H-Bot Endstop problem

Veti

I did so. After G1 H1 X-395 F1800 it crashed.

this is VERY strange.
if M119 says the endstop is triggered the G1 H1 command should stop when the endstop is triggered.

The Firmware version is the correct one.

dc42

I agree, very strange. Typical causes of this type of problem would be:

Endstops pulling the input low enough to light the LED, but not low enough to register as a low input to the processor. But in that case, M119 wouldn't show the endstops as triggered when you test them.
X and Y endstop connections swapped
Axis movement incorrect (easy to do on a CoreXY machine), for example G1 H1 X-395 is actually moving the Y axis, or is moving X away from the endstop switch.

NL61

@dc42
thanks for answering,
I tried M119 serveral times for each axis, it always shows if triggered or not.
the endstops are not swapped and the movement goes to the correct direction.

deckingman

One thing that keeps leaping out at me from the config.g file, is the steps per mm for X and Y but I don't see what this would have bearing on the end stop triggering. But with a 16 tooth pulley, 2mm pitch, and 0.9 degree motors, the value is around 80 so having the steps per mm set to 32 would imply something like a 50 tooth pulley which is pretty huge. 32 steps @16x micro stepping is 2 full steps per mm or 100 mm per motor revolution using 1.8 degree motors.

dc42

@NL61 said in H-Bot Endstop problem:

@dc42
thanks for answering,
I tried M119 serveral times for each axis, it always shows if triggered or not.
the endstops are not swapped and the movement goes to the correct direction.

Can you confirm that the homex.g file looks exactly like the one you posted above when you open it in Duet Web Control?
If you try to home just X (by pressing the Home X button in Duet Web Control, or by sending G28 X), does movement stop if you trigger either the X or Y homing sensors manually?

Until you have homing working, you may wish to reduce the motor currents.

NL61

@dc42
once again the homex.g file:

; homex.g
; called to home the X axis
;
; generated by RepRapFirmware Configuration Tool v2.1.2 on Mon Nov 11 2019 14:44:06 GMT+0100 (Mitteleuropäische Normalzeit)
G91 ; relative positioning
G1 H2 Z5 F1800 ; lift Z relative to current position
G1 H1 X-395 F1800 ; move quickly to X axis endstop and stop there (first pass)
G1 X5 F1800 ; go back a few mm
G1 S1 X-395 F360 ; move slowly to X axis endstop once more (second pass)
G1 H2 Z-5 F1800 ; lower Z again
G90 ; absolute positioning

I tried to home X and triggered both, X and Y Endstop. No reaction

droftarts

Could you try swapping your NPN sensor for a microswitch, just to check it's working correctly?

Ian

NL61

@droftarts
hmm, the micro switch is working in X and Y.
The Sensors are MLM8-3002NA NPN, Schließer.
24 V and a BAT43 Diode at the signal output.
Why does the z-axis work? Its all the same setup.

I have to leave now. Thank you all for your answers, if there are any ideas whalt I can di to get those sensors working, I`ll try tomorrow.

dc42

Those sensors have a blue wire that must be connected to ground. Where on the Duet have you connected them?

NL61

@dc42
I connected 24V and Ground to the Power Supply.

NL61

I did some troubleshooting this morning. I connectet the X-Sensor to the Z-input and could stop the Z-axis with a piece of metal holding it to the X-Sensor. Same with the Y-Sensor. That tells me that the sensors work fine.
Then I connectet Z-Motor and Sensor to the Y-connectors on the duet. The LED triggered but the motor did not stop. M119 says Y-Axis homed.
Is it possible that there is a problem with the board?
Or does anyone have an idea?

dc42

Can you confirm that if you use a microswitch connected to the Y endstop input in place of the inductive sensor, the Y motor does stop when you trigger the microswitch?

If that's the case, then one possibility is that ground noise created by running the X and Y motors is affecting the input voltage at the endstop inputs. It may work better if you connect the blue wire to the ground pin on the endstop input, instead of to ground at the PSU.

NL61

@dc42
I connected microswitches to X and Y now, so I can run my first print. They are both working.
I already connected the ground wire to the input of the endstop but it also didn't work.

dc42

@NL61 said in H-Bot Endstop problem:

@dc42
I connected microswitches to X and Y now, so I can run my first print. They are both working.
I already connected the ground wire to the input of the endstop but it also didn't work.

If your Duet is revision 1.04, then you might get the inductive sensors working if you remove the diodes and connect the output directly to the endstop input. This disadvantage is that as you are using a 24V supply, if the ground wire to the sensor breaks then that will probably feed enough voltage into the endstop input to blow the indicator LED. The microcontroller itself is protected against excessive input voltage on the version 1.04 Duet.

Veti

@dc42

how does that correspond to the documentation here?
https://duet3d.dozuki.com/Wiki/Hardware_Overview

The Z probe input improved to tolerate inputs up to 30V.
The endstop inputs will tolerate at least 8V.

to me, this says do not connect the 24v on the endstops directly.

dc42

@Veti said in H-Bot Endstop problem:

@dc42

how does that correspond to the documentation here?
https://duet3d.dozuki.com/Wiki/Hardware_Overview

The Z probe input improved to tolerate inputs up to 30V.
The endstop inputs will tolerate at least 8V.

to me, this says do not connect the 24v on the endstops directly.

An NPN output sensor doesn't apply 24V to the endstop input, except that under fault conditions (ground wire disconnected) then it may apply something approaching 24V.

NL61

@dc42
I removed the diodes and the NPN Sensors are working properly now.
Thanks all for your help.