Hello Kees, thank you for your answer.
Yes, exactly. In the 2nd picture, it should read a 16 bits data for register 3059. That is exactly what I want to do and it works good. But if I try exactly the same thing with the register 30827, it returns an error (value 32768).
However, I think I found why I have this behaviour. I think you can help me finding a solution :
The power meter i am connected to respond with 32768 if you try to read the two registers with 16 bits separately instead of the 32 bits at once.
In python, I pass the first adress, then another input to say it is register oflength 32 bits and then I do not know what it really does but it works good. If i ask the same adress and tells the lenght is 16 bits, it returns the same value error 32768 than what I get in Labview (instead of the first 16 bits of the answer i get when I asked the 32 bits at once)
In Labview, I call register by register so 30827 and 30828 separately. And then only when I have the values in each register I convert them from haxadecimal to float32.
What is weird in the equipment behaviour is that does work good with this method on some "32 bits". register but returned an error 32768 for other "32 bits" register.
So what I have to do to fix the problem is to ask the 32 bits "all in once" and not 16 bits twice. Is there a way to do this in Labview ?
My connection is TCP / Modbus with an ethernet cable.
Thank you
