Apprentice here, sorry if this is basic. Replaced an inductive prox on a cylinder end stop yesterday and the only one left in stores was a 2 wire type. The old one was 3 wire I think, it went in the bin before I looked at it properly.
PLC is a 1214C DC/DC/DC and the sensor is on I0.3. Input stays on with nothing in front of it. I put the meter across the input terminal and M with the target away and got 6V, so I moved the sensor over to I0.4 and it did the same, then I tried a second sensor out of stores and that did the same as well.
One of the sparkies says the board's damaged. Spec says an input needs 15VDC to be on and mine is sitting on at 6. Two inputs and two sensors all sat at 6V though, would a damaged board really do that on every one of them?
Two numbers. What's the sensor's off-state leakage on its datasheet, and what does the CPU spec say for a guaranteed 0 on that input, voltage and current? Both are on the sheets.
Sensor datasheet says leakage current 1.5mA max. CPU manual table says 0 state max 5VDC at 1mA, 1 state min 15VDC at 2.5mA, though I'm not sure I'm on the right table.
So I don't get how it's on at 6V when on needs 15.
A 2 wire sensor has no separate supply. It powers itself through the load, your input, so even when off it passes a small current to keep its electronics alive. That is the 1.5mA, and the CPU only promises a 0 under 1mA. Through the input's impedance that current makes a voltage, 6V in your case. Above the 5V the CPU calls a guaranteed 0, below the 15V it calls a guaranteed 1, so it reads whatever it likes. Yours reads 1.
Fix: a bleeder resistor from the input terminal to M, something like 4.7k, 1W, check the dissipation. The leakage gets another path and the input drops under 5V. Or a 3 wire sensor, no resistor.
The input types and thresholds are in here: https://plctr.com/understanding-the-basics-of-input-output-modules-in-automation/
ok. So the sensor is fine and the input is fine, they just don't suit each other, and the resistor gives that leakage somewhere to go so the input sees less voltage. Have I got that right?
Right. Fit the resistor and measure again.
Fitted a 4.7k 2W across I0.3 and M. Target away, 2.9V, input off. Target in front, 22V, input on. Cylinder sequence runs.
So it was the sensor's leakage sitting above what the CPU is willing to call a 0, and the resistor gives that current somewhere else to go. I lost a shift to the wrong sensor out of the wrong bin.
What nobody can tell me is why stores has 2 wire sensors when every input on this machine was 3 wire. Thanks marek, ed.
Stores buys what's cheap. Put a note on the bin.