Integer to Real

[ozzye]

Have a device sending me 2 integers one is the value and the other is the divisor. Would there be a compact way to divide this value not have to convert it to 10ths and display it as a real?

Thank you

 

[TConnolly]

What PLC?
What Display?

Did you mean that one integer contains the whole number and the other contains the fraction after the decimal point?

Or is it returning a 32 bit IEEE float as two 16 bit integers?

 

[ozzye]

PLC is compact logix and display is panelview plus.

 

[dmargineau]

For Logix plaforms I'd use the Modulo(MOD) instruction.

 

[ozzye]

So if source a 12345(value/integer) and source b is 2(divisor/integer) I would think the destination would equal 123.45(real). Getting a zero in the real.

 

[TConnolly]

Based on what you said, it looks like Source B is giving you how many points are after the decimal.

Divide 12345 by 10².

CPT DESTINATION_tag SOURCEA_tag/(10**SOURCEB_TAG)

If destination is a DINT you'll get the unrounded result 123 in this case. If destination is a REAL you'll get 123.45.

 

[ozzye]

Always dividing by 100, how could the divisor determine this?

Destination: GAUGE_PROFILES.PROFILE1_REAL

Expression: GAUGE_PROFILES.PROFILE1_MEASURED_VALUE / (100 ** GAUGE_PROFILES.PROFILE1_DIVISOR)

 

[TConnolly]

You are not always dividing by 100.

You are dividing by 10^N where N is the number in the second integer.

So if N = 0 then 10^N= 1 and you get 12345
If N = 1 then 10^N = 10 and you get 1234.5
if N=2 then 10^N = 100 and you get 123.45
If N=3 then 10^N = 1000 and you get 12.345
and so on.

If that isn't what you need then please supply more detailed information about what you are doing. I want to help you with the solution but I am still guessing about what it is you really want with the limited details you have provided so far.

 

[ozzye]

Ok, got it, typed in a 100 for 10**.

Destination: Real_Value_Tag

Expression: Int_Value_Tag / (10 ** Int_Divisor_Tag)

Thank you

 

[TConnolly]

Glad that was what you needed. Good luck.

 

[kittydog42]

Anytime you're dividing with a dynamic denominator, you should put a compare before the compute to ensure that it's greater than zero. If you don't you could end up with a processor fault that may be hard to find.

 

[TConnolly]

That is good advice. Its always good to trap for divide by zero whenever it is possible that it might happen.

In this case we are OK with skipping a divide by 0 trap since the divisor is always 10 raised to some power. There is no power to which ten can be raised that will equal zero. Put another way, Log₁₀0 is undefined.