Data conversion

[dms]

Hello to everyone I need some help here.

I've got a 2 byte data in this form:

bit15: Sign
bit14: 2^6
bit13: 2^5
bit12: 2^4
bit11: 2^3
bit10: 2^2
bit09: 2^1
bit08: 2^0
bit07: 2^-1
bit06: 2^-2
bit05: 2^-3
bit04: 2^-4
bit03: 0
bit02: 0
bit01: 0
bit00: 0

I want to convert a 2 byte data in this form to a float point data and save it in a variable. I need a calculation, if you want to know, this is related to the digital sensor DS1631.

Help me help me

thank you
___________________________
This message was ported from CCS's old forum
Original Post ID: 11287

[R.J.Hamlett]

:=Hello to everyone I need some help here.
:=
:=I've got a 2 byte data in this form:
:=
:=bit15: Sign
:=bit14: 2^6
:=bit13: 2^5
:=bit12: 2^4
:=bit11: 2^3
:=bit10: 2^2
:=bit09: 2^1
:=bit08: 2^0
:=bit07: 2^-1
:=bit06: 2^-2
:=bit05: 2^-3
:=bit04: 2^-4
:=bit03: 0
:=bit02: 0
:=bit01: 0
:=bit00: 0
:=
:=I want to convert a 2 byte data in this form to a float point data and save it in a variable. I need a calculation, if you want to know, this is related to the digital sensor DS1631.
:=
:=Help me help me
This is just an integer, scaled /256.
Assuming your two bytes are held in a variable declared as a signed integer - if not, use a union, and have a signed int16 as one of the methods of accessing the data like :
union data {
int8 b[2];
signed int16 w;
} value;

Then you can access the individual characters as:
value.b[0], and value.b[1], and the whole 16bit word as value.w

Then just code as:
result=(float)value.w/256.0;

This 'casts' the integer into a float, and then divides this by 256 to give the correct result. However note that I am _forcing_ the operation to be done on a float, by using a float constant, and doing the conversion before the arithmetic (CCS C, has a tendency to try to save time, by converting latter...).

Best Wishes
___________________________
This message was ported from CCS's old forum
Original Post ID: 11289

[dms]

:=:=Hello to everyone I need some help here.
:=:=
:=:=I've got a 2 byte data in this form:
:=:=
:=:=bit15: Sign
:=:=bit14: 2^6
:=:=bit13: 2^5
:=:=bit12: 2^4
:=:=bit11: 2^3
:=:=bit10: 2^2
:=:=bit09: 2^1
:=:=bit08: 2^0
:=:=bit07: 2^-1
:=:=bit06: 2^-2
:=:=bit05: 2^-3
:=:=bit04: 2^-4
:=:=bit03: 0
:=:=bit02: 0
:=:=bit01: 0
:=:=bit00: 0
:=:=
:=:=I want to convert a 2 byte data in this form to a float point data and save it in a variable. I need a calculation, if you want to know, this is related to the digital sensor DS1631.
:=:=
:=:=Help me help me
:=This is just an integer, scaled /256.
:=Assuming your two bytes are held in a variable declared as a signed integer - if not, use a union, and have a signed int16 as one of the methods of accessing the data like :
:=union data {
:= int8 b[2];
:= signed int16 w;
:=} value;
:=
:=Then you can access the individual characters as:
:=value.b[0], and value.b[1], and the whole 16bit word as value.w
:=
:=Then just code as:
:=result=(float)value.w/256.0;
:=
:=This 'casts' the integer into a float, and then divides this by 256 to give the correct result. However note that I am _forcing_ the operation to be done on a float, by using a float constant, and doing the conversion before the arithmetic (CCS C, has a tendency to try to save time, by converting latter...).
:=
:=Best Wishes


Works fine!


THANK YOU THANK YOU
___________________________
This message was ported from CCS's old forum
Original Post ID: 11292

[R.J.Hamlett]

:=:=:=Hello to everyone I need some help here.
:=:=:=
:=:=:=I've got a 2 byte data in this form:
:=:=:=
:=:=:=bit15: Sign
:=:=:=bit14: 2^6
:=:=:=bit13: 2^5
:=:=:=bit12: 2^4
:=:=:=bit11: 2^3
:=:=:=bit10: 2^2
:=:=:=bit09: 2^1
:=:=:=bit08: 2^0
:=:=:=bit07: 2^-1
:=:=:=bit06: 2^-2
:=:=:=bit05: 2^-3
:=:=:=bit04: 2^-4
:=:=:=bit03: 0
:=:=:=bit02: 0
:=:=:=bit01: 0
:=:=:=bit00: 0
:=:=:=
:=:=:=I want to convert a 2 byte data in this form to a float point data and save it in a variable. I need a calculation, if you want to know, this is related to the digital sensor DS1631.
:=:=:=
:=:=:=Help me help me
:=:=This is just an integer, scaled /256.
:=:=Assuming your two bytes are held in a variable declared as a signed integer - if not, use a union, and have a signed int16 as one of the methods of accessing the data like :
:=:=union data {
:=:= int8 b[2];
:=:= signed int16 w;
:=:=} value;
:=:=
:=:=Then you can access the individual characters as:
:=:=value.b[0], and value.b[1], and the whole 16bit word as value.w
:=:=
:=:=Then just code as:
:=:=result=(float)value.w/256.0;
:=:=
:=:=This 'casts' the integer into a float, and then divides this by 256 to give the correct result. However note that I am _forcing_ the operation to be done on a float, by using a float constant, and doing the conversion before the arithmetic (CCS C, has a tendency to try to save time, by converting latter...).
:=:=
:=:=Best Wishes
:=
:=
:=Works fine!
:=
:=
:=THANK YOU THANK YOU
Glad to help.
You phrased your question clearly, and by thanking well, have made it a pleasure to help you.

Good luck on the rest of the project.
___________________________
This message was ported from CCS's old forum
Original Post ID: 11296