kai/pounder_test
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Updating float conversion

Browse Source
This commit is contained in:
Ryan Summers 2021-05-07 14:02:25 +02:00
parent b73a4d9e59
commit 923790b0b9
1 changed files with 7 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
src/hardware/dac.rs
View File

@ -80,13 +80,14 @@ impl Into<f32> for DacCode {
// signal then passes through a 2.5x gain stage. Note that the DAC operates using unsigned
// integers, and u16::MAX / 2 is considered zero voltage output. Thus, the dynamic range of
// the output stage is +/- 10.24 V. At a DAC code of zero, there is an output of -10.24 V,
// and at a max DAC code, there is an output of 10.24 V.
//
// Note: The MAX code corresponding to +VREF is not programmable, as it is 1 bit larger
// than full-scale.
let dac_volts_per_lsb = 10.24 * 2.0 / (u16::MAX as u32 + 1) as f32;
// and at a max DAC code, there is an output of (slightly less than) 10.24 V.
(self.0 as f32) * dac_volts_per_lsb - 10.24
let dac_volts_per_lsb = 10.24 * 2.0 / u16::MAX as f32;
// Note that the bipolar table is an offset-binary code, but it is much more logical and
// correct to treat it as a twos-complement value. TO do that, we XOR the most significant
// bit to convert it.
(self.0 ^ 0x8000) as i16 as f32 * dac_volts_per_lsb
}
}