I stick with integer arithmetic. In my particular case, I use a known=20
(standard) pressure, temperature, etc.) and record a half dozen readings=20
over the range to be used. Plug the readings into an Excel spreadsheet,=20
do a XY graph, and generate a formula. Generally a linear (y =3D mx +b) is=
=20
accurate sufficient. When moving the formula to the CPU ("C" or ASM,=20
etc.) need to use a little algebra to move terms around and scale (maybe=20
multiply by 100 and then when display/print) adjust the number to=20
prevent over/under flow. In your case, could enter the freescale=20
equation into a spreadheet, and then genreate the formula.

On 12/27/2011 6:03 AM, David wrote:
> I am using one of the Freescale pressure sensors, the application note
> for the I2C interface is below.
>
> http://www.freescale.com/files/sensors/doc/app_note/AN3785.pdf
>
> On page 15 it explains the coefficients used in the compensation
> algorithm.  They do also give sample code for the whole calculation, but
> it is a bit of a mess.
>
> How can I convert these 16 bit coefficients to floating point types in
> C?  I know this will drag in the floating point library (I use HiTech)
> but I am mostly interested in this for debugging, not long term use.
>
> David
>   =20
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.