Stupid endstop question kick me if necessary

Okay. I3 rework with makerbot 1.2 pbc enstops. I have had them connected as normally open and working fine, the LED's work they trigger correctly and I have been running this way for months printing fine. But I am on a safety kick now so I want to configure them as normally closed, for fail-safe operating.

Can these board be setup normally closed, which wires go to which pins on ramps and what do I change in Marlin?

I currently have them connected to y-min and x-max on ramps 1.4 with the signal pin from the endstop board to signal on ramps and the ground pin on the endstop board to ground on ramps. I have set in Marlin disable pullup for these two endstops (as I believe they have built in pullup resistors) and set logic to false. However they show as triggered when they are physically open.

If I reverse the logic they work correctly but if I disconnect a board to check for fail-safe operation they do not show as triggered.

What blindingly obvious stupid thing am I doing wrong?

it would need to be a modification on the endstop board. Do you have the 4 pin or 3 pin version? it should be as simple as moving the NO trace to the NC pin on the switch. I think that will also reverse the logic on the LED, so it would be off when triggered. but there are a few iterations of that board, so it's not a cut and dried answer.

Do NOT just try this as it may short out your ramps and destroy the arduino.

post a picture of your endstop board, maybe front and back.

Here is the endstop board, its a 4 wire version but both central pins are GND.

[onedrive.live.com]
[onedrive.live.com]

I've tried every which way with this board, as far as I can see the +5v on the right (as you look at the front of the board) is connected to NC on the switch, also goes via a resistor to signal, and via another resistor to the LED. There is also a capacitor from GND (both middle pins) to the signal. Ground is connected to NO on the switch, and signal is connected to COM on the switch.

I've tried connecting just the signal and GND with and without pullups enabled with logic for this endstop to false. I get a triggered response to m119 when it's physically open and vice versa.

I've tried removing the r1 resistor and only using the internal pullups, and I've removed the capacitor as I cannot see what it is likely to be there for. This does not help.

The only time it behaved partially like a NC endstop is when I had internal pullup off and logic set to false, if I disconnect the endstop it shows triggered, however when connected it shows triggered when its physically open and vice versa so no use.

Based on this [www.sainsmart.com] I also reversed the com and signal pins in the connector, this also does nothing it just behaves like a NO switch, except when you connect 5v then it shorts the ramps and causes a usb coms failiure. Needless to say I only did this for a fraction ofa second.

So for now its back to NO endstops which work, but I am always reading that we should all use NC for safety. Perhaps I ought to just make some endstops without fancy PCB's and LED's and have done with it.

Simple solution just use some plain microswitches configured to NC and everything works fine, not sure why these little PCD boards exist except that they have an LED on them.

Quote
DjDemonD
Simple solution just use some plain microswitches configured to NC and everything works fine, not sure why these little PCD boards exist except that they have an LED on them.

NC switches are what most people use (including me), and if they are of reasonable quality and used correctly, they are reliable and give highly reproducible results. Some people use opto or Hall effect endstops instead, but it isn't clear that they give better results. Opto endstops may be affected by sunlight (which contains a lot of IR if there are no clouds in the way), and Hall effect endstops may be affected by the Earth's magnetic field and hence the orientation of the printer. I have never seen any discussion or measurements of these effects reported.

In practice, reproducibility of endstop switches isn't usually very important. In a Cartesian printer, the reproducibility of X and Y endstops doesn't matter unless you pause a print and re-home before restarting it. The reproducibility of the Z endstop (if you have one at all) doesn't matter if you use a Z probe either to perform auto bed compensation or just to set Z=0 before printing.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thanks, I have no problems with mechanical endsotps they've worked fine for me for a long time. I do not rehome during printing so reproducibility isn't high on my list. My frustration was that the pcb endstops seemingly cannot be configured to give a NC output and therefore I wonder why they seem to be so popular, their only saving grace is that they have an LED on them and do not require any soldering.