Creating a zeroing function

[Walkerok]

I realize that this is not strictly a PLC question, but we have some awfully experienced people here that might be able to keep me out of trouble. I have a customer that uses an extremely low range "guage" pressure transducer +/- 30" WC (or +/- .0294 Bar) used to control a process with operating control points between (-0.5) to 6" WC. Because this sensor is so low in range as you might imagine there is a real and ever present possibility to change the Zero point of the pressure transducer from process realities that permenantly deform the sensing disk of the sensor. The customer want the ability to place this sensor into atmosperic conditions and call whatever the sensor sees as the zero position.

The way I look at it I have a choice. Since I am making calculations to put the sensors raw values into engineering values, I can either shift the high and low ranges of the sensor depending on the zero up front or I can wait till the end of the calculation and then just make a positive or negative shift to make it show zero on screen and use this shifted value to also be what the program looks at to make the control decisions.

What I am looking for is if anyone has done this, what are the pit-falls (if any) do you see to doing the zeroing either way, or am I looking at this all wrong and need to rethink how to do the function in general?

I am thinking that if I do it at the end I can write in a comparision of the total shift to what zero should be and then I can prevent the user from using what is effectively a broken sensor by only allowing a maximum shift from where the zero point should be, but I do not want to creat new problems.

Comments would be much appreciated.

Thank you in advance

Keith

[Damian]

I realize that this is not strictly a PLC question, but we have some awfully experienced people here that might be able to keep me out of trouble. I have a customer that uses an extremely low range "guage" pressure transducer +/- 30" WC (or +/- .0294 Bar) used to control a process with operating control points between (-0.5) to 6" WC. Because this sensor is so low in range as you might imagine there is a real and ever present possibility to change the Zero point of the pressure transducer from process realities that permenantly deform the sensing disk of the sensor. The customer want the ability to place this sensor into atmosperic conditions and call whatever the sensor sees as the zero position.

The way I look at it I have a choice. Since I am making calculations to put the sensors raw values into engineering values, I can either shift the high and low ranges of the sensor depending on the zero up front or I can wait till the end of the calculation and then just make a positive or negative shift to make it show zero on screen and use this shifted value to also be what the program looks at to make the control decisions.

What I am looking for is if anyone has done this, what are the pit-falls (if any) do you see to doing the zeroing either way, or am I looking at this all wrong and need to rethink how to do the function in general?

I am thinking that if I do it at the end I can write in a comparision of the total shift to what zero should be and then I can prevent the user from using what is effectively a broken sensor by only allowing a maximum shift from where the zero point should be, but I do not want to creat new problems.

Comments would be much appreciated.

Thank you in advance

Keith

I would use option 2 (shifting after the fact). It is probably the least amount of code, and won't goof up your scaling. It also, as you stated, gives you a good baseline in regards to deviation from calibrated zero.

That is a pretty typical method of "Taring" your reading.

[Walkerok]

Thank You very much for the input. Thats helps me put my mind to rest on this subject.