Concerning safety in the wiring stile, are 24VDC sink or source PLC outputs better?

[VinceS]

I have build using both sink and source PLC outputs. I have a preference for source outputs or relay outputs, and have always tried to stay away from sink outputs. My rational for avoiding sink outputs is that the output device could activate in the event of a short circuit.

Is my rational unfounded? Is there a standard concerning the use of sink PLC outputs?


Thanks in advance...

 

[GeorgeB]

I have build using both sink and source PLC outputs. I have a preference for source outputs or relay outputs, and have always tried to stay away from sink outputs. My rational for avoiding sink outputs is that the output device could activate in the event of a short circuit.

Is my rational unfounded? Is there a standard concerning the use of sink PLC outputs?


Thanks in advance...

I believe that sink outputs are electrically more efficient in conventional electronics; I can saturate the NPN transistor with internally available voltages where to saturate a PNP (sourcing) requires a base voltage a volt or so higher than the DC supply.

On the shorting issue ... valid, but I suspect that shorts occur more often in the device than in the field, and IN MY OPINION more frequently in the PNP than the NPN.

I have not done any hardware design in years, but NPN devices (at the same power levels) used to be slightly less expensive as well.

 

[ELA]

Wouldn't the safety level depend on other things such as:
Is the PLC power supply bonded and if so is it a negative or positive ground?
What is the fail safe state of the output?

I have also used both sinking and sourcing. In my experience the majority were sinking when using US devices. As we started using more European devices we experienced more sourcing devices.

We more often found PNP (sourcing) output drivers blown. Most often suspected from the result of a technician accidentally shorting the output to ground when - the design was the plc negative was bonded to ground.

If safety was of the utmost concern a safety PLC output might be better.
How about using two outputs. One a relay output that sourced the positive side and a npn output that sinked the return? Then maybe add a could of monitoring inputs (one for each output)?

 

[cornbread]

I prefer a relay output. I've have seen trouble with impeadance matching. In fact last week I had a problem where someone had replaced coil relays with solid state. The system worked fine until you lifted the common to the PLC output card, with out a reference the all the outputs on the card (feeding solid state relays) floated high, high enough to trigger the relay.

http://www.plctalk.net/qanda/showthread.php?t=47484&highlight=1746