analog I/O

For the analog I/O module for the Allen Bradly 1200, I want to use voltage reading over amperage readings which seems to be what most people use.Is there a reason that current is used over voltage for analog I/O? difficulty? We are using 2 IR sensors to center an object for a robot manipulator. The IR sensors use voltage to determine distance. In a utopian world I want both sensors to read the exact same voltage in order to determine that the object is centered. I realize though, this is difficult to fine tune movement in order to get them equal. My question is are there tolerances that can be used in the compare functions? Like that IR one is equal to IR two within a small +/- tolerance. Is this possible?

I dont know rslogix but with other plc's there is exactly that.
It is called a zone-compare.


Something like; zone compare register for the value 90 to 100
(if less than 90 =false. if between 90&100 =true. if greater than 100 =false)

I do not know the syntax in this case though

Hi PLCGirl,

Current is used more often than voltage due to the fact that voltage will drop as cables become longer especially with very low voltages! this effect is seen to a lesser extent with a small current signal for example 4-20mA.

There is no reason why you cannot use a voltage signal with the system that you have, although depending upon the accuracy that you need I would allow for a slight difference between your sensors.

I would if I was you scale the inputs of the sensors to between 2 numbers, 0 and 1000 maybe (analog modules will normally give a number between 0 and 4000hex, 0 and 16384 decimal) and then compare the two numbers, this should itself give enough tolerance for you.

Regards

Barry

Well, yes there is!

For the analog I/O module for the Allen Bradly 1200

Click to expand...

I assume you mean the MicroLogix 1200 here and yes, you can use either current or voltage for inputs to an analog module. The reason current is often used over a voltage signal is to overcome the effects of distance. Voltage loss due to excess distance will degrade your sensor's accuracy. Also, current signals are far less affected by electrical noise sources than a voltage signal is.
So, now to the other question.

My question is are there tolerances that can be used in the compare functions?

Click to expand...

You can if you set up the proper code and one way to do that is with the LIM instruction. The LIMit test instruction lets you set as large or as small a +\- tolerance as you need to achieve your goal. If the LIM instruction is not available in the ML1200 then you can do the same thing by using a couple of rungs with > and < instructions some bit code in between.

Current analog signals, typically 4-20 mA, are typically used when the signal is in an electrically noisy environment or must be transmitted some distance. The 4-20 mA signal is more noise immune than a voltage signal, and can be transmitted longer distances. The 4 mA also gives a "live zero" signal: the control system is able to differnetiate between a zero process value (4 mA) and a broken wire (0 mA).

When I want to use compares with a dedaband or tolerance, I use five registers.

Tolerance
Nominal Value
Nominal + Tolerance
Nominal - Tolerance
Actual readhing

Compare your actual reading to the nominal + tolerance and nominal - tolrerance in two different compares. In your case substitute actual reading of one of the sensors for thre Nominal value, and you are all set.

Compare your actual reading to the nominal + tolerance and nominal - tolrerance in two different compares. In your case substitute actual reading of one of the sensors for thre Nominal value, and you are all set.

Click to expand...

If I understand your logic correctly Tom, that is quite clever.

Are you saying, you are using two compares, one looking for 'greater than' to set a bit and the other looking for 'less than' to set another bit. Then using the two bits in series to say it is between the tolerances.

I suppose I am lazy in that the main plc I use has this function already - but I like that logic.

Others have addresses 4-20mA vs. 0-10 Volts. If you want more, just search on this site on analog or 4-20 and you should get enough reading material for several hours.

This is a classic "A minus B" measurement. You should probably start by subtracting one of the values from the other with the SUB instruction.

Send the number through a LIM instruction to give you a GO/NOGO output for if the object is in tolerance (you will have to scale the signal to find the numbers that correspond to your tolerance for the upper and lower limits you need).

---------------------------+--+---------+ 

                           |  |       MEAS1  |

                           |  |       MEAS2  |

                           |  |     MEAS_DIFF|

                           |  +--------------+

                           |

                           +--+-[LIM]--------+   IN_TOL

                              |      LOW_LIM +-----( )---

                              |    MEAS_DIFF |

                              |     HIGH_LIM |

                              +--------------+

.

_{.
If you need to know which direction the object is offset, use GRT or LES to determine the sign of the difference. The sign will tell you which direction the offset is.}

Well, Goody, if it isn't simple I can't do it! You have the logic correctly interpreted.

One of the things I like about doing it this way is it lets me set a flag coil for "Too High" or "Too Low" that can be used in my logic, subsequent math (I usually use real integer math on the Koyo's) and so on. I can also latch the compare until a certain event is completed, etc.