Step on it! Have some fun.

I tend to provide exposition, to prevent newbies from falling into the assumption trap we've seen all too often. It gives the impression of me wandering off-topic, but there really is a method to the madness.

 

While the motor curves are intersting, different motors can behave quite differently, and its not just the motor as the battery can be less efficient at high current. While the OPs plots are interesting its not clear it relevent -- the discarge on the PiP can be more than an order or magnitude higher then the battery plots provided. (using 2.4kWh in about 15min). However, I do agree that overall, the cost of hard EV acceleration is not near as wasteful as the cost of aggressive ICe acceleration.

It would be interesting for someone with a OBD2/Torque setup and a PiP to actually measure the difference in SOC for an agressive acceleration and a normal acceleration. I did such measurements to quantify the cost of agressive acceleration in the Volt (discussed in this thread,) which was about .2 to .3 miles of EV range (depending on what a full charge give you). that is given up for an an agressive (0-60 in ~9 seconds) compared to doing the same 0-60 in ~20 seconds.

 

Can you really accelerate aggressively in a PIP with out the ICE kicking on? I guess it is relative.

 

definitely. you can watch the hsi mirror the throttle position and as long as you can keep it out of the power zone, the ice won't start. is that aggresive? it's not a tesla, but it is compared to the way some of us normally drive.

 

we need to get bob wilson one of these babies for a few weeks. or his prote'ge', FL8.

 

I got both (Torque and ScanguageII) but don't know how to work those babies.lol

 

There is a combined motor/inverter efficiency contour for the 2010 Prius in the lower right on page 10 of this ORNL report:
http://www1.eere.energy.gov/vehiclesandfuels/pdfs/merit_review_2011/adv_power_electronics/ape006_burress_2011_o.pdf

I'm not 100% sure I read that correctly, but I believe you'd draw a curve from the origin passing through the left-most bulge of each contour to map out a max. efficiency acceleration curve. From that I'd say that at <2000 RPM you want to be at ~30% power, increasing to about 60% power by 4000 rpm, and about 75% percent by 6000 RPM. Now how those RPMs and power levels map into speed and throttle position is a whole different question given the variable gear ratio and ECUs mapping of throttle position to power requested and gear ratio. If its possible to get the software to hang MG2 right around 6000 rpm with pretty heavy throttle through your whole acceleration that looks pretty ideal.

I would say the graph supports the original claim. It seems like being too lite on the throttle during acceleration would have you spend much more time in low efficiency regions than being too heavy on it. Even at the max torque/power line efficiency is not that far off the ideal line. The Li battery should be pretty ideal vs. current output, but wiring losses could be a factor. Measuring the voltage sag at the max output would give you an idea of how much of a factor that would be.

Rob

 

I did a test tonight and learned some things....Ive only had my PiP a week or so and my typical 13 mile drive was just barely able to make it on EV. My road is a 55mph country rd with no stops for 5 miles, then 1.5 or so stop and go.

when i would run 50mph down my rd....i would not make it home without ICE starting when I was about 2 miles away.


So today i took a slightly longer route of 14.5 miles round trip, however the road has less traffic and I was able to go much slower at 35mph on average. Going these slower speeds I made it there and back and still had miles left, so i drove further getting my best total EV range today of 17.2 miles!

So the way i see it, the slower speeds and lower acceleration moved me from approx 12miles on EV to 17.2 on EV, all from just going slower. Now this makes sense to me...but i had to try it for myself. So much of this talk about stepping on it makes little sense to me, when i step on it the EV range drops super quick.

Man I love this car!


two pics attached.....
PIC 1: 445 miles total on my new PiP and MPG is now at 98mpg

PIC 2: Even at 17.2EV I had .1 miles remaining!

 

I have tried this every night on my drive home from work for the past month. It's frankly to hard to replicate on my route. I always run out of EV within about 1/2 mile regardless of a heavy throttle night or a light throttle night. I find the amount of stop lights I hit is what causes the 1/2 mile discrepency.

My personal opinion based on my results is creeping up on the throttle vs. accelerating at 75% (in a 45 MPH zone)are so close other variables will wash out the difference.

 

Interesting plot.. but I don't see it supporting your conclusion about hard acceleration. When you start at a stop (i.e. at RPM) if you accelerate fast you are requiring a very high torque, so are way at the top of the chart which is an inefficient region. if you accelerate slowly you have a more moderate torque so more in the lower part of the plot as the RPMs increase. (Yes too slow is bad, as well as too fast).

Also unclear from the slides what motor this is, though I agree it is likely MG2, but that alone would not define total torque.. so that is a tad confusing. Maybe it was with the ring gear freely rotating, so all power from MG2.

 

17.2 of EV range is awesome. I also agree with your conclusion that EV range drops super quick when stepping on it.

I think a discussion about the efficiency of the electric motor is one thing. But efficiency, in this context, doesnt mean much if you're trying to get maximum EV range. Discussing a motor's efficiency simply means that using more electricity (harder acceleration) will translate to an "equal" amount of acceleration (i.e. no wasted electricity). Meaning if I use 1W of electricity, I will get "1W worth" of acceleration. If I use 2W of electricity, I will get 2W worth of acceleration. But ultimately I am still using more electricity, thus decreasing EV range more than if I accelerated lightly.

 

I think that what is not intuitive to many, based on their experience with gasoline engines, is that lead-footed acceleration in EV mode does not necessarily use that much more energy to reach a particular speed than a light foot. In a 100% efficient system, the acceleration is irrelevant; you need to expend a certain amount of energy (kwh of electricity, in the case of EV) to get a given mass to a given velocity (E= ½ mv²), and it doesn't matter how quickly you do it. What is being discussed here is, in the real world, how much do you lose due to inefficiencies in the system (like resistive heating in the wiring), by applying a high torque (and thus higher amperage) for a shorter time, vs. lower torque for a longer time.

 

sweet! that's what i have been getting with similar driving. one thing i don't get is, you say slower speeds and lower accelleration. but you only mention slower speed. if you only accelerate twice each way on your commute, your probably not testing the o.p.'s premise.

 

There are more detailed reports from ORNL about the test conditions for this graph, but I'd need to go back and look at them. I believe this is a bench test with MG2 removed from the transaxle. I don't know if the inverter data is taken independently and then combined, or if the combined data is a test of the two together.

One of the major complicating variables in applying this data is the behavior of the PSD during acceleration. If this was a traditional EV with a fixed gear ratio, the ideal acceleration would be as I described. Fairly light throttle at launch, with increasing throttle as RPMs increase. However, I don't know that with the PSD in action speed, throttle position and MG2 RPM are linearly related. There is an intervention there of the Hybrid ECU to convert throttle position to torque demanded of the motor and effective gear ratio.

If we assume they are linearly related, I believe the graph indicates that too much throttle is more efficient than being too light on the throttle, but that in general the efficiency bands are pretty wide allowing similar efficiency for a fairly wide range of accelerations. At 2000 RPM for instance, the ideal about 60NM or ~30% torque (throttle). However at double that or 120NM/60% throttle, efficiency only degrades from ~88% to ~87%. Even at max throttle it degrades to about 78%. By comparison, at 1/2 the ideal torque (throttle), or ~30NM or 15% efficiency is still 87%, but starts to fall off quickly below that. So essentially at 2000 RPM, there <1% difference in efficiency between 30NM/15% throttle and 120NM or 60% throttle. At 4000RPM, the efficiency band shifts further to the hard acceleration side. The ideal position is at about 80NM or 57% torque/throttle. The 1% down efficiency band spans from about 45NM to 125NM, or 32% to 90% torque/throttle. At max throttle efficiency is only down by 1.5%, vs. at 10% throttle its down 10-12%. At 1000 RPM, peak efficiency is about 80% at 55NM or 27% throttle, and the 1% efficiency loss band stretches from 30NM to 110 NM or 15% to 55% torque/throttle. In this case max throttle results in about 66% efficiency, while 10% throttle results in ~60%.

So to summarize that, the Max Efficiency to within 1% bands are roughly:
1000 RPM: 15-55% throttle/torque
2000RPM: 15%-60% throttle/torque
4000RPM: 32-90% throttle/torque

All other things being equal, acceleration starting at about 50% throttle and increasing to 90% throttle should be just as efficient as steady acceleration at 32%, or light acceleration from 15-32%.

Motor/Inverter efficiency should be the dominant factor in energy usage. Assuming uniform efficiency, it doesn't take any more energy to accelerate up to a given speed quickly than slowly. Remember that F=mA, where torque is basically Force, m is mass and A is acceleration. Torque is proportional to Amps, and at a given Voltage also electric power (kW). The amount of energy consumed, or work it takes to accelerate a mass up to a given speed is the force times the time the force is applied, or the power (kW) time the time = kWh consumed from the battery. With uniform efficiency, using twice as much power (current, torque, force) will accelerate to a given speed in half the time. Since 2*F * 1/2* t is equal to F*t, the amount of work done or energy consumed for both accelerations is equal. To account for efficiency, mechanical power is basically proportional to electrical power * efficiency. If efficiency is the same for the two case as it is above, then 2*kW*1/2*t*Eff = 2*F*1/2*t = F*t = kW*t*Eff. Electrical energy consumed is still equal between the two cases.

One more variable that comes into play in the PiP, is the algorithm that the ECU uses to estimate range. Depending on what that algorithm is, it is certainly possible that the higher current used in the faster acceleration even though used for less time, will cause the estimated range to be lowered. To test this you'd have to see if the reduced range is accompanied by a reduction in charge accepted by the battery. This would indicate that the ECU is underestimating range because of the higher usage, rather than indicating that harder acceleration is actually using more energy. If that turns out to be true, it could be the case that lighter acceleration at the lower end of the peak efficiency band could help convince the ECU to use as much of the battery capacity as possible and extend the effective range.

Whew.

Rob

 

Well there are several stops in the mix where i really dont accelerate hard ...I agree a quick accel then level off here and there is not gonna kill too many EV miles, but if you accel hard all the time and not get it back to low range on HSI you can just watch the EV miles tick away fast!

I just am glad to know the initial projected range for EV is not grained in stone, it all comes down to how you drive it that session as to how many ACTUAL miles you get. Really is the most interesting car to drive, I love trying to use EV then HV on n off for each type of speed and accel condition.

Dan

 

Aerodynamic drag is proportional to the square of the velocity, thus the 2.5 number. So as you point out, the target speed very definitely matters due to aerodynamic and rolling friction, but how quickly you get to that speed really doesn't matter that much, as long as the gasoline engine doesn't start. So you can peal out, as long as you don't get above 35mph, and stay in EV mode.

 

okay, here goes: i did my first, completely unscientific test of o.p.'s theory tonite. my daily commute, which i have been doing in ev for 4 months now is 14.8 miles and i usually arrive home with apprx. 2.5 miles of ev range left, no a/c use. tonight i accellerated as quickly as possible vs my usual gentle, slower than molasses routine. this takes some practice, and it's difficult to get close to the power bar and not start the ice by accident. i was mostly just a little to the right of the green ev car icon. another thing is that it takes some practice to accellerate like this and not go faster than you usually do. so this isn't perfect by any means. but when i got home, i had .5 ev range left. given the variables i encountered, i'm guessing that hard acceleration vs easy cost me about 10-15% of my ev range. i could be completely wrong though, and i have no idea where i was on the motor efficiency curve. i would also like to thank the o.p. for bringing this up because it was a lot of fun and i was surprised that i was still able to do it all in ev!