Step on it! Have some fun.

Thanks.. nice analysis. But I think you may be ignoring the connections in the HSD. Also I think of hard acceleration as well past 55% throttle, I think of it as near 100% (or at least 75%). And I don't see the throttle/torque being at all the same. don't think its a linear relationship. Throttle can stay the same position but as the car speed changes the torque changes (as does the RPM.. as RPM and MPH are tied by a constant).


Since we know PiP in general the equations relate MG1, MG2 and MPH are
MG1 = 3.6 * ICE - 2.6 * MG2 (RPM)
MG2 = 59.1 * MPH

If ICE is off, then if MG2 is at 2000rpm, the car is at ~33MPH and MG1 is at -5200RPM, but even there I don't really know the torque split between them both ( I guess we could calculate it, but maybe someone knows a good source).

At 33MPH, its well past the bulk of the high-load acceleration stage. The "agressive" stage, in terms of torque demanded, is really what happens in the first 30mph and if you only do EV then understanding the actual torque and the split between MG1 and MG2.

It cannot be that hard for someone to actual do some measurements..

 

Pretty similar to my results.....driving like a turtle got me 17.2 miles.....driving 50mph same route gets me about 11-12miles. I tried hard accel then level off but watching the EV miles tick off that fast makes me think its using miles much faster.

 

Keep in mind that its not just the increasing drag that increases the energy used to drive faster. Its also that it simply takes more energy to accelerate a given mass to a higher speed. Again its basically energy consumed = acceleration * time of acceleration. Accelerating to twice the speed takes twice the energy because its takes twice the time at a given rate of acceleration.

If the observation seems to be that accelerating quickly but not driving faster results in less EV range, it would be interesting to know how much charge it takes to fill the battery back up in each case. Its possible there are other inefficiencies that come into play that make it less efficient to accelerate quickly, but its also possible its just a factor of the range estimating algorithm. For the latter case you'd expect to see the lower EV range case also take less ac energy to recharge the battery.

Rob

 

i did another trial tonite. but i tried to hold to the speed limit whereas, last nite i was accelerating so quickly, i wasn't letting off the gas fast enough and would wind up 5-10 mph over. similar acceleration as last nite, lower top speeds (more in keeping with what i usually do in my typical grandmother mode) and when i got home, 1.3 miles of range estimation left. so now, i'm thinking fairly hard acceleration is costing me 1 out of 17 miles of range or about 6%.

 

I think I did qualify the analysis as assuming a linear relation between throttle and torque, and rpm and speed, so you're quite right that's a simplifying assumption.

My understanding was that generally MG2 is providing the drive power, MG1 is mostly just spinning because its there and always connected, and the ICE is spun when needed to keep MG1 from over revving. If these equations are right, then MG2 RPM does seem to be linear with speed. That seems to make sense as MG2 is I believe still on the output side of the PSD.

You are right that torque varies at a given throttle position as rpms change. However, I think the claim is valid at any given rpm, which is I give different throttle positions at different rpm points. In a PWM controller, 50% throttle would equate to 50% duty cycle or 50% drive current output. At a given RPM torque output is essentially linearly proportional to current as long as the motor hasn't saturated. So I estimated throttle position at a given rpm as the torque on a given point in the contour divided by the max torque output at that rpm. That assumes that as this is a curve of combined inverter/motor efficiency, the top line represents the torque generated for 100% inverter output or 100% throttle.

So using the MG2 equation, the table above (filling in the 3000rpm point) would become:
17 MPH: 15-55% throttle/torque
34 MPH: 15%-60% throttle/torque
51 MPH: 24-71% throttle/torque
68 MPH: 32-90% throttle/torque

Now the relation between pedal position and what I'm calling throttle is a question for the Toyota software engineers. The ECU is going to map pedal position to throttle / torque requested in some probably non-linear fashion. I also seem to recall that it changes based on the selected driving mode. In eco mode you might have to press 80% pedal to get 50% throttle. In power mode it might be more linear? I'd bet there's a good chance that the Toyota software is already working to keep you in the high efficiency range under a wide range of pedal inputs. How that all relates to the power feedback display is also going to be up to the software. It should be possible with a scan gauge or the like to get a sense of pedal position, current (throttle/torque) and the power display all relate.

Agreed direct observation is always good, but is in this case hampered by the software between you and the hardware. That can make it difficult to interpret results because there's a translation step in between. In that situation I find it helpful to come at things from both a physical and theoretical direction to try and make sense of it all. I certainly don't claim to be an expert, just know enough about general EV stuff to look at the data and speculate.

Rob

 

BTW, more details on the testing used to generate the contours is available here starting on page 62:
http://info.ornl.gov/sites/publications/files/Pub26762.pdf

Its important to note that they bypassed the boost regulator and drove the dc section of the inverter directly with fixed voltages as noted on the graphs. The data we've looked at so far is only for the 650V graph. In normal operation I believe the boost converter would only put out this high a voltage at high speeds. That helps explain why the peak efficiency area seems skewed to higher rpms. Looking at the combined efficiency contour for 225V should give a better sense of low speed efficiency, as the boost converter should be turned off under those conditions. From that contour, I get torque/throttle ranges within 1% of peak efficiency of:

500RPM/8.5MPH: 20NM - 70NM out of 120NM = 17 - 59% torque/throttle
1000RPM/17MPH: 25NM - 70NM out of 120NM = 21 - 59% torque/throttle
1500RPM/25MPH: 25NM - 65NM out of 90NM = 28% - 72% torque/throttle
2000RPM/34MPH: 20NM - 65NM out of 80NM = 25% - 81% torque/throttle
2500RPM/42MPH: 20NM - 70NM out of 70NM = 29% - 100% torque/throttle
3000RPM/51MPH: 15NM - 45NM out of 55NM = 27% - 82% torque/throttle

Its also possible to estimate the efficiency penalty of max throttle (assuming the software even allows 100% pedal to translate to 100% throttle, possible in power mode, less likely in eco mode) at a given rpm:

500RPM/8.5MPH: 5%
1000RPM/17MPH: 5%
1500RPM/25MPH: 5%
2000RPM/34MPH: 4%
2500RPM/42PMH: 1%
3000RPM/51MPH: 1.5%

At some point here the boost converter will come online, making the higher voltage contours relevant. Its also important to note that with the boost converter in place the max current output will be limited, making the actual max torque output lower than whats shown on the 650V contour in particular. This will skew the max efficiency ranges higher by an unknown amount by reducing the max torque value in the denominator.

Rob

 

that's what i thought.

 

What does all that mean in English?

 

step on it! have some fun.

 

Who do you think you are, Bob Wilson?

 

Great data turtle, thanks! As you say, I think that fills in the missing piece of the puzzle.

 

How long does the PIP take to accelerate in loop 1 vs 2 to 35? I'm wondering how big of a difference we are talking about. Any one have the time 0-50mph?

 

Turtle may have real data, but based on his comments, 6 stop signs + initial acceleration = 7 acceleration events. 2:03 longer for the loop / 7 events = 17.5 s longer to accelerate to speed on loop 2 vs. loop 1 assuming same cruising speeds and deceleration profile on each loop.

 

the problem with acceleration is, you can't floor it due to ice kicking in. so our feathering it which means everone does it a bit differently.

 

i also wonder about long term battery health in constant hard acceleration vs gentle.

 

That can't be right. How long does it take to get to 35? I am asking because I am looking into getting a PIP, but this makes it seem that it is very hard to stay in all EV mode and get along with traffic.

 

oh no, it's got some kick to it, you can keep up with traffic just fine until you hit the highway. then you're gonna need the ice to go above 62.