Calibration of PLC Timer??

daithi1979

Member
Join Date
Sep 2011
Location
Limerick
Posts
11
Hi All,

The company I work for has recently been audited and one of the sitings was that we are not calibrating our PLC timers. The question posed by the auditor was "If a process is controlled at a temp for x time how do we know the timer is correct"

I am having difficulty finding detailed information explaining that there is no standard to calibrate a PLC timer and also that PLC timers are stable. Has anyone come across this situation before?

Common sense is out the window here, looking for solid data to reply to the auditors

Many Thanks,
David.
 
What is your process that requires an auditor to ask this question?. To what level of calibrated time does he want? as you will struggle with most conventional PLC's as they are scan based (info available in manuals) unless the PLC clock can be corrected by a calibrated input, i.e atomic clock or GPS.
 
The auditor wants to know if the time of the temperature is within limits.
The easiest way to be sure is by logging the date/time and temperature.
for example milk needs to be sterialized for 300 seconds on 120 celsius.
So when it is a batch proces the time starts when the temperature is reached and stops when it falls below.
They have a separate unit to have this in a datalogger for audit.
So you can have it in the PLC but it must be logged on paper or on file.
 
I think you most probable nailed it Shooter, but logging time still doesn't calibrate the timer. Yet another post we can all play the guessing game.
 
The auditor wants to know if the time of the temperature is within limits.
The easiest way to be sure is by logging the date/time and temperature.
for example milk needs to be sterialized for 300 seconds on 120 celsius.
So when it is a batch proces the time starts when the temperature is reached and stops when it falls below.
They have a separate unit to have this in a datalogger for audit.
So you can have it in the PLC but it must be logged on paper or on file.

Thats an idea but the problem is like the quote:

"The man with one watch always knows what time it is, the man with two watches never knows what time it is"

In the above case if the datalogger says that it was on for 290seconds and the PLC for 300seconds which one is wrong?
 
U: it is the datalogger that is correct as that one is calibrated in temperature and in time. even when this time should be wrong it is documented how much and that is certified, the time in the PLC can be changed due to inner processes and program.
It is very difficult to certify a plc program.
 
In a normal PLC you usually have a clock and timer functions, with the clock is easy to confirm calibration by comparing with an NTP server. The timers are harder to calibrate because it depends on so may things within the PLC. So you could change your timers so that you instead use the clock to get timings, or use the clock to confirm the timer operation. So at the start of a timing I would use a rising edge to store the current clock time, at the end of the timing use the falling edge to store the current time, then subtract one from the other, compare with the timer setpoint and you can confirm the timer operation. Through the day check PLC time against NTP time to confirm the clock.
 
I know of a system where an auditor questioned the reliability of a plc timer. There are several PLC's networked together.

The ridiculous solution was to run a second timer in a different plc and wait until both timers had elapsed.

Auditor was satisfied with that because it had redundancy of the timer.

IMO all PLC 'timers' in modern systems are 'type tested' by the manufacturers and most process don't need millisecond accuracy
 
Back to the OP's first post, calibration infers that adjustments can be made to match a certified standard. How can a PLC timer be adjusted so that a programmed 1000 millisecond timer will always time to 1000 +/- x%? Or if I change the preset to 3000 will it be 3000 +/-x%. So now, the linearity of its output is a factor that has to be certified. Its not something that PLC manufacturers made available and as pointed out, there are many factors to consider. So an external timing device that can be calibrated is really the only true solution.
 
Best solution is to use the real time clock in the plc and use the nts service to synchronize the clock to a known standard. I would also use a standard timer as a check. That gives you two independent timers
 
Back to the OP's first post, calibration infers that adjustments can be made to match a certified standard. How can a PLC timer be adjusted so that a programmed 1000 millisecond timer will always time to 1000 +/- x%? Or if I change the preset to 3000 will it be 3000 +/-x%. So now, the linearity of its output is a factor that has to be certified. Its not something that PLC manufacturers made available and as pointed out, there are many factors to consider. So an external timing device that can be calibrated is really the only true solution.

THIS.

Go back to the auditors as ask them to clarify their definition of "CALIBRATION", and then provide the definition of "VALIDATION", and proceed to tell them that calibration of PLC timers is not possible, however validation of them is by testing against a known source...IE hand-held stopwatch.
 

Similar Topics

Hi everyone, i hope you are doing well i am using fbs 2lc module , i have calibrated the weight by calibrated weight but my value is not...
Replies
0
Views
85
Hello everyone, Just for general knowledge, do you think PLC analog input & output module/point needs to calibrate using external calibrator?
Replies
17
Views
14,987
Omron AD081-V1 Analog Input Card Offset & Gain Adjustment Entering Adjustment Mode 1. Set the input card in adjustment mode (Turn ON Dip SW No-1)...
Replies
0
Views
87
Hello, I have a problem calibrating siwarex ms with siwatool ms Whatever I have tried, I have communication error and numerous retries to...
Replies
0
Views
74
5069 IF8 with a 4-20ma 0-100psi, sealed fixed, no calibration, transducer attached. IF8 is scaled 4ma= 0 psi, 20ma=100psi. If over time the...
Replies
12
Views
2,608
Back
Top Bottom