EV Mode: Why I'm Lovin' it!

Interesting stuff.

 

If you take a look at the money saved with the 30 MPG SUV instead of a 20 MPG, vs the 60 MPG hybrid instead of a 50 MPG, the former saved more money...

Of course anyone will agree that a Prius consumes less fuel than a SUV but that was not what I was trying to highlight.

 

Short Answer: Yup, EV mode during short periods of acceleration is exactly the way to use it.

Long Answer: Here is why...

Back to basic principles...

1. Energy is measured in joules and can also be called "work".
2. work = force * distance
3. force = mass * acceleration
4. power = work / time

In simpler terms: It requires more energy/work/joules to go from 0-60 in 5 seconds than it does to go from 0-60 in ten seconds. This is a law of physics and applies regardless of what you have under the hood...

5. ICE is about 20% efficient.
6. Electric motors are about 80% efficient.

Fundamentally, this efficiency difference is why a hybrid makes such sense.

Since acceleration is the most energy intensive step, you want to do it with the most efficient power-plant you have, this would be the electric motor. After you accelerate up to speed, when the only force needed is what is required to overcome drag and friction (flat land, no speed change), your most efficient power-plant is the one that has the fewest energy conversion steps, your ICE. It gets complicated when you factor in rolling hills, stop and go traffic and desired battery lifetime, but fundamentally the concept holds.

To say that using the traction battery too much will lower your MPG is only true if you are using it when you are not accelerating.

The key to getting good gas mileage is that you want to use the chemical energy stored in the gasoline at the slowest rate possible. You can do this one of two ways: Accelerate *REALLLLLYYYYYY SLOOOOOWWWWLLLLLYYYY* if all you have is an ICE, or defer the acceleration portion of your trip to a power plant that is much more efficient.

So consider the following scenario: You accelerate from 0-25MPH and then cruise at 25MPH for a period of time that is longer than it took to go from 0-25MPG. Ideally, the electric motor would have propelled you from 0-25MPH and then the ICE would take over to keep you cruising at 25MPH. A small amount of ICE energy would then be diverted during cruise phase to the battery to make up for the energy spent getting from 0-25MPH. Compared to the energy the ICE is using to overcome friction and drag, the diverted energy is probably only a small percentage, even if you take into account losses in the electrical system. The key here is that the charging phase happens at a much slower rate than the acceleration phase, which means the ICE puts out less power to cover that 0-25MPH than if it had driven it directly. This ultimately translates to less physical gasoline burned to do the same amount of work.

Keep in mind, we have not created a perpetual motion machine here. If you use 200 joules of energy from the battery, you will have to burn more than 200 joules worth of gasoline to replace that 200 joules of battery energy. However, you get to "spend" the gasoline to replenish that acceleration at a slower rate than it was used. Or put another way, you are still ultimately using the ICE to produce the energy for the acceleration phase, but the hybrid makes it the equivalent of a *VEEEERRRRRYYYY SLLLOOOOWWWW* acceleration that it "hides" inside of the cruise phase.

So am I saying that P&G is not real? Oh heck no, I think I even proved it further. The key to P&G is that you accelerate to an optimal speed quickly and then do not spend any continuing energy to overcome friction and drag. The ICE would ideally run during the glide portion only to slowly replace the energy lost during the pulse portion. If the glide is long enough and the ICE charges slow enough, it is roughly equivalent to a *VERRRRRRYYYYYYYY SLOOOOOOOOOOOOOOOWWWWWW* ICE-only acceleration.

 

Not to start a :flame:, but I cannot leave this as is...

True for a regular Otto engine.

The 2ZR-FXE Gen III Prius Atkinson engine gets around 35%.

MG1 and MG2 are about 95% efficient. This doesn't take into account energy loss from other conversions (battery charge/discharge).

This is an oversimplification of what an hybrid car is. The Prius efficiency comes from many aspects, one of which is to prefer the use of the electric motor when it is more efficient than to run the ICE in a non efficient manner. Other improvements, to name fews, are: low coefficient of drag, Atkinson engine, simulated CVT (PSD) to keep the ICE at its peak efficiency most of the time, use regenerative braking, variable speed electric A/C, variable speed water pump to keep heat in, etc.

Certainly true (95% efficiency is better than 35% efficiency). But you are forgetting that your battery charge comes from the ICE. You should therefore compare (in a simplified manner) 95%x35% with 35%. Then 33% is not better than 35%. This doesn't take into account conversion losses in/out of battery (which would make it less than 33%).

Most of the time (below 45 MPH if my memory is correct), the reverse is true: Use the ICE to accelerate while it can be used inside its most efficient range, then, when power requirement goes down and running the ICE chocked with the throttle closed would be inefficient (pumping losses), turn it off and use the motor with battery charge to keep your speed.

Unsure how you came to this conclusion. You're saying that draining the battery more, by accelerating quickly with the motor, and charging it back with conversion losses, is more efficient than avoiding conversion losses and using the motor only when it is more efficient than the ICE (at low power requirement)...

Your are ignoring the fact that an ICE is efficient only in a limited range. Accelerating very slowly with an ICE is inefficient.

I may not have the adequate level of technical background to confirm/infirm your reasoning but this contradict everything I have read from knowledgeable PriusChat forums contributors. Do you have any empirical data to back this?

Still don't understand your conclusion. Why would using the ICE at its peak 35% efficiency to generate 200 joules of energy to charge back the battery, with conversion losses, be more efficient than using the ICE at its peak 35% efficiency to accelerate the car forward directly with 200 joules of energy...

You really need to talk to Toyota's engineers as then have got it completely wrong by configuring the HSD to do exactly the reverse: accelerate with the ICE (second part of the HSI) and glide with the motor/battery (first part of the HSI).

 

"Normal" acceleration using only the electrics will also be a heavy drain on the batteries. This type of repeated heavy drain is harder on the battery pack and will have a detrimental effect on their lifespan in the long run. How much of an effect is hard to say, but it is something to be aware of...

 

Just to add to this, my understanding of expert Prius hypermilers is that they don't try to run in EV and will definitely accelerate using the engine in normal driving for precisely the conversion reasons Philosophe explained. They'll accelerate at a rate at which the engine is operating efficiently.

They make use of EV only where the ICE would be operated inefficiently: basically that means situations of low power demand like stop-and-go traffic and declines. In these situations it's better to use the EV because ICE_efficiency_at_charge * EV_efficiency > ICE_efficiency. EV is also advantageous in the glide of pulse-and-glide at higher speeds where, even though the engine is not using fuel, the battery is used to keep the engine spinning in order to protect the motors.

As GWhizzer mentioned, you also don't want to use heavy EV because deep discharging your battery will shorten its life. Hypermilers monitor the state of charge and make sure they don't use many bars.

Having written all that I can own up to use of EV just for the fun of driving with the car humming. Ironically, just after I last gassed up the car I drove 0.6 miles home without using the engine.

 

I use EV on the way to work. I start in a 30, go about a mile and a half (it's in warmup, so no EV) then down a hill. by this time the ice goes off, and i glide on battery through a village. i then go about a mile in a 60, then back to a 30. at this point i go into ev mode, over three roundabouts and one set of lights, and a bridge. that's about a mile and do it all in EV, end up with 3 bars or so left. then up a hill into a 60, and 8 miles of 60 after that (battery goes back up to 6 bars). get to work, and once on site stay under the ice threshold (so ev, but not EV mode) to find a parking space. way home is the same, except once back in my village i do go into EV, and usually get home with 3 bars. if i -don't- EV in the village i arrive with 5/6 bars, but next morning they've gone down to 3 bars anyway. so i would say ALWAYS EV close to your house, as that energy is already in your battery and will definitely be lost overnight if you don't use it!

My method yields ~62mpg, my girlfriends method of "just drive it" yields ~59mpg.

 

Good analysis. However, the flaw is that (I think) none of knows the actual efficiency of the ICE during an acceleration event from 0-25 mph (at some rate of acceleration) nor the same for the motors. The often cited numbers of 35% for the Prius ICE is some kind of average or some peak steady state value.

It could easily be that the ICE is only 10% or 20% efficient during the acceleration from stopped and the motors are 80%. And if the battery is recharged while the ICE is operating at 35%, that would be a net of 28%. So it is possible to get net mpg improvement while using EV mode and not be violating any laws of thermodynamics.

3PriusMike

 

is anyone buddy-buddy with an auto mechanic that has an emmision tester gizmo and a dyno or similar way to "move" the car without moving?

Might be able to make a graph of emmisions vs. accelerator pedal position vs. speed and extrapolate the efficiency from there.

 

This has been analysed already. See this thread.

 

Well I tried it on Friday and I think it is a great method. My SOC was down to two bars by the time I got to the parking garage. Which is the lowest reading ever. Normally it would be 5 or 6. But that is OK since in reality the SOC is much higher than what we see on the graph. Also, in my case, driving out of the garage (going down floors) the ICE is running anyway (warm up) so it might as well be charging the battery. It didn't seem the ICE ran any longer to charge up the battery, it just seem to charge at a higher rate (per ScanGauage).

 

Tom,

That's because spinning the generator to charge the battery back is not free. To generate electricity, MG2 as to exert torque against the ICE, which is then working harder than it would if the battery was already full (by not using it to accelerate). This consumes more gas.

If you accelerate with the ICE (middle of the right part of the ECO zone on the HSI), you will not have to recharge the traction battery (or only merely) as in your technique.

Seing arrows on the display doesn't tell you about how much current is going to the traction battery (and how much gas is burned to do so). If you don't believe me, buy a ScanGauge which will show you what I'm explaining.

Compare the two methods on a identical circuit, with an already warmed ICE, by resetting your trip meter at the beginning of each. If you're looking only at your instant MPG display, this doesn't tell you anything about your overall efficiency.

I'm saying that using the motor with (too much) battery power is inefficient. Whether your power request, with ICE off, is near the middle of the HSI indicator or to its right, both are not the most efficient way to accelerate a Prius. You should rather accelerate, with the ICE, with the HSI near the end of the ECO zone (not in the PWR zone, which gets the ICE out of its peak efficiency).

To put it simply, but using the HSI to drive the car, you should follow theses rules, for each of its quarters:
1. GOOD. Use to glide. Only a small amount of battery power is used and the ICE is off, as it would be inefficient at this very small power request.
2. NOT SO GOOD. More battery power is used and will need to be replaced, at a cost.
3. NOT SO GOOD. The ICE is used bellow its peak efficiency range.
4. GOOD. Used to accelerate. ICE is used at its peak efficiency.
Avoid keeping the indicator in the second quarter of the ECO zone or by using EV to extend electric mode over the middle of the HSI (too much battery charge is used and will need to be replaced later: inefficient).
Instead, adopt pulse and glide techniques (not pertinent on the highway, at speed over 45 MPH as the ICE will turn anyway to prevent MG1 from spinning over its designed limits).


This is from this excellent hobbit's piece on the subject (image from ken@japan, a contributor here)

I hope this clarifies things.

 

This has already been discussed. Unless you begin your morning commute with a long descent that could help you charge the battery back (regenerative braking), you should end your day with a full battery. No doing so tricks your trip MPG for that day and make it look nice, but you are simply reporting gas consumption to the next day. When you will charge the battery back up the next morning, it will be with a cooler ICE that will therefore be less efficient.

 

15 miles after the parking garage when I get home, the battery is charged.

There is no doubt that everyone's commute will alter the efficiencies/inefficiencies of this method. Apply one methodology to everyone's situation is narrow minded. However P&G is certainly a great place to start for hypermilers.

I plan to track this closely and see if there is any benefit.

 

"Context is everything."

I doubt that EV mode is useful in other situations, but in the one the OP described, I can see how it might be. The main reason being that stronger acceleration will cause the ICE to start, and in the low range of mph, when the ICE runs it also generates electricity (and the lower the speed, the higher the % of electricity vs physical torque). And if you're moving from stoplight to stoplight, you're getting plenty of regen from braking also.

What may be happening is that when using the ICE to accelerate, only a little power is going directly to the wheels while most of it is being translated to electricity by MG1, then sent to the battery and on to MG2 for propulsion. EV mode may be reducing losses by eliminating the ICE/MG1 portion of that equation - at least temporarily. And if most of that energy is recaptured by regen, the ICE/MG1 portion of the equation never comes into play.

 

I meant to have this out sooner, but I have been quite busy lately...

No problem, correcting misinformation is not the start of a :flame:. I am glad you spoke up.

Correct, for the actual power plants themselves. I got my 80% electric efficiency and 20% ICE efficiency from what I understand actually hits the ground, not the power plants themselves.

Everything I have read says that an additional 15% is lost from the flywheel to the pavement on an ICE. I figured similar losses happen for the traction motor too (Yes, I know you cannot just subtract percentages. I was working towards a rough estimate of actual traction motor efficiency at the wheels. Feel free to correct me.).

I spent a lot of time thinking about this over the last week and I think I identified the mistaken reasoning on my part. Unfortunately, it lead to more questions than conclusions. Perhaps I should explain and then hopefully you can chime in.

Short Path - Energy moves from the gas tank, through the ICE, CVT and into the wheels. About 20% efficient on average.

Long Path - Energy moves from the gas tank, through the ICE, through the electrical system, through the motor, into the traction battery, back out of the traction battery, through the motor, CVT, and into the wheels. (My original reasoning was incorrectly based on only half of the long path)

Here is where things get a bit murky for me. By my rough calculation I get about a 25% average efficiency in the long path:

ICE: 35%
Electrical System: Not sure, but I assumed 95%, maybe that is where I am going wrong.
Traction Motor/Battery: 95%
Traction Battery to Wheels: 80%

.35 x .95 x .95 x .80 = about 25%

Since these are only simplified average values, the only logical conclusion I can come to is that there are some driving profiles where the short path is actually more efficient than the long path. Otherwise, there would be no reason to have the ICE direct drive the CVT at any time (this is where my original reasoning stemmed from).

So now closing the textbooks and back into the real world... It appears that what you are saying is that it is more efficient to use the long path for gliding and the short path for acceleration? Or conversely, the short path is more efficient than the long path when it comes to accelerations and the reverse is true for gliding?


This also brings up another question... Are we approaching an ICE efficiency point where the long path will never be more efficient than the short path? Or is that point simply beyond the theoretical thermodynamic capability of an ICE?

 

I can't believe there is an argument about this. EV mode is a toy. In the right situation it MIGHT save a little gas. It's mostly cool for a quiet car coming home late or rolling around a parking garage.

 

You got numbers, charts, etc. to back up this claim?