2023 Prius Prime EV Range

Excuses, excuses, and handwaving. And a refusal to properly define (as you demanded of me) and then actually calculate it out.

Your baseline is too short. You need longer distances of regen, as Salamander has called out.

You could also try a hill, up and down, then also a similar flat road (if pavements are comparable roughness) at the same speed, and compare SOCs.

80%, then 90%, then 100%. This is a fish story, it keeps gets bigger with each telling. Next up, 110%, then 120% ??

I'd like to attempt some theoretical calculations. What is your speed at the start of your regen braking, and also at the end of that regen cycle if you don't come to a full stop? And then you accelerate back up to the start speed? And how many miles into your trip is this regeneration/acceleration cycle performed? Or if performed multiple times during the trip, what are the numeric details of several of them?

Thought without inside knowledge into Toyota's display filtering algorithm, my calculation attempt is likely to be an exercise in futility.

 

Again, after tonight's 15.5-mile round trip, which involved slow driving, dozens of traffic lights, very gentle acceleration, and very gentle deceleration, I am absolutely convinced that regenerative braking is close to 100% efficient under these conditions, the actual number probably being around 90%.

In both legs, I got 6.9 mi/kWh at the end. The return leg of my trip was probably a better evidence, as the mi/kWh had already stabilized. I started the return leg at 6.9 mi/kWh, and at every green light, it dropped to 6.7–6.8 mi/kWh, and at every red light, it went back up to 6.9 mi/kWh. This is simply not possible if regenerative braking isn't around 90% efficient unless the mi/kWh in the eco diary is doing something strange. There is no funny business here, I always report my numbers for round-trip driving. When gravitational energy isn't excluded through round-trip driving, all bets are off. The current generation Prius Prime has clearly perfected regenerative braking for very gentle acceleration and very gentle braking.



Note that May 20 and 21 are mostly HV driving, and you can ignore those numbers. May 23 is an actual 15.5-mile EV round trip.

Regarding mi/kWh, I don't believe anyone can get much above 7.0 mi/kWh in a round trip. 7.0 mi/kWh is already 44 mi on full charge. If you are seeing 8.0 mi/kWh on the eco diary or people are seeing 50 miles on the EV DTE, this is because you and they are only using the EV mode for downhill driving and using the HV mode for uphill driving. 7.0 mi/kWh and 44 mi is more or less the upper limit for round-trip EV driving. I could believe slightly higher but not much higher.

 

The trip meter shows the last leg was 6.2 miles. That means that that 15.5 mile round trip was not symmetrical: The car rested for some time in the last 1/3 of the trip. Or it was only a 6.2 mile trip and the picture was snapped when the numbers looked good.

I have to agree with others. If you want to claim that the regen is near 100% efficient, all you need to do is drive 10 miles on a flat road in EV mode. A country road works best. Note the SOC % at the beginning. Use a cycle that is 100% gentle acceleration to 35 MPH followed immediately by braking gently to 10 MPH or so and then repeat. It should never engage the ICE nor the brake pads and should never be coasting. At the end of 10 miles the SOC % will tell you how much energy was lost.

Keeping speed under 35 minimizes the aerodynamic drag. Keeping speed above 10 keeps the brake pads out of the equation.

The kWh/mile is not a measurement of how efficient the regen braking is.

 

It was a 15.5-mile round trip, but I stopped at four different stores. The picture shows the end of the last leg.

Again, the SOC doesn't have enough precision to calculate deltas. 50.49% and 49.50% both show as 50%. You would calculate the delta to be 0%, but in reality it is 1.0%.

On the other hand, the mi/kWh rate better reflects the deltas, as the computer doesn't round off before the mi/kWh rate is calculated. You could try it yourself when you drive in the EV mode to see how it decreases when you start from a green and then how it increases when you approach a red.

 

This is where I don't understand why you think you can calculate the regen efficiency from those numbers you have observed. The number represents only the daily average cumulative of the total EV miles driven on that day and kWh used. Here is the hypothetical breakdown of your trip at the end.

Say for a total of 15.5 miles driven on EV mode, you are at a 15.0 miles mark stopped on a red light and the car is showing 6.9miles/kWh. On green light, you gently accelerate for 0.3miles, the number dropped to 6.7miles/kWh on Eco Diary. Then preparing to stop again, you gently decelerate for the next 0.2miles using regen and come to a stop on a red light. The number is now back to 6.9miles/kWh. You are saying this fluctuation is proof that regen is near 100% efficient. OK, hypothetically if this is how you finished 15.5 miles of EV drive for that day. Now let's look at the number and see if it shows near 100% regen efficiency.

EV efficiency for the first 15 miles was 6.9miles/kWh. This means you have used 2.173912kWh of energy from the traction battery for this distance. (15.0/6.9)

Now, you gently accelerated the next 0.3 miles and the number dropped 6.7miles/kWh. This means you have now used 2.28582kWh of energy from the traction battery for a total of 15.3 miles distance. (15.3/6.7).

This means you have used 0.1096669kWh (2.28582-2.173912) of energy for the acceleration part of 0.3 miles EV drive.

Then you gently decelerated the next 0.2 miles and the number went back to 6.9miles/kWh at the end of 15.5 miles. So overall, you have now used a net 2.246377kWh of energy from the traction battery (15.5/6.9). This means you have regained 0.03721kWh of energy from the regen in the last 0.2 miles of deceleration (2.246377-2.28582). Note that since this is a net gain of kWh, the number is negative.

Now, compare the kWh used for the acceleration 0.1096669kWh
And the kWh gained for the deceleration 0.03721kWh

That's only 34% (0.03721/0.1096669) regen efficiency. According to your claim, this number should be close to 100%. On this hypothetical trip, even though the last 0.5 miles segment of the trip showed the fluctuation of number from 6.9miles/kWh to 6.7miles/kWh and again back to 6.9miles/kWh, that does not show near 100% regen efficiency at all. The problem here is that the first 15.0 miles of the trip is much longer than the last 0.5 miles of acceleration/deceleration segment so the daily average number presented does not reflect the actual EV efficiency of the short segments at all.

 

There's a mountain road I drive semi-frequently. The theoretical gravitational potential energy of the elevation difference is ~5,700 Wh. In my Prime descending that hill in B mode with very little application of the brake pedal (and when it is applied it is well within regen limits), I gain approximately 70% SOC. I've never made a controlled experiment out of it, but given those numbers we're looking at ~70% efficiency, which is pretty dang good.

I need an app which can show me a fairly precise actual SOC and voltages for the traction battery to carry out some more detailed experiments. Even better would be something that logs those variables, and maybe even traction battery temp and input current and voltage. Recommendations?

 

And this 70% efficiency is under very challenging conditions, putting the system near its limits and probably requiring the occasional use of the brake pedal.

There is no reason why the regenerative braking efficiency can't be around 90% for very gentle acceleration and deceleration. The lithium-ion charging system is no worse than 95% efficient and the motor/generator is more than 95% efficient as far as I know. There is really not much other loss unless the friction brakes are being used.

 

If only you weren't carless and you could test this on your own Prius Prime.

The observation is based on much more than the very last traffic lights:

(1) For example, during a very long stretch of cruising, mi/kWh does not keep increasing above 6.9 kWh. In fact, it also drops to 6.8 or 6.7 kWh occasionally depending on the load.

(2) The amount of mi/kWh drop during acceleration and mi/kWh increase during deceleration keeps decreasing after every traffic light when more miles are added to the trip, as they are averaged over a longer trip. Originally, it is much higher than 0.1–0.2 mi/kWh, but eventually it is no more than 0.1 mi/kWh.

Based on the continuous observation along the whole trip of (1) and (2) above, there is no other explanation than regenerative braking being very efficient. I am guessing about 90%, and I don't think it can be less than 85% based on my observations.

I am repeating this from my above post: There is no reason why the regenerative braking efficiency can't be around 90% for very gentle acceleration and deceleration. The lithium-ion charging system is no worse than 95% efficient and the motor/generator is more than 95% efficient as far as I know. There is really not much other loss unless the friction brakes are being used. However, for hill descent or quick deceleration, you are putting the system at its limits, and the efficiency could be a lot worse.

 

The motor has to feed the generated power back through the charging system, which is ~90% efficient. I don't think anyone on Prius Chat knows enough detail about how the regenerative system works with the charging system to estimate system losses. For example, how is the voltage/amperage from regeneration regulated? Do we get more voltage as regeneration increases, more amperage, or both? How does that have to be regulated/converted so it can be fed to the charging system? There's no way Toyota is feeding raw power from regeneration directly into the batteries. That would be battery suicide. Every stage has losses. Even if regen were 90% efficient, by the time it was fed through the charging system it would be around 80%.

Just to achieve the 70% I tentatively observe, the charging system and the regeneration system have to be around 84% efficient.

 

Sorry, but your observation and guess are not based on mathematical calculation. In the spreadsheet, I can plug in any distances and numbers of acceleration and deceleration along the way of a total of 15.5 miles of EV trip. Yes, I can simulate the multiple stops-and-go traffic with each time the miles/kWh bouncing back to 6.9miles/kWh after regen. But as long as the acceleration and deceleration segments are short compared to the total travel distance you will not see anywhere close to 100% regen at any point. I just do not see how you can make a reasonable estimation of regen efficiency by just glancing at the bouncing daily average mile/kWh number before and after each traffic stops.

 

I think the lithium-ion battery itself has a very high charging efficiency, and they have perfected their electronics that their charging system is very efficient when it converts the voltage and current. The motor/generator is extremely efficient.

 

For the older Nissan Leaf, regenerative braking efficiency was reported to be 70%. The Prius Prime is a more efficient EV than the Nissan Leaf, and this should be higher, especially with gentle deceleration.

Analysis of the regenerative braking efficiency of a latest electric vehicle

Moreover, when I observe the efficiency by looking at mi/kWh, I am looking at the relative efficiency in comparison to the cruising efficiency. Therefore, what I perceive to be 90% efficiency for regenerative braking is probably more like 80% when the drivetrain losses also present during cruising are taken into account.

 

Per that link... "While the propulsion efficiency approaches 90%, the round trip regenerative braking efficiency reaches the 70%,". With a very high propulsion efficiency and much lower round trip regenerative braking efficiency, the obvious conclusion is that the regenerative braking efficiency is way less than 70%. Probably 50% or less. This is based on the probable computation of round trip regenerative braking efficiency as the combination of the very efficient propulsion averaged with the not so efficient braking.

 

No, make the regenerative side about 80%. Then multiply the two directions together to get the round trip figure. I.e. 90% (propulsion side) X 80% (regeneration side) ==> 72% (round trip). Then this lands close to a pile of the industry literature, which shows plenty of estimates from 60 to 70%, and a smaller number scattered across the 70-something range.

Back in the Roadster days, a decade before the Prime, a Telsa engineer wrote this:
" We must also remember that, even though the battery-to-wheel conversion efficiency is pretty good (up to 80% or so), the energy makes a full circle back into the battery and it gets converted twice for a net efficiency of at most 80% * 80% = 64%."
The Magic of Tesla Roadster Regenerative Braking

 

So that means you are resetting the mi/kWh display after accelerating up to cruise speed? And having no previous decel/accel cycles mixed in with that cruise segment, let alone multiple such cycles?

Without such a cruise-speed reset, then you are comparing not against cruise speed energy consumption, but instead against an average of cruising and regeneration cycles, with undisclosed weighting between the two modes.

Back at post #85, Salamander displayed an example of the computations possible when knowing more details of the drive profile. You need to provide us more such detail, including a more complete profile of the preceeding portions of the drive cycle, for us to do additional computations. Otherwise, your seat-of-the-pants guesses are little more than handwaving.

 

No resetting.

What I meant was that when judging what is going on during regenerative braking and acceleration from mi/kWh, you're implicitly measuring the efficiency with respect to cruising efficiency, which is your baseline. So, if regenerative braking appears to be 90% efficient according to mi/kWh behavior, it would have an absolute efficiency of more like 80%.

In any case, this is something you can try yourself if you have a Prius Prime. Just watch mi/kWh being dynamically updated on the eco diary during a round trip.

 

The reason I did this was not to obtain an actual efficiency number for regenerative braking but to show that very gentle acceleration and braking, as opposed to cruising all the way, has no noticeable effect on the EV range.

You can easily do this experiment yourself by watching mi/kWh being dynamically updated on the eco diary during a round trip. If acceleration and/or braking are decreasing your EV range, you are accelerating and/or braking too hard. I can't observe a measurable effect of very gentle acceleration and braking on the EV range. I am maxing out my city round-trip EV range around 44 miles, and there is no way I can add much to that even if all traffic lights were eliminated and the only acceleration and deceleration happened at the beginning and end of the trip, respectively.

 

1. You're arguing that as long as you accelerate and brake gently enough your range will not be negatively impacted. This is the perpetual motion machine I'm talking about. If you can move the car any non-zero distance with 0 net energy, it means you can move it infinite distance with 0 net energy. It's impossible.

2. The rate of acceleration has no effect on the energy required to perform that acceleration. KE = 0.5m * v^2 regardless of the time it takes to reach v. If the rate of acceleration impacts the energy required it means there are losses in the system, thus making #1 impossible.

 

The same losses also happen during cruising. Very gentle acceleration and deceleration is when the regenerative braking is the most efficient, and according to my observations, the net additional loss due to them is negligible and unobservable. Don't forget that even during flat-land cruising there is always acceleration and deceleration as you're constantly going over mild humps such as railroad crossings. There is no such thing as acceleration-and-deceleration-free cruising, and constant speed on the speedometer does not indicate mild gravitational-potential changes, which are acceleration/declaration. Moreover, you are constantly adjusting your speed for other cars during cruising, which is acceleration/declaration. For that reason, if you go through the traffic lights with very gentle acceleration/declaration, you aren't impacting your EV range at all.

Again, do the eco diary miles/kWh experiment yourself.

 

The EV range as predicted by the eco diary is not a valid source of data. The miles/kWh is a number that's calculated and it is impacted by your environmental controls among other things. There are formulas that use the car's weight, coefficient of drag, etc to calculate the energy needed to move the car X distance. You should be using that formula for your base level, and any change in SOC that is less than predicted would be the regen.