Question about normally open vs normally closed endstops.

Is there any reason to choose using normally open vs normally close endstops?

It seems the typical method is to use them as normally closed, so they open the circuit when touched. But isn't that wasting electricity, even if only a tiny trickle, having it constantly flowing through the switch.

I'm using a Sanguinololu board, and according to the wiki it wants them to be normally open. This seems like a more efficient method. Racking my brain trying to figure out why this isn't standard I'm just guessing maybe for fault detection? Like if a switch goes bad, it will always show open, even when its touched, whereas having them in normally closed configuration, if it goes bad and opens up, you would know right away its open? Just a random guess.

Can anyone shed some light on this, or explain if I am missing something?

The reason usually is because, if one of the switch wires breaks or falls off, the motor will not keep trying to drive through it (or something like that). Having said that, I have never used that configuration coz' I use reed switches - never had a problem as they are so reliable.

---

_______________________________________
Waitaki 3D Printer

Waitaki is spot on with failing to a safe state vs allowing the printer to potentially damage itself.

Quote
scubaguy
But isn't that wasting electricity, even if only a tiny trickle, having it constantly flowing through the switch.

Depending on the particulars of your electronics, the current is extremely small. A MEGA 2560 board for instance has pull-up resistors of between 20k and 50k ohms, so .25-.1 milliamps per end stop. It's less than a rounding error in the scheme of everything else with your printer.

This is really a holdover from larger scale industrial machines common control design practice (and makes sense at this scale as well). If a limit switch fails on a punch press that handles sheets of 12 ga steel, someone can get seriously hurt (not to mention property damage) if the machine does not default to a halt state when a sensor defect occurs.

Saving some power is of little concern if you keep an employee from being cut in half if a cable gets damaged.

Just for the record, it's more than just concern about the limit switch failing. The connections, wiring, circuit board traces and internal circuitry associated with the limit switches could fail as well, causing loss of limit protection. These additional failure points will almost always fail to the open condition, so if you use a normally open configuration the system will interpret these failures as simply the travel limit not yet being reached. With a normally closed configuration any of these failures result in immediate axis stop.

It's good to have confidence in your switches, but it's wise to create a setup which protects against other limit protection system failures as well.

I assumed (my bad) that by "limit switch failing", readers would understand I meant EVERY component required to get the limit switch signal detected correctly. Connector and wire integrity is much more likely to be a point of failure than the switch actually going bad. I should have been more clear with my wording.