G5 PHEV track days - battery cooling, charge, and discharge - figures and observations

Warning: This is a long nerdy post that looks more like a report. TLDR skip to the underlined parts.


This week I took my Prius Plug-In to a racetrack twice. I had the whole track for me in both occasions so I took the chance to run some 'tests' and figured I'd report my observations here in case someone is interested.

All figures come from the Dr. Prius app connected to the OBD port through an old Bluetooth Carista OBD scanner that I bought for my G2 years ago. I also tried to sniff on the CAN bus through the same OBD port with another adapter and my laptop but everything is unsurprisingly behind a gateway, similar to the G4.

The Dr. Prius website does not list the G5 in the supported models, so there is a possibility the readings are incorrect. However, to get the readings the app should send messages through the OBD port with specific diagnostic codes, as the sniffing above showed that no information are exposed on their own. These codes may have not changed from G4, so I'm inclined to believe the figures (exceptions below).




Battery cooling
This was the information I was most interested in. I already knew that the cooling circuit activates when one of the sensors reaches slightly below 40 °C (104 °F), but I wanted to find out whether it happens anyway even if the AC is off, and conversely if the car would opportunistically cool the battery if the AC is on but 40 °C are not reached.

The liquid battery cooling activates on its own even with the AC OFF (I never turned AC on before that, on purpose), till the battery temperature returns to at least around 32 °C (90 °F). I forgot to check afterwards. The rate of cooling is not dramatic but significant, you see the temperature decreasing in real time.

The cooling circuit does not opportunistically cool the battery if the AC is on. The posted screenshot shows battery temperature after 5 minutes of cooling the cabin with the car in P, drive mode Normal (no Eco AC), and the AC set to Auto, 22 °C (~10 °C less than ambient). At screenshot time the refrigerant is already cold as the discharge current is already low at 3.9A, while it sucked 9.9A when I turned the AC on. In these 5 minutes battery temperature decreased by about 0.03 - 0.05 °C, regular heat dissipation and nothing else going on. With the car in P and the battery cooling circuit active the battery cools down significantly faster. This should show that the Panasonic/Toyota engineers are confident in the battery being ok at temperatures < 40 °C (104 °F).

As expected, the battery quickly heats up with repeated hard discharge/charge currents. It went from 25 °C (77 °F) to 35 °C (104 °F) in a few laps.

Charge and Discharge Currents
With the battery in optimal conditions (temperature, SOC) Dr. Prius reports a max charge of 48 HP (35 kW) and a max discharge of almost 85 HP (62 kW).

Now, from specs the max power of "the" electric motor (MG1? MG2? I will ignore this question for simplicity) is 163 HP (120 kW), and I've seen everywhere inferred that the max. power in EV mode is therefore ~160 HP. This is likely false. That power may be probably neared at some point in HV mode when both the battery and the engine are contributing.
In EV mode, 0-100 or 0-60 in ~11/10 seconds are also not consistent with 160 HP. The max power in EV mode is given by the max power the battery can discharge, not the max power of the motor.
However, the 0-100, 0-60 figures are likely not consistent with 85 HP moving 1500 kgs of mass either, even accounting for a 'fuller' power delivery curve with respect to an ICE.

My theory is that the numbers reported by Dr. Prius indicate the max sustained power that the battery can charge/discharge (other option: they are simply wrong). If you floor it, the discharge current goes to more than 300 amps, which when multiplied with the battery pack voltage returns 80-90 kW, ie. 110-120 HP, which IS consistent with the 0-100, 0-60 figures. It is also ~2/3 of "the" 160 HP electric motor.
The same goes for charge. Just pull your foot up at high speeds in B mode set up at max regen, or brake, and VxA results in more than 35 kW (the discharge current square in the screenshot turns into charge current, green).

Few more notes on those values:
- The max charge value of 35 kW applies from ~75% dashboard SOC downwards. With higher SOC the value is lower.
- B mode works differently in the Plug-In than the regular Prii as an almost 1-pedal drive. This applies even to 100% dashboard SOC, when simply the regen force is noticeably lower in B mode, like in all EVs. However, in situations where the dashboard SOC is near 100% AND the requested regen is too high (eg. down long hills, brake pedal firmly pressed at speed) the ICE will turn on, to warm up or maybe even immediately provide additional resistance. This does not seem to happen in D mode, and it's likely to protect the battery in "old B mode" scenarios.
- Going down a long hill is the perfect experiment to see the max charge value gradually decrease in real time. This depends on the time of continuous regen, which also gives credit to the theory above that the figures reported by Dr. Prius are related to sustained, not peak, powers. After a few seconds of discharge or no charge, the value quickly returns up. If the max charge value gets too low and the engine is cold, it will probably turn on even only to warm up just in case.
- As the SOC gets lower, the max discharge value decreases too. How much depends on battery temperature. In the summer is almost unnoticeable. In the winter, when the battery temperature is low (< 5-10 °C - 40-50 °F) AND the SOC is low (~15% dashboard SOC or lower) the max discharge value goes down, fast. My observations indicate that when the max discharge value reported by Dr. Prius gets to ~60 HP (45 kW) or less, the ICE will turn on if cold. This is why some drivers report the ICE turning on on its own in EV mode in the winter. Note that after the warming-up has completed, the ICE will only turn on again if the pedal is pushed hard to make up for the power the battery is now unwilling to discharge. Typically you can then reach 0% dashboard SOC in EV mode without calling in the ICE again.
- The ICE running in EV mode is NOT equivalent to HV mode operation. The engine outputs less power, and it consumes less fuel, even in idle with the car still. Easy to see it by switching between modes while the ICE is running. I would guess something in the transmission is disengaged as it's not necessary to charge the battery like HV mode would, but it's just a guess.

Track observations
- It was fun
- I just rotated the original Yokohama BluEarth-GT AE51 tires (195/50 R19) that came with the car, so the traction tires were basically new, yet being 195s I was frankly a bit surprised. I thought they would give up quite sooner than they actually did. That said, a good set of larger tires would do much better. That's the first thing you should change if you plan to track this thing.
- The brakes seem good. They do not look very large, especially with the gigantic 19" wheels, but combined with the regen they do their job.
- I'm not a track expert by any means but I think the car handles quite well, as many reviewers have said.
- Suspension/chassis feeling: very similar in principle to my Ducati, which is good on tracks but built for road use. Grab the front brakes and initially the suspensions give in for road confort, then you can clench all you want but the bike doesn't sink much more. The Prius, initially a bit of body roll, most noticeable between sudden change of directions, but after that, not much more. The car stays quite flat in the corners, and pulls well ouf of them, though with a bit of understeer if you push it.
- Acceleration is great for newbies, fun but not intimidating, but you really can't use all the 220 HP out of corners with the stock tires. I mean you can, but you do lose a bit of traction.
- Something smells weird afterwards. Burning weird. Don't know if it was tires, brakes, or oil. Probably tires or brakes. But, no more rust on the brake discs. I'm always in B mode so I nearly never use them, let's see how long they last clean.

 

Good points! I did not think about scaling which is indeed a thing when interpreting data coming from the CAN bus.
Next winter I will also try to accelerate hard with the engine on in EV mode, that might give us the answer regarding whether the difference is due to the engine only idling without charging or if it's a different mode altogether.
There is also this video on Youtube showing acceleration in EV mode after passing the EV-only threshold of about 135 km/h (143 km/h on the dashboard, apparently), where the engine seems to be working less than in HV mode.