Long distance pulse signal to Plc

[123MB]

Hi All

We have materials handling application with a run of 500m maximum out to a photoelectric sensor/photo eye which is being used as an underspeed sensor. The sensor will be wired from the field to a PLC input via 30 core SWA PVC/PVC control cable untwisted and unshielded.

The sensor will detect the rotation of the non-driven pulley of a conveyor and be used to detect a problem with the conveyor where the motor might be running but the conveyor isn't moving.

The pulse frequency we are expecting from the photo electric sensor is a maximum of 4Hz.

What I am concerned about is the capacitance between cores in this cable, and from core to armour, effectively slowing the rise time of our pulses and causing problems.

I am trying to find capacitance values for the 30 core SWA PVC/PVC cable to calculate rise times, etc, but manufacturer's dont publish this data for normal PVC/PVC control cables.

Also I am having a hard time understanding the mechanisms for determining the capacitance. For example:

1) I am happy that the capacitance from one core to others increases with the number of cores, but how to quantify this based on data for core to core capacitiance? Consider the capacitances in parallel?

2) Does the presence of armor make the capacitance worse?

The closest data I have found for a 30 core SWA PVC control cable has a capacitiance value of 380nF/m which is very very high and I am not sure it is accurate

Sorry for the long post guys I hope someone can help out and that the responses help others in future

Cheers

 

[VEBill]

One option in the worst case would be to put the underspeed decision (a controller) next to the sensor. Then just send the decision (underspeed) over the longer distance. A possible solution in the worst case.

This assumes a constant decision point.

 

[IRstuff]

Seems to me that you're not caring whether the pulses get attenuated or are misshapen, just their presence? Any sort of RS-422 differential driver should be able to handle that distance at that frequency.

TTFN
I can do absolutely anything. I'm an expert!
faq731-376 forum1529

 

[djs]

There are available photo cells with built-in timers which will do the missing pulse detection for you. No timing issues there. I would be more concerned about signal interference, induction pickup etc. over a 1500 ft run.

 

[snarkysparky]

380 nf/m does sound too high. Buy a meter of this cable and use a multimeter with capacitance measuring ability.

If the sensor is 24V out I would think it could drive this 500 m cable at 4 hz without problem. Determine the output drive capability of the sensor in milli ampere and calculate the time to charge the cable with it using I= C(Dv/Dt)

 

[benta]

For distances as long as this, I'd be inclined to use RS-422 over optical fibre. More expensive yes, but not prohibitively so, and much more reliable.

Benta

 

[123MB]

Hi Snarkysparky, I would love to do that but I don't really have the means (the cable is a long lead time item and we need to have the design all locked away long before the cable arrives)

I agree that 380nf/m sounds high

Since I wrote this post I've found that most proximity switch manufacturers say that you can extend the wiring 200m max - their reasons are:

1) Excessive capacitance increasing rise/fall times (they quote a maximum value of 20-100nF)
2) Too much capacitance in the output causes the capacitance to be discharged through the transistors which is an accumulative heating affect which eventually damages the electronics
3) Volt drop, interference

So I am leaning towards a proxy with the built in evaluation of the under speed, as has been suggested.