to 0V and work your way along the 4 pins in turn. All you're looking for is some sign of changing output when the Z axis should be moving and isn't, so any one pin is probably enough.

So you mean the pins that the motor wires are connected to? 1A 1B 2A 2B?
I held the probe on 1B and when Z was supposed to move during print, the led went a bit dim and fully lit again (or vice versa, it was fast). Like a very fast blink maybe.

Quote
JamesK
OK, that all sounds good for signals getting through. You took the 1/2 step setting into account when you calculated the steps/mm for Z? As long as it matches up ok 1/2 step should be fine, if a bit noisy.

Yes, I took that into account and calibrated Z with half step.


P.S. I just tried a new stl model just in case if the first model was broken or something, but it made no difference, same problem here as well.

Edited 4 time(s). Last edit at 04/09/2016 11:08AM by White.Renard.

Yes, those are the pins. So I think you've proved that everything is working like it's supposed to in terms of getting signals from the controller to the motor, but for some reason the motor decides it doesn't want to play, but only sometimes. I think that takes us to the speed, acceleration and jerk settings. Try setting the Z jerk to 0, and the Z accel to 0.1 and see if it makes any difference. And bump up the current to the definitely warm but not burning hot level, even if it's noisy

Quote
JamesK
Yes, those are the pins. So I think you've proved that everything is working like it's supposed to in terms of getting signals from the controller to the motor, but for some reason the motor decides it doesn't want to play, but only sometimes. I think that takes us to the speed, acceleration and jerk settings. Try setting the Z jerk to 0, and the Z accel to 0.1 and see if it makes any difference. And bump up the current to the definitely warm but not burning hot level, even if it's noisy

IT MOVED! IT FREAKING MOVED!!!

I'm printing a 20mm cube right now!

I increased the current by about 50% and changed the accel and jerk like you suggested! I'm not sure which one made the difference, but I'm gonna try the previous settings after this to see if it was the current or the settings... or both?!

Anyway, I'm super happy right now, thanks a lot man!
Probably would have taken me weeks to fix it by myself!
I appreciate it!

You just made my day Enjoy yourself, it's just awesome when you get one of these things going. Don't expect miracles straight away, it takes a while to get everything tuned up. Have you seen the calibration guide - loads of helpful stuff for getting everything dialled in.