Stanley, you do not understand how Toyota engines work. They use pure electric power most of the time, with the engine running only to supply the electricity. When cruising at fixed speeds, the engine rarely if ever directly drives the car. And it turns out (based on observations, i.e. evidence) that it is more fuel-efficient to cruise in EV mode, then let the engine run later on to recharge the batteries, than it is to let the engine run all the time. Remember that ICEs have efficiency peaks; if they run too slowly, their efficiency actually lowers. So for low-load situations, it is inefficient to run the engine when you can use battery power, then run the engine later on with a higher energy output to recharge the batteries.

EDIT: Oh, this thread is larger than I realized. I only read the first page; apologies if my post is redundant.

 

CGameProgrammer — I'm afraid that about the only correct statement in your post is your final EDIT. The TCH (and indeed the Toyota Hybrid Synergy Drive system itself) CANNOT have the car in motion with the ICE running without the ICE contributing mechanical power directly to the wheels. It's a mechanical impossibility. Examine the behavior of the planetary gear set (Power Split Device) — see the references I cited earlier in this thread. The TCH never uses "pure electric power" when its ICE is running, and it is incorrect to say: "When cruising at fixed speeds, the engine rarely if ever directly drives the car." The whole question that I'm trying to address here is whether you (and others) are actually correct when you say: "So for low-load situations, it is inefficient to run the engine when you can use battery power, then run the engine later on with a higher energy output to recharge the batteries." I contest this statement, and I have provided data to support my claims.

Droid13 — Can you actually provide some meaningful data to support your claim that, when using forced pure-EV mode you CAN indeed later recharge the NiMH battery under conditions that result in BETTER fuel efficiency than had you not forced pure-EV mode in the first place? I have already provided data to show that this is probably wishful thinking, not fact. To address your claim that a 40-mph comparison isn't meaningful (I say that it IS meaningful), I have now used the data provided in Miller's paper to compute the net benefit at all the speeds for which he gives data for the Prius I — it's always a NEGATIVE benefit — and here are the results (I repeat the previously given 40-mph data for comparison purposes):

1) Running in pure-EV mode at 40 mph and then recharging at 40 mph consumes (2459 - 1618) = 841 g/h more gasoline than just running in normal ICE mode at 40 mph, where the consumption rate is 1618 g/h. This makes the fuel consumption 52% WORSE.

2) Running in pure-EV mode at 40 mph and then recharging at 50 mph consumes (2109 - 1618) = 491 g/h more gasoline than just running in normal ICE mode at 40 mph, where the consumption rate is 1618 g/h. This makes the fuel consumption 30% WORSE.

3) Running in pure-EV mode at 40 mph and then recharging at 60 mph consumes (2408 - 1618) = 790 g/h more gasoline than just running in normal ICE mode at 40 mph, where the consumption rate is 1618 g/h. This makes the fuel consumption 49% WORSE.

4) Running in pure-EV mode at 40 mph and then recharging at 70 mph consumes (2047 - 1618) = 429 g/h more gasoline than just running in normal ICE mode at 40 mph, where the consumption rate is 1618 g/h. This makes the fuel consumption 27% WORSE.

5) Running in pure-EV mode at 40 mph and then recharging at 80 mph consumes (1757 - 1618) = 139 g/h more gasoline than just running in normal ICE mode at 40 mph, where the consumption rate is 1618 g/h. This makes the fuel consumption 9% WORSE.

You see, although 40 mph is a worst-case scenario, there is ALWAYS a net LOSS by doing this. You're NOT better off, because you CAN'T actually achieve a POSITIVE net benefit by recharging the NimH battery at some more favorable speed. Only if you could jump to 80 mph to recharge your battery, after a spell of forced pure-EV mode at 40 mph, would you almost (but not quite) break even! [The ICE is running near peak thermodynamic efficiency at 80 mph.] You can't actually do this, of course. So, in the absence of hard data to support your case, I'm afraid that I have to reject your speculations.

Stan

 

Very good read Stanley!!! I think your about the only one that has this figured out lol. keep up the good work.

 

Wrong. If you allow MG1 to be open circuit/free spinning, then you can direct virtually 100% of whatever energy is being produced by an idling or faster running ICE into spinning MG1, with virtually 0% going to the ring gear. The only residual energy is from the very small PSD frictional losses, and any MG1 EM forces at open circuit -- nearly zero. In this case, MG2 is free to drive the wheels as though the ICE and MG1 were not even part of the system.

As to the rest of your very excellent data (I mean that sincerely), you are somewhat misapplying it with near static assumptions (such as constant loads or speeds or accelerations, etc.). In the real world, the PCU is making lots of optimizations without all the relevant data. For example, while the HSD may have the optimum MG1/MG2/ICE balance for the given speed/acceleration/load/SOC/etc., all of which may of be first order significance to the energy optimization, it does not have other first order inputs dependent on what is 1 sec or 1 minute ahead; did the driver just notice a yellow light ahead in his, or the cross traffic? Did the driver just decide that there is plenty of extra time for this trip? Such considerations can be taken into account by the driver, who may be able to realize an additional optimization of the energy equation, but your formulas cannot account for them.

Stan, I am not arguing that it is, or is not possible for a human with a deep understanding of the HSD system to better optimize the SOC/ICE tradeoffs for long term efficiencies, but I am arguing that it is impossible for you to prove it is not possible. You are over-relying on your insufficient (but otherwise excellent) analysis.

Personally, I believe Toyota did a magnificent job optimizing the control strategy for energy optimization given the assumptions they needed to make, and minimizing the extremes of errors caused by those assumptions. That is, if the HSD were designed for urban San Francisco rush hour driving, it would have a different optimization than it would for rural Florida weekend driving, but I suspect the HSDs of any given model have the same program delivered to both dealer destinations. Given that necessary compromises were made, which included not only the programming parameters, but also things like the NiMH cell capacities and voltages, and MG1 and MG2 KVs, etc., that I think there is lots of room for energy improvements to be made through human manipulation of the program by forcing or avoiding EV mode contrary to the program. Your arguments assume the program is "optimized" and forcing EV against the program is necessarily going to be a loss of optimization. Even if this were a linear system, which it is very far from being, then you would still be wrong per your analysis. The many non-linearities make your analysis all the more difficult to accept. Driving conditions routinely impose non-linear inputs to the energy equations, and these are the same kinds of inputs that humans do a remarkable job of handling sub-consciously. It is not too unreasonable to allow that the driver might also make good energy optimizing decisions, even if the conscious understanding is wrong, and even though the substantial evidence that we do get better results though forcing EV is completely anecdotal.

That is to say, you could be right, but you are far from proving it. And, you could easily be wrong. Your position is far from factual.

-- Alan

 

Just remember one thing, energy is neither created nor distoryed. All the energy that the TCH uses is from gasoline and recovered energy. The ICE is more efficient than most gasoline engines, the fact that the recovered energy allows one to travel without the use of the ICE, and last but not least, the ICE only runs when needed, allows one to achive high FE.

 

alan_in_tempe — You are quite right! I apologize for overlooking this possibility, which could apply if the ICE was warming up, and doing nothing else, while MG2 was propelling the car using battery power. But this is pointless, since the ICE is then using gasoline to no effect. Does it ever occur in the TCH? I think not, because it wouldn't make sense from any point of view.

The rest of your points are also well taken. I have been arguing from the only hard data that is available to me, namely Miller's paper, which may or may not contain errors. Yes, this means that I cannot address all possibilities; and, yes, it is possible (but I think unlikely in most cases) that in individual circumstances the driver might be able to do better than Toyota's generic optimization. But, what this thread is addressing is the belief among many TCH owners that the more pure-EV mode driving they can force, the better their fuel economy will be. This "feels" right — after all, the car is so quiet when the ICE is "off" and using no gas at all, that one surely must be saving gas overall!

Everybody — One way to try to resolve this issue would be for the proponents to conduct a series of careful experiments and report the results to us. This could convince me. But, to have any statistical significance, enough trials must be run, and with appropriate controls, which are difficult to achieve. This is what I believe would be the best way to do the experiment. It should be conducted by people who have a regular unbroken commute of at least 30 km (~20 miles) each way to and from work every weekday. (This reduces the contribution of the warm-up period to the data.) You MUST (most important!) drive equally moderately whether trying to force pure-EV mode as much as possible, or just letting the TCH do its own thing. Do this for two weeks (10 return commutes), and alternate (i) "forced pure-EV mode" driving BOTH ways on one day, with (ii) "Toyota's choice" BOTH ways the next day. This alternation will help reduce the effects of temperature and weather changes on the results. At the end of each 1-way trip, read and note down the trip fuel economy display readout (be quick — it only stays for a second!). Average the "outbound" FE with the "return" FE to get the day's FE figure. After two weeks you will have accumulated 5 return trips' worth of FE data for each of the two modes. Average the five type-(i) FE data sets, and the five type-(ii) FE data sets separately. How much difference is there between these two averages? Report all the data to us at the end of the 2-week period.

Any takers?

Stan

 

Stanley, you act like we purposely run in EV mode. It happens automatically and is out of our control. The only way to ensure the ICE doesn't turn off would be to use a heavy foot on the throttle, which obviously just wastes gas.

It is reasonable to assume Toyota programmed it this way because their testing found it to be most fuel-efficient in average situations, though it's possible it reduces pollution more than gas consumption... that was the intent of the original Prius after all.

 

I have found the driver has a great deal of control over EV mode. For example, assume level ground and a desired 40 MPH: Just accelerating to 40, and then setting cruise, the TCH is unlikely to go into, or stay in EV. However, accelerating to 41 or 42, and then going completely off the accelerator will cause the TCH to go into EV if the SOC is fairly high (~over 75%), while the car will drop 1-2 MPH in the process. Then with the lightest possible accelerator pressure to "barely" hold 40, you should be able to keep it in EV to well under half SOC, which could be many miles (while small road undulations or wind gusts might be enough to knock you or the cruise out of EV mode otherwise, and the TCH is slow to return to EV on its own).



I like Stan's idea on how to test the overall efficiency of forced EV. I would like to add a minor suggestion: keep track of the time of each trip (with a stopwatch, preferably). This is only minor check that the average driving speed is similar. My biggest concern for this experiment is that given the described setting, the shut-down reporting of FE is not likely to be usable in that this kind of driving (with or without forced EV), as it is likely to have FE over the max possible dash gauge reported 42 MPG. This experiment may have to be limited to those with the nav package who will reset the digital FE for each test run (in which case the round trip can be measured as a single value as there would be no need to measure each direction and then average, except to exclude mid day sorties).

-- Alan

 

I have a commute that is approximatly 75 miles one way. I do this about 4 days a week, and some times 5 days a week, so I would be perfect to do this "experiment", but my only concern is how do you take into account traffic variations. About 65 miles of my commute is almost no traffic or very light traffic, and at 60-65 MPH. The other 10 miles is on a road that has a 45 - 50 MPH speed limit where on day traffic is traveling at 35 MPH stopping at every stop light spaced approximatly 3/4 to 2 miles appart. The next day traffic could be travveling at 55 MPH with all green lights.

In addition, the wind can change from a south wind to a north wind on a daily basis. You could easily have a north wind on Monday, a south wind on tuesday and a north wind again on wednesday. Then the wind is usually a lot stronger on the trip home than the trip to work as the sun has not come up yet to start the thermal winds for the day.

It would be a great experiment to run, but I am affraid there would be way too many variables to take into accout.

The only good part is I have the nav, so I could document the results very well.

Also, I find it very easy to force the car into e-mode when it would not normally go into e-mode, with the exception that I have noticed when going to work after driving at highway speeds for a bit, I can't force e-mode and when I come to a stop, the engine does not turn off, and the energy management screen shows no energy flow. The FE gauge just sits at 0 till I start moving again.

 

Sounds like U B DA man!!!!!!!