C51: Transmitting Floating-Point Numbers

Information in this article applies to:

• C51

QUESTION

How can I send and receive floating-point numbers via the on-chip UART? I have tried using a printf/scanf combination, but I cannot get it to work.

ANSWER

A floating-point number is stored in memory using four bytes. A good way to transmit the number is to simply send those four bytes, using a pointer to access the bytes.

Here is an example:

float f = 4.7838322;

void main(void)
{
  unsigned char *ptr, i;

  ptr = (unsigned char *)&f;
  for (i = 0; i < sizeof(float); transmit(*(ptr + i)), i++);
}

where the function transmit() is your custom function that transmits an unsigned char using the on-chip UART or whatever method/peripheral you choose.

To receive the floating-point number simply reverse the process, again using a pointer.

MORE INFORMATION

• Refer to Floating-point Numbers in the Cx51 User's Guide.

SEE ALSO

• C51: Support for Floating-point Numbers
• C51: Are 32-bit SFRs Supported?
• GENERAL: byte-wise access to float variables

Last Reviewed: Thursday, February 25, 2021