RaceChronoDiyBleDevice/can_db/ft86.md

7.5 KiB
Raw Blame History

FT86 cars (Subaru BRZ, Toyota 86, Scion FR-S)

Here is information on some of the PIDs that you're most likely to use with RaceChrono and equations to get the right scale, etc.

PID 0x18

Update frequency: 100 times per second.

0x18 is a strange PID. Judging by the low PID number (which in CAN networks implies higher priority) and the high update frequency, one would expect it to have some important data, like data for ABS or ESC systems. But based on what is currently known, it only has one the steering angle as a useful data channel. The steering angle graphs are usually relatively smooth, and this data is also available over 0xD0, along with some much more important data, such as accelerometers.

Channel name Equation Notes
Steering angle bytesToIntLe(raw, 0, 2) * -0.1 Also available in 0xD0
??? C or bytesToIntLe(raw, 2, 1) The value is 112 most of the time for me
??? D or bytesToIntLe(raw, 3, 1) 014 sawtooth
??? E or bytesToIntLe(raw, 4, 1) The value is 0 most of the time for me
??? F or bytesToIntLe(raw, 5, 1) The value is 0 most of the time for me
??? G or bytesToIntLe(raw, 6, 1) The value is 0 most of the time for me
??? H or bytesToIntLe(raw, 7, 1) Strange data channel. It changes in a strange way when the car turns. It also has a 0-14 sawtooth over some otherwise smooth curve.

PID 0xD0

Update frequency: 50 times per second.

Channel name Equation Notes
Steering angle bytesToIntLe(raw, 0, 2) * 0.1 Also available in 0x18
Z rate of rotation bytesToIntLe(raw, 2, 2) * -0.286478897 The multiplier for º/sec appears to be ((90 / pi) * 100)
??? E Some flags?
??? F Some flags?
Lateral acceleration bytesToIntLe(raw, 6, 1) * 0.2 Not 100% sure about the multiplier, but looks about right
Longitudinal acceleration bytesToIntLe(raw, 7, 1) * -0.1 Not 100% sure about the multiplier, but looks about right
Combined acceleration sqrt(pow2(bytesToIntLe(raw, 6, 1) * 0.2) + pow2(bytesToIntLe(raw, 7, 1) * 0.1))

PID 0xD1

Update frequency: 50 times per second. Length: 4 bytes

Channel name Equation Notes
Speed bytesToIntLe(raw, 0, 2) * 0.015694 May want to check the multiplier against an external GPS device
Brake position min(C / 0.7, 100) The third byte is the pressure in the brake system, in Bars. The 0.7 divider seems to be a good value to get 100% at pressure slightly higher than those you're likely to use on the track for cars with no aero. You can use 0.8 or 0.9 if you see 100% too often.
??? D Always 0?

PID 0xD4

Update frequency: 50 times per second.

Channel name Equation Notes
Wheel speed FL bytesToIntLe(raw, 0, 2) * 0.015694 Use same multiplier as for speed in 0xD1
Wheel speed FR bytesToIntLe(raw, 2, 2) * 0.015694 Use same multiplier as for speed in 0xD1
Wheel speed RL bytesToIntLe(raw, 4, 2) * 0.015694 Use same multiplier as for speed in 0xD1
Wheel speed RR bytesToIntLe(raw, 6, 2) * 0.015694 Use same multiplier as for speed in 0xD1

PID 0x140

Update frequency: 100 times per second.

Channel name Equation Notes
Accelerator position A / 2.55
Clutch position (B & 0x80) / 1.28 On/off only
??? B & 0x70 Unused?
??? B & 0x0f 015 counter?
Engine RPM C + (D & 0x3f) * 256
??? D & 0x80 Always 0?
??? D & 0x40 1 when accelerator pedal is released, 0 otherwise
Accelerator position E / 2.55 Not clear what's the difference from the other two
Accelerator position F / 2.55 Not clear what's the difference from the other two
Throttle position G / 2.55 Not tested
??? H Some flags

PID 0x141

Update frequency: 100 times per second.

Channel name Equation Notes
Accelerator pedal position? bytesToIntLe(raw, 0, 2) Follows A from 0x140 closely with ~9860 for 0% and ~11625 for 42%. Needs more testing.
Throttle position? bytesToIntLe(raw, 2, 2) Follows the accelerator pedal position with some extra smoothness. Independent at idle revs, and you can see it changing in response to modulating the clutch in stop and go traffic. Scale is unknown.
Engine RPM E + (F & 0x3f) * 256
??? F & 0x80 1 when accelerator pedal is released, 0 otherwise
??? F & 0x40 Always 0?
Gear (G & 0xf) * (1 - (min(G & 0xf, 7)) / 7) It's basically just G & 0xf but neutral is reported as 7, hence the complex math to turn it into a 0. The reverse gear is reported as 1. The value can be wrong when the clutch pedal is depressed.
??? G & 0xF0 I saw values of 128, 160, 192 here.
??? H Equals to 16 when I lift off the accelerator, then turns to 8, then 0.

PID 0x360

Update frequency: 20 times per second.

Channel name Equation Notes
Engine oil temperature C - 40
Coolant temperature D - 40
Cruise control ON (F & 16) / 16 Means the mode is "On", but not necessarily "Set". Not tested much.
Cruise control set (F & 32) / 32 Not tested much.
Cruise control speed H In the same unit as the current speed display units? Not tested much.

TODO: would be great to find how to read the ambient temperature, and maybe the intake temperature.

TODO: find how to log the fuel remaining.

Connections

Besides the CAN pins in the OBD-II port, there is a CAN bus male port hidden behind the car multimedia head unit:

Hidden CAN port

It's very close to the glovebox, which makes it great for putting your CAN reader in the glovebox. Such a placement makes sure it's out of the way and you won't accidentally hit it with your leg while on the track, and also allows quick access for troubleshooting and experimenting. There's also a second 12V port inside the glovebox, which makes it easy to use a 12V-to-USB adapter instead of adding a 12V-to-5V converter to your hardware design.

You can use the two middle pins of a Toyota radio harness to connect to the CAN bus in a reliable way.

CAN socket wiring

It's recommended to use a ~6090 cm (23') twisted pair cable between that port and your CAN reader. I would strongly discourage you from using the screw terminals on the MCP2515 board, as if the cable comes loose it can cause a short circuit in the CAN bus (ask me how I know), which will throw a MIL at best, and who knows what at worst if it happens on the track. Instead, I attached a female JST SM connector to the cable.

Assembled device in the glovebox

Typical histogram of PIDs

Here's what the distribution of PIDs looks like in the CAN bus while idling in a parking lot:

PID Decimal PID Number of packets received over a 10 second period
0x18 24 1000
0xD0 208 500
0xD1 209 500
0xD2 210 500
0xD3 211 500
0xD4 212 500
0x140 320 1000
0x141 321 1000
0x142 322 1000
0x144 324 500
0x152 338 500
0x156 342 500
0x280 640 500
0x282 642 167
0x284 644 100
0x360 864 200
0x361 865 200
0x370 880 200
0x372 882 100
0x374 884 10
0x375 885 10
0x37A 890 10
0x3D1 977 84
0x440 1088 25
0x442 1090 25
0x44D 1101 25
0x46C 1132 25
0x4C1 1217 10
0x4C3 1219 10
0x4C6 1222 10
0x4C8 1224 10
0x4DC 1244 10
0x4DD 1245 10
0x63B 1595 20
0x6E1 1761 10
0x6E2 1762 10