Need help condemning breakout board

[billbyrd1945]

Need help condemning breakout board

I was changing out some limit switches on my homemade CNC machine. I goofed and failed to power down the system, I clipped the two wires going to one of the switches and got a spark. I looked over at my main panel and saw that all three drivers that normally have green leds, had turned red. I powered down and restarted to see all drivers now green and everything worked. Later in the day, the limit switches (all except the estop) quit working. by "quit working", I mean that each of the three axes could bump the switches without triggering them. It no longer mattered whether they were in their normally closed position or if I manually opened them.

Each of the four systems (x, y, z, estop) has its own circuit coming from the breakout board. When I disconnect the wire controlling the estop, I get the excepted alarm. Removing any of the other wires does nothing. I went ahead and ordered a new breakout board, but I would really like to be able to use my meter to confirm the short/open. All stepper drivers work fine. Only the limit switches are inop. These are the values I  have so far (see image). Can anyone tell me if these values confirm the breakout board as the culprit? Any help will be appreciated.

[Paul Moir]

Safe to assume the GND to switch common reading is 5v?  If so the X axis input being hauled up the 5v does look awfully suspicious.  It would be pretty normal for the E-STOP to be on a different circuit and so not involved since it shouldn't have logic elements involved.

If you post a picture of the front & back of the board I'm sure we can confirm the fault and perhaps ID the failed chip if you like.  There is something missing in the story since you shouldn't be getting a spark from clipping those two wires.  Were the cutters touching the frame of the machine?

[billbyrd1945]

Thank you for your reply. Last night I got to thinking that it would be simple enough to do a beep-continuity test off each limit circuit. I didn't do it earlier because it didn't seem logical to me that any one of them could be stuck in the open configuration without tripping a reset alarm. But I decided to check them anyhow. The culprit turned out to be-- the x+ switch had failed to remake after the gantry backed away from it. So it's an intermittent switch. Sometimes it works, sometimes it doesn't. I would have told anyone that if it failed to make up, it MUST trip a reset. But, not so. So I'm guessing that when it's stuck, it's still got enough continuity going through it so that it doesn't sound the alarm. It was clearly the problem.

[Gyro]

If they are standard cheap microswitches (no gold contacts) then one thing you could try is adding a 100nF capacitor directly across the switch contacts. The resulting small current spike as the contacts close can produce a 'self cleaning' action.

[shakalnokturn]

If they are standard cheap microswitches (no gold contacts) then one thing you could try is adding a 100nF capacitor directly across the switch contacts. The resulting small current spike as the contacts close can produce a 'self cleaning' action.

I believe (but could be wrong) that the capacitors commonly used across switch contacts are either used for mechanical "debouncing" or to extend contact life (as suggested) but by acting as a snubber to kill the arc when the switch opens.
When closing, of course the capacitor is extra power to dissipate for the contacts so perhaps that melts their surface and self cleans, I would have expected the opposite though.
As for the cheap switches for some reason they just seem to build their own layer of insulating black (oxide ?) without any abuse. Just avoid cheap switches...

Edit: 555'th post, need to celebrate that with some downtime and beer 

[billbyrd1945]

I've ordered magnetic proximity switches to replace them.