Do brake lights come on each time you apply the brakes?

What I called "the solenoid" is probably better called "the relay" described in post #11above by Elektroingenieur

 

In post #11, Elektroingenieur wrote about several things, including RAV4 HV models.
I can only say what I see and hear.

After some reading and experimenting, I have the data refresh rate down to 0.6-0.8 seconds. Not as good as I hoped (only about 7 PIDs per second), but good enough to assist the discussion. From the latest log there can be no doubt that the stop light PID does not activate at (gentle) brake pedal pressures, enough to activate the stroke sensor and the friction brakes (gently, but enough to feel in my seat-of-the-pants meter). When the brake pedal is pressed a little harder it causes pressure in the regulator pressure sensor, and the brake lights also illuminate.

It would probably not be an issue in an old car. I remember a MG Magnette: beautiful car but you really had to push the pedal to get a response from the drum brakes in those days. Then some of the early disc brakes were quite lethal when they engaged, but you still had to push the pedal a ways to engage them. All good for a brake light switch mounted near the pedal arm.

But in hybrids, drivers are encouraged to brake gently to maximize regeneration. The car is still slowed, but as I have shown a brake light switch mounted near the pedal is now less fit for purpose. IT is much harder to adjust,and keep in adjustment, with the required finesse. Hopefully TMC will think about using stroke sensor output (or whatever actually activates the brakes in future) to activate the stop light, as suggested above.

This graph is from a raw log, so it looks a little different to those above. I did not log regen braking as it comes from a separate ECU (which may sow down the data rate). But from the logs above, regen torque generally follows a request from regen cooperation. So this should still help.

 

Thanks again guys.

Wouldn't this show up in the stroke sensor, and so meet the quoted regulations?

Of course, both an ECU and a switch mounted near the brake pedal can be used. This should ensure that the brake lights always operate with the brakes, and comfort anyone who does not trust multiplex-bus messages (and might therefore not use dynamic cruise control or any of the CAN-dependent safety features in recent Toyotas?).

Here is another log. It has a bit slower refresh because of the added PIDs, but it is still interesting and taken with those above I think it shows:
1. Stroke sensor basal reading is 1.24 -1.25V.
2. Stroke sensor above basal reading (eg 1.29V detected) causes friction and/or regen braking (depending on car speed, regen cooperation etc).
3. Stroke sensor above 1.4V corresponds with brake light (operated by a separate switch).
4. Stroke sensor above 1.5V corresponds with increased regulator pressure.

I hope it helps someone.

 

ChapmanF, you were right that my guess about that clunk was wrong. It is quite audible while sitting in the car in park. It clunks at stroke sensor voltage of 1.35-1.36, well below the brake pedal pressure to activate the brake light (and it seems to correspond with first detectable pressure in the wheel cylinder while parked). All visible in a live display of those PID readings while parked.

I don't think this alters the main messages in the thread at all, but I wanted to acknowledge that you were right that the clunk is not the brake light relay.

 

It certainly makes sense that the loud clunk when the brake pedal is pushed comes from the switching solenoid valves (diagram attached). There are also quiet clicks when only the stop light switch plunger is prodded (the brake pedal being held in for this purpose), but it is necessary to get ones head near the brake pedal and listen carefully to hear those. I have not tried to relate them to any PID outputs.

 

Bob, I checked - my fridge light works every time (geez it is cold in there, and no PIDs). Shame the same does not apply to Toyota HV brake lights yet, but we can hope for the future.

ChapmanF, my diagram was for the AVA (gen 4 rav4 HV) so the AXA (gen 5 rav4 HV) may well have the greater complexity of the gen 4 prius (probably they all have the solenoid clunk). Thanks for the link to the diagram from Elektroingenieur; it shows nicely why the brake pedal stroke sensor should also be suitable for brake light control during emergency braking.

 

The Toyota service technician said that the stop light was fine. I think the sophisticated test applied was to press the brake pedal firmly and confirm that the stop light illuminated. (Even though I explained the concern in detail to the service consultant, it was probably not passed on to the technician). It is your choice whether to trust someone who: (a) could not manage a 5-wheel rotation according to the Toyota owner's manual, even though an enlarged copy of the correct method was placed on the front seat; (b) 'forgot' to reconnect the dashcam that was disconnected for some reason during the service; and (c) inflated tyres to pressures way above those specified by Toyota.

If you want to check your own car, it is easy in a dim garage where you can see the glow from stop lights in the rear view mirror. If you press the brake down very gently and hear the solenoid clunk before the stop light comes on, your brakes are coming on before your stop lights activate. No need to climb into the fridge!

ChapmanF, it is a bit off topic, but I looked more into PID refresh rates. Basically the rate-limiting stage is not the 500 Kbaud 11 bit CAN bus, or the OBD reader UART at 115.2 Kbaud, or BT3 at 24 Mbps. What matters is the match between unregulated features of the serial port protocol (SPP, which operates under BT to handle the RS232 data from the UART) in the OBD reader, the OBD software and the host device. Data rates down to 8-10 Kbps have been measured. This explains why (allowing for the full cycle of request, receipt, processing and transmission of roughly 125 bit PID response frames) users experience below 60 PIDs/sec. Although I don't know much about BLE, I understand that it uses GATT, which is typically slower than SPP in BT. Rates can be faster over USB, but even there, something in the software seems to throttle the refresh rates, which vary a lot between OBD apps. There are claims of gateway modules that limit data rates from CAN to OBD, but I have not seen evidence for this. When sellers of OBD systems say "maximum X PIDs/sec" what they mean is "You will never get anything approaching X PIDs/sec". Of course, none of this matters when one just wants a few gauges, but as you say it is a limitation when one wants logs of multiple sensor readings at sub-second intervals (eg to study braking).