PLC DI card wiring

Status
Not open for further replies.

fireryan

Senior Member
Location
Minnesota
IMG_0360.PNG


How many of you PLC guys are familiar with this kind of DI wiring? Im having a hard time wrapping my head around this. What is the advantage of dong it this way instead of just having a 24v input back. Also cant figure out how the diode and resistor stop this from being a direct short
 
Surprised you did not get more replies.

Usually a DI input is looking externally for an external dry contact closing or a transistor switch. I would expect a DI would not use an external power supply, I would expect trouble. The DI is powered internally usually.

Page 17 looks like DI inputs, page 18 looks like a specified (24 v) voltage input. The diode is an LED driving an optoisolator circuit at the input, the resistor probably ballasts the LED current.

Some more clarification from, the manufacturer appears to make both types.
 
170204-2011 EST

fireryan:

The drawing on page 18 is very clear.

The diode shown is the LED inside the optical coupler. The resistor, 750 ohms, is in series with the LED. The LED voltage drop plus the resistor drop determines the current thru the LED.

If we assume a 1.5 V drop across the LED, then from the 24 V source shown the current will be (24 -1.5) / 750 = 30 mA. This is a high driving current for typical optical coupler. But, note this IO module is designed to respond to 50 kHz.

I don't know if it was mentioned somewhere, but if reverse polarity was applied, then you would likely destroy the LED. Possibly a reverse biased diode is shunted across the LED and not shown. If this existed, then reverse polarity would not damage the LED.

What is the advantage of dong it this way instead of just having a 24v input back.
I don't know what you mean by "input back". This type of input provides DC isolation of your input, up to 1500 V, from the PLC internal logic common and thus likely from the system EGC.

.


 
Last edited:
The only advantage of using one of these is the isolation. Otherwise think of it as regular input, you can still use the same power supply that the PLC uses.
You can still tie all your commons together and the input card still switches on a high signal. Instead of running just one wire to each input, now you need both.
You can just run common to the card and jumper that to all the common terminals and then it will seem like any other DI card.

Now say you need to monitor inputs from machines with isolated power supplies. Normally you would need to use a relay to provide an input to your PLC, because
wiring direct may cause problems due to differences in voltage potential. With this DI card you are already isolated, there is no electrical connection. Instead
it is optical. So you would just wire it straight in to your card.
 
170204-2011 EST

fireryan:

The drawing on page 18 is very clear.

The diode shown is the LED inside the optical coupler. The resistor, 750 ohms, is in series with the LED. The LED voltage drop plus the resistor drop determines the current thru the LED.

If we assume a 1.5 V drop across the LED, then from the 24 V source shown the current will be (24 -1.5) / 750 = 30 mA. This is a high driving current for typical optical coupler. But, note this IO module is designed to respond to 50 kHz.

I don't know if it was mentioned somewhere, but if reverse polarity was applied, then you would likely destroy the LED. Possibly a reverse biased diode is shunted across the LED and not shown. If this existed, then reverse polarity would not damage the LED.

I don't know what you mean by "input back". This type of input provides DC isolation of your input, up to 1500 V, from the PLC internal logic common and thus likely from the system EGC.

.



I'm referring to a switch closing and providing the input signal back to the card like on a standard input.
 
The only advantage of using one of these is the isolation. Otherwise think of it as regular input, you can still use the same power supply that the PLC uses.
You can still tie all your commons together and the input card still switches on a high signal. Instead of running just one wire to each input, now you need both.
You can just run common to the card and jumper that to all the common terminals and then it will seem like any other DI card.

Now say you need to monitor inputs from machines with isolated power supplies. Normally you would need to use a relay to provide an input to your PLC, because
wiring direct may cause problems due to differences in voltage potential. With this DI card you are already isolated, there is no electrical connection. Instead
it is optical. So you would just wire it straight in to your card.

What is the advantage to the dc isolation
 
The led is turned on when the switch closes
it emits light (photons)
these strike a photo cell that changes resistance when light hits it
no light = high
light = low
acts like a switch

complete electrical isolation
a transformer provides the same with electromagnetic force
this with light
 
The led is turned on when the switch closes
it emits light (photons)
these strike a photo cell that changes resistance when light hits it
no light = high
light = low
acts like a switch

complete electrical isolation
a transformer provides the same with electromagnetic force
this with light

Ok now I understand
 
Typically you don't see this used on a PLC, because typically a PLC is used on a machine and is relatively close at hand, so the wire length is not an issue. But that manual is for an Emerson DCS (Distributed Control System), not a PLC, which generally means it is not controlling a single machine, it is controlling an entire processing plant of some sort. So when you need specific inputs to a DCS from a device that is maybe thousands of feet (or even yards) away, voltage drop and exposure to damage becomes an issue. So if you attempted to use the same power supply that the cards and processor use inside of your panel and something goes wrong in the circuit, you lose everything connected to that power supply, which in a processing facility can mean a major disaster. So what you do is put a power supply out in the field with the sensor so if something happens, you only use that one sensor.

But now what you have done is introduce a risk of having noise or voltage problems in that "foreign source" sensor circuit bleeding into your entire controller environment causing disruption, errors and even failure. So by having that foreign source isolated through an LED optocoupler, there is no direct electrical connection of that foreign source and the control system power.
 
Typically you don't see this used on a PLC, because typically a PLC is used on a machine and is relatively close at hand, so the wire length is not an issue. But that manual is for an Emerson DCS (Distributed Control System), not a PLC, which generally means it is not controlling a single machine, it is controlling an entire processing plant of some sort. So when you need specific inputs to a DCS from a device that is maybe thousands of feet (or even yards) away, voltage drop and exposure to damage becomes an issue. So if you attempted to use the same power supply that the cards and processor use inside of your panel and something goes wrong in the circuit, you lose everything connected to that power supply, which in a processing facility can mean a major disaster. So what you do is put a power supply out in the field with the sensor so if something happens, you only use that one sensor.

But now what you have done is introduce a risk of having noise or voltage problems in that "foreign source" sensor circuit bleeding into your entire controller environment causing disruption, errors and even failure. So by having that foreign source isolated through an LED optocoupler, there is no direct electrical connection of that foreign source and the control system power.

This is actually the first time I've come across a Dcs system around here. All of the other plants around here are using plc to control entire processes in the plant. Is this a newer type of control?
 
No, actually older from the standpoint of plantwide control. When PLCs first came out, they were relay replacers for the most part, the first 2 or 3 generations were not capable of doing floating point math, so they could not do PID loop controls. Even after that capability was added, you could do a few loops before the number crunching would start to interfere with the logic execution time. DCS systems were built on dedicated large computers that could handle thousands of PID loops and provide a graphical display, long before PLCs were capable of that kind of processing power. The lines are definitely blurred now however, so the days of dedicated DCS hardware are probably numbered because PLCs can be used as part of a DCS, which is how A-B and Siemens do it, but it's a lot harder to make a DCS system control a single machine, at least economically.
 
No, actually older from the standpoint of plantwide control. When PLCs first came out, they were relay replacers for the most part, the first 2 or 3 generations were not capable of doing floating point math, so they could not do PID loop controls. Even after that capability was added, you could do a few loops before the number crunching would start to interfere with the logic execution time. DCS systems were built on dedicated large computers that could handle thousands of PID loops and provide a graphical display, long before PLCs were capable of that kind of processing power. The lines are definitely blurred now however, so the days of dedicated DCS hardware are probably numbered because PLCs can be used as part of a DCS, which is how A-B and Siemens do it, but it's a lot harder to make a DCS system control a single machine, at least economically.

We use GE PLCs here with a combo of Delta-V and Bailey for our DCS. As confusing as it is having both controlling systems as long as PIDs and 4-20 type devices are on the DCS and digital/motor control is on the PLC it makes it a little easier I think.
 
Status
Not open for further replies.
Top