Endstops that are triggered during power off not recognized on power up.

Strange problem I can't seem to figure out. My machine will not recognize my Z end stop if its triggered during power up. For some reason when I home all my axis trigger the end stops and remain there, but if I power off and on the end stops are no longer recognized as being triggered. My understanding of Marlin is that its suppose to trigger an endstop than back off 1-2mm to untrigger. As of now I have to manually back off each axis before power down. I could hack around this with GCode but I'd rather know what's causing this behavior.

Using the latest Marlin no LCD.
I just switched electronics from a melzi to rambo1.3(melzi fried)
Everything else works flawlessly as long as I don't power down while any endstop is triggered.

If anyone can point me in the right direction I'd appreciate it greatly. Thanks!

The latest Marlin is a bad idea because it is full of bugs. If you don't have an LCD I would recommend my fork as it is known to work on Mendel90. If you want a more recent version that supports an LCD then I recommend Neil Darlow's fork, which is also known to work.

Ideally Marlin should have a stable branch but it doesn't at the moment and people are making sweeping changes to tidy up the code but that inevitably introduces bugs. Some serious like turning on all the heaters at start up!

---

[www.hydraraptor.blogspot.com]

Quote
nophead
The latest Marlin is a bad idea because it is full of bugs. If you don't have an LCD I would recommend my fork as it is known to work on Mendel90. If you want a more recent version that supports an LCD then I recommend Neil Darlow's fork, which is also known to work.

For that very reason I have not updated my fork for some months. Its present state is stable and I have implemented bed levelling with it (not for a Mendel90 but it is functional).

I will be porting my changes to a newer Marlin branch shortly but will not be updating the repository until I am sure it functions as well as the current code.

---

I try to write with consideration for all nationalities. Please let me know if something is unclear.
Printing with Mendel90 from fedora 25 using Cura, FreeCAD, MeshLab, OpenSCAD, Skeinforge and Slic3r tools.

Upon further inspection it seems I am running a fork #define STRING_CONFIG_H_AUTHOR "ErikZalm+nophead+neildarlow" // Who made the changes. Perhaps its a fault in the rambo board?

Configuration.h

I can't think of an electronics hardware fault that would stop it recognising the switches after power on, or stop it backing off. But if you are using Neil's firmware it doesn't have either of those faults on Melzi.

What does M119 report at the start when the endstops are triggered?

---

[www.hydraraptor.blogspot.com]

If I manually hold the Z axis endstop and turn on the printer it'll return all values as open. If I let go and M119 again it remains open and pressing it again will cause no change in reading. I've also replaced the endstop in case it was a faulty switch.

Actually it backs off and then hits a second time, so it does leave the end stops in the triggered state. If I power on and off M119 reports triggered. If I home it backs off and hits again.

On Mendel90 soft limits are 1mm from the endstops so after the first move all axes will move away from the endstops and never be able to hit them again except during homing.

---

[www.hydraraptor.blogspot.com]

My guess is you don't have internal pullups enabled. Melzi has external pullups but perhaps rambo doesn't.

Edited 1 time(s). Last edit at 03/30/2015 01:38PM by nophead.

---

[www.hydraraptor.blogspot.com]

I have endstoppullups defined in my config file, is that not internal pull ups? I'll have to go back through my config and see what its doing.

Yes that should enable them. I can't think of any other reason why they would read low when the switch was open instead of going high.

---

[www.hydraraptor.blogspot.com]

Note that when Marlin says "open" it actually means "not triggered", which is a closed switch. I reported this as a bug today.

---

[www.hydraraptor.blogspot.com]