Cold weather hill climb induces engine ON

Thanks, I've corrected my earlier post. My understanding is you had two of which one was a family member not seating the J1772.

Bob Wilson

 

I had one on a practice run an hour ago:

So I pulled off and power cycled the car. I did three more cycles and it did not happen again.

• Use "," to backup a frame and "." to forward a frame.

Bob Wilson

 

So 27F (-3C), I was able to record:

• cruise control engine ON - power on reset back to EV, 0.7 mi, 46 mph, 98% SOC
• cruise control engine ON - power on reset back to EV, 1.1 mi, 47 mph, 94% SOC
• cruis control failed engine ON - 1.8 mi, 39 mph, 90% SOC
• hill climb engine ON - power on reset back to EV, 13.0 mi, 47 mph, 16% SOC

Previous history or prep does not appear to suppress either effect. However the SOC, <=90%, appears to suppress the cruise control engine ON although I hadn't reached mid 40 mph.

I also learned that Google Earth, route profile does not strictly follows the plotted route. When drilling down to the exact location of the hill engine ON, I observed the arrow pointing to the spot wandering up the hill side, off the road and route giving an impossible 78% grade. I had seen earlier symptoms when the route profile followed the bank down to the creek instead of using the bridge. What this means is to get a proper grade record, I'm going to need to record the three axis GPS data or use my three axis accelerometer.

The reason for knowing the grade is to map speed vs grade to identify 'safe' speeds when climbing but I may be overthinking the problem:

• Google Earth route profile wanders off the road and up the hill side giving unrealistic grades.
• Yes, dog is my copilot.

It appears the power bar reaches the max EV limit and within second, the engine ON occurs. So it may be adjusting the speed to keep the power bar from the max might work. Perhaps targeting the text "EV" as the maximum rate might inhibit the switch to engine ON. I find it curious that the descent, engine ON event also occurs when the regeneration reaches the maximum value.

FYI, this shows where I am planning to mount the dash cam on the roof liner. The angle gets a clear shot of the dash instruments as well as a clear image of the windshield. Right now, I'm using the passenger sun visor but there is more vibration than I want. However, I have a plan for a light weight, rigid mount that won't make a permanent change in the liner.

Bob Wilson

 

With temperatures around 43F (6C), I tried some more hill climb and descents using GPS recorded from Speedometer 55:

1. descending 35 mph, 11% grade
2. descending 46 mph, 4.3% grade
3. descending 48 mph, 7.2% grade (not plotted on chart)
4. ascending 62 mph, 2.4% grade

These were cruise control set speeds. But we've seen similar behavior under manual control. I was hoping the grade times the speed which should be proportional to the energy gained would be constant. But these are Googe Earth reported grades. I really need to work with the raw GPS data. So far, the only mitigation is to pull over and power ON reset to return to EV mode.



UPDATE

Using just the GPS recorded data, it appears there is a pattern on hill descents:

• Each line ends with an 'engine ON' event.
• Two down grades exceeding 10% appears to trigger the engine ON
• Cruise control was used to control the speed

I normally don't drive in hilly area and Dixie doesn't have a lot of cold weather. I would likely run the engine in HV mode early, long enough to warm up the engine. Then I would switch to EV confident the car is warmed up for these unexpected, 'engine ON' events.

Nothing stands out on the hill climb 'engine ON' event. Curiously it did not happen when climbing the first hill for the descents:


Bob Wilson

 

Sorry, I mistyped the hill climb plot:

With temperatures around 43F (6C), I tried some more hill climb and descents using GPS recorded from Speedometer 55:

1. descending 35 mph, 11% grade
2. descending 46 mph, 4.3% grade
3. descending 48 mph, 7.2% grade (not plotted on chart)
4. ascending 62 mph, 2.4% grade

Nothing stands out on the hill climb 'engine ON' event. Curiously it did not happen when climbing the first hill for the descents:


Bob Wilson

 

With a day of 50-64F (10-18C), I replicated my earlier route and got similar results: two descent 'engine ON', and one 'engine ON' during a climb. All were on cruise control keeping up with traffic.

The descent 'engine ON' at 63-64F were at:

• 54 MPH, N34.66771 W86.51012 (*)
• 32 MPH, N34.65931 W86.50098 (**)

The previous 'engine ON' at 42-43F were at:

• 41 MPH, N34.65711 W86.51364 (*)
• 43 MPH, N34.65706 W86.50700
• 48 MPH, N34.65959 W86.50062 (**)

This suggests it is the grade that triggers these 'engine ON' events. The temperatures not so much and speed not so much although we aren't looking at deceleration rates, yet.

The ascent 'engine ON' at 62F was at:

• 64 MPH, N34.70861 W86.52871 - 'engine ON'
  • 44 MPH, N34.71808 W86.56335 - reverted to EV

The previous 'engine ON' at 43F was at:

• 63 MPH, N34.70652 W86.52492

Absent a safe shoulder, I reduced speed to ~45 MPH and soon reached the crest. So I descended while in 'engine ON' mode to see what happens. The car was in regeneration for the next 2.2 miles until it returned to EV mode. This is consistent with earlier 'engine ON' events.

I need to do one more test without cruise control to remove it but I am forming an opinion that our Prius Prime has trouble with EV driving on hills. I had run the same route with our BMW i3-REx and it remained in EV mode the whole time.

There is a bug with cruise control, 'engine ON' under 50F (10C) and SOC in the 96 - 100% range. There is a temperature component that freezing temperatures allow lower SOCs, 91 - 100% to work.

Without cruise control, descents were normal, no ‘engine ON’ events. Still had one on the hill climb again just before the crest.

Bob Wilson

 

I noticed @john1701a drives in ECO so I repeated the hill tests just before the temperatures dropped with the next cold front. There was one descent 'engine ON' and one hill climb 'engine ON'. Both were at familiar spots from previous tests.

Bob Wilson