Basic question on wiring a digital input

[Deac99]

I've got a real basic question on how to wire a digital input.  All my sensors until now have just been dry contacts so wire from the 24V to the contact and from the other side of the contact to the digital input.

I now have a thru-beam sensor that is attached to a circuit board lets say pins 1 and 2.  When the beam is broken there is 24VDC between 1 & 2 where 1 is + and 2 is -.  When the sensor is not broken it goes to 0V

How would I hook that up to my digital input I was thinking I would simply connect pin 1 to my input and pin 2 to the common but that did not seem to work.

[Duncan]

you are right to connect 2 to the 0v and 1 to a digital input - can you see the input go positive using live view?

are you using a separate power supply for the beam sensor? you might need to connect the 0v of the beam device supply to the 0v of the loxone system in order to get a common 0v reference, although connecting pin2 should have already done that

[Deac99]

Thanks Duncan

I did some more testing.  When the wires are disconnected from miniserver and the beam sensor is blocked I indeed get 24V (and 0 when unblocked) on the wires that I would connect to the miniserver so I know my wires are OK

The power supply at the gate is a long distance from my cabinet (probably close to 100m of cable)

When I measure the voltage from the - on the wire to the - on loxone it is about 0.2V difference so I was thinking that would be insignificant.

When I connect the +24 from the sensor and measure to the - which is not connected to the miniserver I still get the 24V

As soon as I connect the - to the miniserver the voltage at my input goes from 24V to 3V.

Do you think this is due to different grounds?  And if so are you proposing that I connect a wire from the - of the 24V supply at the gate to the - of the 24V supply at Loxone?  Just want to clarify that part.

[Robbie]

This sounds like the Miniserver digital input is dragging the 24V down - which I wouldn't expect, as the input impedance of the digital inputs is 10kΩ. The impedance of your connecting cable, even at 100m long, should only be a fraction of that. Maybe test the impedance/resistance of your cable? Is there a minor break, poor solder joint, or faulty/corroded connector in the path causing a high impedance & resultant volt drop?

As the digital inputs need over 8V to register, as a temporary fix, you could use an analogue input, and set the threshold at, say, 1.5V. But really you need to determine why the 24V is dropping to 3V.

On Wednesday, October 18, 2017 at 3:41:28 PM UTC+1, Deac99 wrote:

[Deac99]

I just went out, disconnected the 2 wires from the sensor twisted them together and measured resistance with both wires disconnected at the miniserver.  17.6 ohms.

Then I connected the + to a different digital input on Loxone (just in case that input resistor was damaged, - to the - on loxone and again it goes from 24V to 3.3V

[Robbie]

Very odd!

All I can guess is that the your sensor's internal pull up resistor is of a very high value?

[Paulus]

It sound to me like the beam has an open collector and will need a pull up, what happens if you put a resistor, say 10K across the two wires (disconnected from Loxone) and measure the voltage across the resistor?

Have you got the specs for the beam?

[Deac99]

Interesting - I don't have any info on the sensor itself as it was delivered as sort of a black box.  I just followed the wires to the circuit board to find the output.  I disconnected the 2 wires from loxone, put a 10K resistor across those ends blocked the sensor and measured 7.1V across the resistor. 

This is were I get a little lost, not sure what that means.

Thanks for any input on this.

On Wednesday, October 18, 2017 at 11:28:55 AM UTC+2, Deac99 wrote:

[Duncan]

If it's open collector then the output should jump up to around 24v with the pullup when it's not triggered and drop to around 1v or less when triggered

Possibly bad beam controller?
If you can swap the input to an analogue input as already suggested you can use a greater than 3v or similar to detect a positive but it does sound like it's not working fully

[Deac99]

I'll go the analog route but this is a curiosity, thinking about it more, the sensor goes to a terminal block on a circuit board and the board has a relay which latches to stop the gate from closing if the beam is broken.  Maybe that relay is the culprit taking too much of the energy.