How to transform 32-bit FLOAT number to 16-bit UNSIGNED

[zde92]

Hi there,

I am trying to write 32-bit single precision number to two 16-bit unsigned modbus regsiters. Any ideas how could that be done? 

thanks in advance!

[cbutcher]

How much precision do you need, and how do you want to use the value later?

[GerdW · Accepted solution]

Hallo zde,

maybe this is an inspiration to you:

[cbutcher]

I got this far, but yours is much nicer. I didn't realise before looking into this how miserable single-precision is...

[zde92]

For now, I don't have specific requirement for precision. I'am trying to do the modbus master/slave simulator. The SGL value will be written in input register of slave, then read with Read from input registers function on master. Since the modbus functions accepts the U16 format, the conversion is needed. 

[zde92]

Hi there, I tried something similar using type cast to transform to U32 and the used Split Number. But probably, I would need to create the opposite conversion on master where I read the values from register? 

[GerdW]

Hi zde,

I tried something similar using type cast to transform to U32 and the used Split Number.

So you get the same result using more functions!?

But probably, I would need to create the opposite conversion on master where I read the values from register? 

Yes, when you want to "simulate" this master too…