From 934b18ef1b0b17a62ee1fa2af5e5cf58e1c8f937 Mon Sep 17 00:00:00 2001 From: Timur Iskhodzhanov Date: Sat, 24 Jul 2021 21:17:10 -0700 Subject: [PATCH] Minor clarifications --- can_db/mazda_mx5_nd.md | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/can_db/mazda_mx5_nd.md b/can_db/mazda_mx5_nd.md index 2153e51..b6506d0 100644 --- a/can_db/mazda_mx5_nd.md +++ b/can_db/mazda_mx5_nd.md @@ -25,17 +25,15 @@ The binary representation used for the brake pedal/pressure is very weird. In my limited testing, I've observed the values of 149...154 when not touching the brake pedal. Most of the time it stays as 154. When progressively pressing the brake pedal, the number goes up to 255, then wraps around to 0, and keeps -going up. I've observed values as high as 145 when heavy braking. -It feels very much like +going up. I've observed values as high as 145 when heavy braking, and ~135 when +braking hard enough to engage ABS. It very much feels like the representation is `(cast_to_int(brake_pressure * multiplier) + offset) % 255`, as long as we allow -the `brake_pressure` to go slightly negative, and as long as the binary -representation for the maximum possible value and the binary representation for -the minimal possible values don't conflict with each other. +the `brake_pressure` to go slightly negative. Now out of all 0...255 possible values, this leaves us with only 146...148 values that I haven't observed. For now, I've made an assumption that 146 is "braking even harder than I was ever able to brake", and 147 and 148 are just -values "even more negative than I was able to observe". It is recommended to log +values "even more vacuum than I was able to observe". It is recommended to log the `E` byte by itself in a separate channel (e.g. "Digital 1") for diagnostic purposes so that if you ever see the "brake position" jump from 0% to 100% when you're not braking, or from 100% to 0% when you're braking super hard, we can