Archives

Hi all. Just got an '07 Prius. I have a rather interesting commute and noticed that I always pick up a lot of SoC on the way home from work. Mainly b/c I park in a garage on the 3rd floor and use the 'B' gear to get out of the garage.

Anyway it peaked the question in my mind. How much is a SoC bar worth? So I ran some calculations but you guys seem to have a lot more info on this stuff than I, so I'd like to put it up for review.

NHW20 Prius has 28 cell battery pack. Each cell is 6.5Ah @ 7.2V. That yields 1310.4 Wh (26 green cars of regen). So each bar in the SoC display represents 163.8 Wh (3.276 green cars of regen).

According to the CalCars study (only data I can find), the Prius MG1/MG2 engines operate at about 262 Wh / mile (which I don't thing was P&G). So the gas consumption at that speed is about 40 mile / gal. Converting Wh to gallons of gas (Wh/mi * mi/gal = Wh/gal) shows that 1 Gallon of gas = 10.480 KWh. Or 1oz of gas = 81.875 Wh.

So this means that one SoC bar = 163.8 Wh = 2oz gas = 0.016 gal of gas.

So for every SoC bar I add in a trip, I will take off 0.016 gal of gas from my consumption count, and for every SoC bar I consume, I will add 0.016 gal of gas to my consumption count.

I know it doesn't sound like much but on my 15 mile commute it can swing my MPG calculations up or down 5 MPG. Gives me a better feel for what methods are working and which aren't.

I understand that the purpose of P&G is to get the battery out of the picture, but seems all that potential energy on the 3rd floor should be harvested. And on the way up... why not EV? I just needed something to offset the effect of adding or removing SoC bars.

I also concede that mucking with SoC effects the way the ECU makes MG1/MG2 based decision.

Anyway... let me know your thoughts.

References:
http://www.calcars.org/conversions-factsheet.pdf
http://en.wikipedia.org/wiki/Toyota_Prius

Interesting stuff. My constructive criticism:

Regarding the battery, the battery SoC indicator only covers the usable range of battery charge, which is much less than the full rated capacity of the pack. IIRC, the Prius will only allow the battery to swing between 40% and 80% charge. I don't know precisely how that computes in kWh, but I would assume that you can use 0.4 times the number that you came up with and be in the ballpark.

Another point, which is trivial, is that all bars on the chart aren't equal. I think that the 6th and 7th bars are worth the most, the rest are worth less by varying degrees. Somewhere on the net there is a chart giving the breakdown, which even changes depending on if SoC is rising or falling. Not worth worrying about with this calculation, though.

I don't quite understand what's going on when you calculate the GGE (gallons gasoline equivalent) for each bar. Maybe I'm just not reading it right? I think what you need to do is start with the energy content of gasoline and start soliving in the other direction. (Unless you did and I'm being thick-headed.) You also need to take into account ICE efficiency, which is going to be somewhere in the range of 25-30% for this motor (rough guess, but assuming that it's on the high side of the typical 20-30% estimate for real-world Otto cycle powerplants). That's an important step because the fuel has to go through the engine to get to the pack.

Finally, the most accurate calculation of this kind would have to take into account the electrical and mechanical losses incurred under regen. Not all of the kinetic energy bled into the motor-generator makes it to the battery and I haven't found a clear answer as to how much. The necessity of doing this is debatable, of course. One could argue that regenerated energy is "free" because it would have been "thrown away" by your average car. (Of course we know that any braking comes at a cost, but we could assume that braking was absolutely necessary.) One could also throw it away by making the assumption that none of the energy was regenrated, but rather supplied by taking power from the ICE. That might be the more valuable case for us since that's arguably the least efficient way to get power. There are still electrical losses but it's a less complicated path.

Good work, though! I really like seeing analyses like this.

Because of the 40% - 80% limits, you actually only get about 650 Wh
to play with. Eight green bars means 80% SOC, and what you'd
see as "dead empty" would be 40%. Paradoxically, six blue bars
is the nominal 60% SoC that the system always seeks toward when
other demands [ev-mode, regen] aren't placed on it. The gauge
is nonlinear, slow, and has hysteresis. Attila Vass did a nice
mapping of it, I think -- see http://vassfamily.net/ for where
he documents his stuff.
.
So that 600-odd Wh theoretically only pushes you a couple of
miles under ideal conditions; in real life it usually works out
to less.
.
Now, are you exiting the garage using "B" in EV-mode at less
than 20 mph, so you don't even light the ICE until you reach the
bottom? That would be the ideal way to do it, although you probably
want to start the warmup cycle and get through the first minute
[see the "five stages" (http://techno-fandom.org/~hobbit/cars/five-stages.txt) document] before hitting the main road
and putting a real load on the engine. And it's probably not
really worth EVing all the way back up to your parking spot
since that much rise will really take it out of the battery.
.
_H*

Cool... Thanks for the input.I don't quite understand what's going on when you calculate the GGE (gallons gasoline equivalent) for each bar.This is a method an old Chemistry prof taught me. The basic concept is that all ratios are re-representations of one. Since one the multiplicative identity, it doesn't alter value. So if x/y=1 and y/z=1, then x/y * y/z=1 and this yields (hint: the y's cancel) x/z also 1. So the trick the Chemistry prof taught me is that units can be thought of algebraically and canceled equally as easily.

Take velocity. Usually represented in miles/hour. Let's represent it in Qubits/Fortnight.

60 mi / hour * 24 hours / day * 7 days / week * 2 weeks / fortnight = 60*24*7*2 mi/fortnight

20160 mi/fortnight * 5280 ft/mi * 12in/ft * 25.4 mm/in * 1 qubit / 524mm = 20160*5280*12*25.4*1/524 qubit/fortnight

60mph = 61,916,746.3 qubits/fortnight.

So I simply multiplied the two efficiency ratios (both one) to cancel units yielding Wh/gal.

262 Wh/mil is equivocal consumption of 40 mpg (as the calcars report says).

262 Wh/mi * 40 mi/gal = 262*40 Wh/gal = 10480 Wh/gal

By doing this I get to take two equivocal emperical values and prevent the need for calculating the BTUs in gasoline and thermal efficentcy of the engine. At least as I see it.

262 Wh/mil is equivocal consumption of 40 mpg (as the calcars report says).


Ahh, OK. Dimensional analysis is a beautiful thing. It was the 40mpg equivalence that threw me off. (I should have been more specific...I didn't realize where it came from.) Thanks for the clarification, and please pardon the aforementioned thick headedness. ;)

Hi Dan:

___I have 0 time but an electrically driven gasoline equivalent is a bit off because of costs and an electric motor’s efficiency when it comes to energy consumption. I added my own thoughts below …

___The Prius II has a 1.3 kWh pack with an ~ 40% range of SoC (520 Wh). I know many here that would drive the Prius II a steady state at < 225 Wh/mile w/out any kind of P&G comps. The average driver would be around 250 Wh/mile and you can come up with the max EV distance from that. Also the Prius II’s SoC display slides depending on temps and is not linear.

___And ICE efficiency. A gasoline fed ICE can convert ~ 30 - 35% of the chemical energy to mechanical. When everything attached to the automobile is included, the conversion is down to < 18%.

___Now costs. If gasoline costs $18.00 per gallon would we use it? If electricity costs $4.00 per kWh, would we use it either? Not as they are currently sold. Since electricity costs are ~ $0.07/kWh here in Illinois and a PHEV-20 based Prius II can travel 20 miles on ~ 5 kWh’s, it costs just $0.35 to travel 20 miles on electricity. To drive a Prius II 20 miles on gasoline at 55 mpg and with gasoline costing $2.13/gallon, the calcs yield $0.77 to travel the same distance. I am not sure why you were asking about a gasoline equivalent but when it comes down to $ and ¢, electricity has it all over gasoline at its present price let alone the emissions of the same.

___About P&G. You want to use some SoC to extend glides but you do not rely on an inefficient conversion of maybe 35% mechanical/braking regen to pack power. Use the SoC as you come out of the garage but use it sparingly and only to extend glides or short distances between stops. Don’t just burn it up with a high power accel using both the ICE and Pack for assist is about all I can tell you. The Prius II will use that SoC without any intervention form you but when you want to use it, use it wisely ;)

___Good Luck

___Wayne

I am not sure why you were asking about a gasoline equivalent.By cost, I'm considering cost to FE. My goal is ultimately to record my overall power consumption for a trip. Basically a way for me to penalize myself (in my calculations) for reducing SoC. I could, for example, go full EV all the way from the exit ramp to my house. This consumes both SoC bars 5 and 6 (a very unwise thing to do). But if I went on pure gas/mi estimates it would have looked like a good idea from the trip by trip mpg calculations.

Conversely. If i start a trip with low SoC, and the ECU spends a lot of time redirecting power to the pack, I wanted some way to notate that even though a lot of gas was burned, some of it actually got stored.

The lifetime/tank/day MPG numbers end up taking care of themselves, but b/c of SoC, the trip tallies seem to be skewed. I'm planing on keeping both a MPG trip tally and an Miles/EnergyUsed trip tally. One simple way for me to do that is to say that a SoC bar is 2oz of gas (whether it is or not, as long as it's remotely close). If I add a SoC bar, I deduct 2oz from my Gas Used tally, thus my MPEU goes up. This makes since since I've put that potential energy in the bank so to speak. If I consume a SoC bar, then I add 2oz to my Gas Used tally so that I remember that the stellar MPG I got on that trip was at the cost of SoC (reflected in MPEU).

Without this, I could boast that my best MPG was 74, but it really wasn't, because I used 2 SoC bars that day. When I factor that consumption into an MPEU type calculation the adjusted MPG is closer to 50 (which is about what I'm hitting these days).

Simply a way to normalize my calculations to some degree or another. Really (for myself) only concerned with bars 5 and 6.

I appreciate the feedback and good information.

Hi Dan:

___Your adjusted segment mpg and lmpg is your normalized mpg. Gasoline in and miles driven received. Running .5 miles of a 3-mile trip in EV could skew a segments results but it won’t skew a tank and if you are driving a 3-mile segment, your FE overall would probably suck anyway :(

___As I have tried to convince Hobbit in the past, understanding the technical details is an excellent use of your time but driving for high FE day in and day out is where the rubber meets the road. It becomes more an art then a science for the top tier drivers. I hope we can get you into that top tier very soon … The tech details help tremendously but look at Jared’s and Tim’s back and forths in the daily grind threads. Get the darn thing up in temp in whatever manner possible and drive like you are on your last 2 oz. of fuel. They are pushing and the results in the temps they are experiencing is kick @$$ ;) Understanding the details will get you to a certain level but it is all about performing that gets you to the next. Before the Prius Marathon, I had all of maybe 60 miles in a Prius II and I don’t think much more needs to be added afterwards.

___Good Luck and we look forward to your ever-increasing FE posts :)

___Wayne

Heh ... Wayne, I *did* try to drive to maximize IFE while
steady-state on the highway, and it sorta backfired on me.
I really need to go finish that final writeup from my winter
vacation... it's still hanging while I'm working on this
other stuff.
.
_H*

Hi Hobbit:

___I remember you seeking 60 at 60 or was it 65 at 65 with the science projects. I know how that turned out :( Drive the Prius II to its strengths and we know where that is. Watch what Jared and Tim are achieving in temps that should make it all but impossible to bust the EPA. Remember what Dan Kroushl achieved at HF2006? We need to go for a long ride together and we are going to cover up all the science experiments other then an SG-II and the built in Energy and Consumption displays to see what you can do when the tech isn’t getting in your way ;)

___Good Luck

___Wayne

That's one reason I'm pushing more for an "MPG training" event
at HF2007 rather than an "MPG competition" ... go out in pairs,
and spread the wealth...
.
_H*

Hi Hobbit:

___HF2007 needs the big number publicity of the MPG Challenge to grab attention from outside the community which I wholeheartedly agree with! We are all hybrid drivers even if we don’t currently own one ;)

___Fortunately, the HF 2007 committee members are working on a tech session which includes hypermiling clinics! I have suggested Justin and I perform as many as we can but due to time constraints, we might have a little trouble keeping up with all that may be interested? Any number of CleanMPG members guaranteed to show would be great at instructing but not everybody is cut out for it. I am looking forward to it but there is no guarantee this will happen just yet. If you do come out this way for HF 2007, you are more then welcome to stay here again as a layover before hitting Madison on Friday. You would be a great instructor but you will have to come down off the “deep inside the Prius II” tech talk as it will bury the average driver in details they will not be able to understand. Believe me, I have enough critics of my own hypermiling double speak given very few actually know what DWB, DWL, FAS’ing, or P&G is let alone the Hypermiling 101 Clinic with driver instructions being barked out similar to the staccato of a machine gun! Add in S1 - S4 and a WAI on a poor and unsuspecting soul and he or she will forever be completely messed up as to what to do and when to do it :D

___Good Luck

___Wayne

Hi Hobbit/Wayne:

If it's feasible, I think having both the MPG training and challenge is a good idea... get people excited/draw press/etc about what's possible, then show them what they can go out and do *right now* to start seeing those kinds of numbers themselves.

Having "basic" vs "advanced" sessions is a good idea too. I for one would enjoy a “deep inside the Prius II” session by Hobbit. Those highway numbers you were putting up (high 60s??) are incredible. :cool:

_Jerad

Hi Wayne,

FYI, electricity is more expensive in Maine, Bangor Hydro Electric Co is $ 0.15/kWh. By your calculations electric would still be cheaper, but only by 2 cents. This is not to say that electric cars would be a bad thing, I am all for them if the electricity is produced using renewables or maybe "clean" coal where all emissions are sequestered.

Dennis