Living Dangerously: The Fuel Light

We're on the same team:

There was a five year period I was working in Washington DC while the wife kept our house in Huntsville. So I would leave Huntsville with just enough to reach the Georgia cheap fuel stations just south of Chattanooga. Then I had enough gas to reach Southwest Virginia 50 miles beyond Roanoak where the next cheap gas was found. From there, it was a straight shot to the DC area. Two fill-ups for a 735 mile, 14 hour trip in a 1991 Camry with a partial tank left over at each end.

For three years, I owned a Cherokee 140 and fuel planning with reserve was not optional. I always flew expecting the engine to quit at any time. So running out of gas in a car ... plez!

I also rode a Yamaha 550 with no gas gauge. There was a 'reserve selector' which is how I knew it was time to fill up in the next 10-15 miles.

Bob Wilson

 

Commercial airline pilots know their fuel range to what uncertainty? Roughly 10 minutes? Airlines would not tolerate the tank and gauging uncertainties we put up with in common cars. That would have them wasting huge amounts of fuel just carrying around deadweight, forgoing considerable potential cargo capacity and long routes that should be practical.

While a car's fuel consumption rate is more variable than that of an airplane, the consequences of running out are much less serious. If pilots can figure it out, why not drivers?

Shortly after my parents teased me for days about fearing running their car well below E, as they continued running it even lower, my first car ran out with its gauge still above E. It had short fuel range, and no hope of traveling from my after-college home to my parents without getting some overpriced fuel along the way.

The second car could theoretically make that trip on a single tank, getting reasonably priced gas at one end, and 'free' gas (dad's farm tank) at the other. But the gauge 'E' mark didn't agree. So I intentionally ran it out to see how much margin was below that mark, a test that ran more than two hours after hitting 'E'. Though that margin was narrow for the full trip, it was plenty to bridge the span between reasonably priced fuel stations. And if another Energy Crisis were to crop up, it could make that trip even on days when I might be prohibited from buying fuel. (No 'free' farm gas now, the tax man would disapprove.)

Car #3 covered the whole trip very comfortably. But #4 raised alarms well before reaching the affordable fuel, until I figured out that it had even more of its true range positioned below the Empty mark than did #2. Even its low fuel warning light, which still had plenty of margin, didn't light up until long past E.

By the time Prius came along, this fuel planning process was already deeply ingrained. I see no reason to stop, the longer Prius fuel range just gives me more flexibility for taking advantage of pricing differences. Not everyone has the engineering or math or puzzle solving interest to do this. But for some of us, this is actually fun.

And I do set aside greater safety margin in winter than in pleasant spring conditions.

 

In my opinion, the DTE, cruising range estimate is conservative but actually pretty accurate, that is, on my Gen 3 Prius it always reaches 0 with about 2 gallons left.

And if you think about it, it can never read entirely consistently because of the very nature of how a Hybrid works.
That is we are driving a vehicle that operates with the "synergy" of a gasoline engine and a electric hybrid motor and components. To get from point A to point B, it uses BOTH gasoline and electricity and it blends them based on numerous factors, some under are control as drivers and some out of our control and environmental.

As far as I know there is no way Toyota could offer a system that could predict how long or short are drive is going to be, or how cold it is outside, or how many hills we may be going up or down on any given trip. Since all these factors and more can affect how the gasoline engine is working, I've been consistently impressed with how consistent my DTE and cruising range actually is.

How "quickly" it reads off miles to zero will change but it's pretty consistent once it reaches zero. I typically fill up once mine reaches zero, and it's ALWAYS been 1+, nearly 2 gallons left.
If it wasn't consistent with this, of dipped below having a gallon or more left, I wouldn't use it as a tool. But it's consistency is exactly why I use it "usually" as my primary tool in determining exactly when I need to pull in and fill up. I say usually, only because I don't play games, and sometimes I'll just fill up and reset immediately once reaching the last pip or so. No medals given for running the Prius to empty IMO.

So anyway, given the nature of a Hybrid vehicle, I've actually been impressed with how accurate and consistent the gauges and tools in my Prius really are..
I don't expect some what would have to be magical system that could predict exactly "how" my drive is going to unfold and thus give me a perfectly exact prediction of how many miles I have left.

 

Why would the 43L (I don't know how many gallons that is) specified in all customer and technical documentation be false/wrong? Why should Toyota not give you an exact true number?
I do trust the fact that they in fact as many as declared. Any fluctuations are due to change in volume of the fuel itself based on temperature (though I would like to see a volume profile vs temperature for E10 gas) and how much gas the gas station is pretending to give me at the pump. Them I don't trust, not the 43L of Toyota.
And considering all sorts of inaccuracies of the MFD and all the above mentioned, I first assess over many tanks and different temperatures, what is the MFD error, and then calculate the L left based on the MFD current tank fuel consumption and this error. From that, I use pessimistic, realistic fuel
Consumption to know how much gas left/range I have. And based on how many gas stations there are around, I drive accordingly.
I have to say that finding gas in Europe is usually not a big issue.
Also, if the DTE is going most of times at 0 with 2 gal left, or about 7,5L that is a lot for EU distances. That is about 15% of total tank available (43/45L gen4/gen3) and even if one did an average pessimistic consumption of 5,5L/100km (43mpg) that would yield a range of 136km (about 80mi), and that's a LOT of range for EU gas station distribution.

 

They're required to land with a 30 minute reserve. That means they have to carry enough fuel to land with a 30 minute reserve after a flight that might have them holding on the runway waiting for takeoff for an hour, holding in the air for longer than that, and then landing.

One time, I had a 4 hour flight take 5 1/2 hours in the air due to weather. We still landed with a 50 minute fuel reserve.

 

43L is 11.36 gallons
Toyota's US website says 11.3 gallons

 

I am not a spokesman for Toyota but I'll offer some thoughts:

• Why would the 43L (I don't know how many gallons that is) specified in all customer and technical documentation be false/wrong? - In engineering, we know all measurements have a tolerance, a +/- offset. So with my Gen-3, I feel very comfortable with the precision of my measurements. I also know how temperature, density, and energy-content can change. Just I want mastery of my vehicles and that is what I do … measure and model. For example, my mph vs MPG (km/h vs L/100 km) chart. I started with the EPA roll-down coefficients; known physics and chemistry; published thermal dynamic efficiency, and; adjusted based upon reports in PriusChat because we need to know.
• Why should Toyota not give you an exact true number? - I wish they would but not if it jacks up my cost. For me, making such measurements are just part of who I am.

There is a USA term we use called "sand bagging" that means your product delivers more than the specification. I'm thinking of the unfortunate TDI buyers who thought the EPA numbers were wrong because they were able to beat them so easily … and recently we learned how. The TDIs were compliant in emissions and MPG testing but cheated in operation.

Bob Wilson

 

So I went to fuel up today.
My trip A (tank) reported the following:


So this meant that: 776,2 *4,8*1,062 (km *L/100km * 6,2% MFD overestimation) / 100 = 39,56 L.
I have fuelled up the following (3 clicks after first stop):


Difference - 39,34-39,56= -0,22L. I.e. practically nothing.
This means I had 45-39,34 = 5,66L still available in the tank at point of fuelling up, which at my lifetime average fuel consumption of 5.4L/100km (calculated) yields 105km worth of range (or about 65mi).

This tells me that the DTE is practically useless in EU because having 2 gal of fuel left, or 7,57L, at 5,4L/100km gives me a huge range (140km/87mi). A range of 60km is more realistic. I.e. the DTE would be more useful at 1gal (3,79L, which is a range of 70km/44mi). When it peeps (btw it does that *after* the DTE has reached zero - never really checked how much after though), it simply tells me that I have a good 140km worth of travel, i.e. enough for a few days of commute (3 to be precise) or to reach the next gas station on the highway/regional road.
The gen4, with its improved fuel consumption, will only exacerbate this DTE "issue". Essentially it will be really totally useless. Because 7,57L at 4,5L/100km (for example) yields a range of 168km/104mi, which is enough to almost drive to the next big town, or even in some cases, the next country

To avoid doing these calculations in my head (no way) or dangerously with my iPhone while driving (never ever), I made the following table in Excel (Range Calculator.xlsx) which I printed out and keep in the glove compartment whenever necessary. It is customisable, especially based on your assessment of MFD error and what remaining range you want to consider. It works for MPG and L/100km.

Yes, I assume that if Toyota say it is 45L tank, it will be 45L tank, +/- a negligible amount that will certainly not save me from trying to reach the last gas station after passing at least 3 others to save 1€ because of the price delta of 3€c...

All this exercise is more psychological than anything else. I use it to simply tell myself that if I consumed so and so many L, I will never have less than so and so many L in my tank. I.e. if I have used 40L, I have at least 4 L inside, as the tank is 45L and e.g. 1L fake value on Toyota's part, is practically impossible. So I do really have at least 4L, which at 5L/100km (a very doable fuel consumption), gives me at least 80km to drive to the next gas station. In the EU that is practically always more than enough to reach anywhere a gas station.
Pity that DTE is not customisable.

PS: I can tell you that at least in my 2002 VW Polo, when the DTE was at 0km, you really had like 50km or less to reach a gas station. That is instead too aggressive, and landed me stranded in the middle of nowhere (exactly in the middle between 2 gas stations 60km apart where I skipped the first because I thought it was too expensive... )

 

After an aborted foreign landing and diversion a couple years ago, a fellow traveler who is also an air traffic controller indicated that the rule he was under, was that aircraft were required to enough for the flight, plus one diversion, plus 30 minutes in daylight or 45 minutes at night.

Here is what I dredged from a 2010 LA Times article about American Airlines' effort to narrow its average fuel reserve load:

"American planes are landing with enough fuel to fly an additional 93 minutes on average, airline officials said. The goal is to whittle the average down over the course of a year, they said. Extra fuel is always allocated during bad weather, they said, and to account for the probability that planes will be placed in holding patterns near congested airports, among other factors.

"In an excellent-weather month, we've gotten down to around 88 [minutes]," but never below that on average, Wagner said.

American's policy to have a minimum of 65 minutes of fuel left at the end of a flight is well above the FAA requirement of a minimum of 45 minutes of fuel remaining on domestic arrivals and 30 minutes on international arrivals.

"The idea that you should never burn into your reserve fuel is a misperception," Wagner said. "It is there to be used if needed. But that said, we burn into our reserve fuel less than 0.2% of the time.""

So, AA's average remaining fuel was 88 to 93 minutes? On my last Ford, only Honda, and first Subaru, that low a margin still put them well below 'E' on their fuel gauges, in the zone that many drivers call 'running on fumes'. On my Prii, it is also below the flashing bottom pip warning and even beyond DTE = 0 miles.

FWIW, my latest Subaru gives several warnings spread over 75 to 30 minutes remaining. But my first Ford went dry with the gauge still showing positive fuel remaining. That incident is one of the reasons I actually test cars now. Some all the way to actual fuel starvation, others merely down to an identifiable 'comfort level' somewhat close to the claimed refill capacity.

 

Uh...93 minutes in an airliner is going to get you a heck of a lot farther than 93 minutes in a car - 700 miles versus 100 miles or so.

I've definitely been in situations in a car where 93 minutes of fuel remaining will no get you to the next gas station. Not the case in an airliner once you have crossed the ocean. And regarding the ocean, the ETOPS rules tend to be closer to 300 minutes, depending on plane and route.

 

How is that relevant?

As have I. See my embedded pictures in an earlier post, #25 of this thread.

But this thread is about how much fuel safety margin we have remaining once we have arrived at that next gas station.

Squirrel!

ETOPS is about engine reliability, not fuel range. Originally, it applied only to twin-engine airliners, not four-engine craft such as B747, A340, and A380. Passenger cars won't meet ETOPS rules no matter high high you keep the fuel tank filled.

 

Because 93 minutes in an airliner will essentially always get you to a gas station while 93 minutes in a car won't.

I prefer at least 4 hours in a car, but I'm good with 1 hour in an airplane.

 

Fuzzy1's picture of 150 km to "Sask crossing" gets worse. It's only about 100 mi., BUT the station there closes early and may not be open at all some days in the winter. Been there on many vehicles, worst problem was on the motorcycle. Station was closed and I had to "limp" to the next, which closed 5 min after I got to it. Trust me, 90 km/hr is -slow- on a motorcycle, especially when motorhomes were passing me like I as standing still (most motorcycles have very small fuel tanks and can go around 200 km to empty -"at speed").

To add to Bob's "reasons", we want to know if something "really bad" happens!

 

My first passage there was in winter, and it was open daily with long hours. Its website shows attended hours of 7a-11p daily, and on call the other hours. (I didn't need any, having stocked up on less expensive fuel outside the Park.) Motorhomes and all other traffic were scarce in February, so 90 km/hr was perfectly fine, with faster traffic catching up very infrequently.

Was your decision to 'limp' away a well informed choice? Or a desperate 'Hail Mary' attempt to get out, not knowing if you'd make it? Or something else?

So what would one do if arriving at such a service station to find it closed and no attendant reachable, in a vehicle with fuel gauge just having reached 'E' or giving its first low fuel warning? My different vehicles would have put different constraints on the choices:

(*) My first Subaru? No problem, huge reserve below E, it could get out in any direction, including to Jasper, with decent margin. Not even any hypermiling, just don't be wasteful.
(*) My first Ford? Forget about it, it would likely already be stalled. If not yet stalled, just park in a nearby safe spot and leave it until the station reopens. Get a room at the adjacent lodge, or sleep in the car. Don't even think about finding alternate lodging, beyond walking distance.
(*) My current Subaru? It can't get to Jasper, but will get out south to Lake Louise, or east past Cline River. If going one those directions, merely continue as planned. If Jasper is the intended destination, then (1) get a close room and wait for fuel reopening, or (2) detour east or south for fuel, or (3) find a nice place to camp overnight, within 30 miles, then come back here tomorrow.
(*) Prius? It ought to get to Jasper, but not with comfortable margin. Review same choices as current Subaru.

I know how these choices differ only because their fuel ranges have been characterized beyond the simple vague indications given by the gauges. Without that knowledge, the choice would be based on uninformed guesswork, filled with fears and wishful thinking.

Just west of Banff, the Trans-Canada Highway has the same distance without fuel between Golden and Revelstoke, though both have more services available.

You can camp out in your car overnight -- or pay tourist trap fuel prices -- while I continue on to a comfortable room as planned.

 

I was on my RZ350, and filled up at Lake Louise. Then headed to Sask. Crossing intending to continue to Rocky Mtn. House and Edmonton. I was "playing on the road", I won't discuss speed, but needless to say, my 2 stroke MC drinks mightily when going fast. I thought it would be a "good idea" to fill up at Sask. Crossing, "just in case". Of course, they were closed (under construction). So no more "playing on the highway", and had to make it to Nordegg to fill up. Bright sunny summer day, almost no traffic, the motorhomes were doing 130-150 km/hr, and I was limping along at 90, trying to make it to Nordegg. Made it with a few cups of fuel left in the tank (went on "reserve" well before Nordegg), 5 minutes before they were to close. In fact the attendant was counting the cash, and had to turn on the pump for me. That tiny station closed at 6 PM back then.

Always something to wreck my fun! Too old for that now.

The Prius just sails right through to Edmonton.

On another trip I forgot to fuel up before leaving Edmonton. Same bike. Went on "reserve" at Carrot Creek on Hwy 16, thought I would fuel up at Edson. Followed fast trucks and cars to improve mileage, and ran out of fuel at the top of the railway overpass on the east side of Edson and coasted down the overpass into the gas station, engine dead. Huh!

"The young don't always do what they should."

Now I get nervous when the Prius is only half full. Gee, I wonder why!

 

One of the things I considered when buying a new bike was always fuel tank capacity. My Hondas were so-so, but the one BMW I owned had a 4.5-5 gal tank, which was great.

Sorry, OT, now back to our regularly scheduled program.

 

And if you're stuck on a closed highway at night and it's 10 degrees outside?

 

That is why sensible drivers plan for additional safety margins in foul weather, and carry basic winter survival equipment when driving in winter. In case of a variety of problems, including mechanical breakdown, one may need to walk out for help. And may need to camp overnight in the car before setting out in daylight.

If a driver has a very large uncertainty about how much fuel range remains, or has never before used that lower portion of a fuel tank, then it would be idiotic to continue beyond a closed station in foul winter weather. Don't depend on the unverified hope that some safety margin is there because the owners manual, or some nominal capacity listed on a specification sheet, suggests that it should be there. It might be there. Or like my first car, it might be just an illusion. That could cause a stranding high up in the mountains in far worse conditions than down by that closed station. In really bad conditions, if you camp out next to the station, at least you can break in for better shelter.

In foul weather with a new-to-me vehicle, I always assume that there is no fuel below about the 1/4 mark, or the lowest gauge point that I have previously used. If I haven't personally verified it, then it isn't really there. Foul conditions are not the best time to make such adverse discoveries. When good weather and benign conditions are available, then I go out and deliberately run some low-tank tests. At least then, recovery from adverse findings won't be complicated by additional 'challenges'.

 

Theoretically, when the Satnav is in use, and if topographical data were in the maps - (i.e if it knew where the hills are and how steep they were), the car could do a much better estimate of predicted fuel use, assuming you followed the Satnav's directions.

I've thought for some time now that if the car "knew" where you were going, it could do more intelligent planning of the traction battery reserve, and use the hybrid system more efficiently. Has Toyota considered this, does anyone know?

Changing the subject, to the above quote, do I understand correctly that the Gen4 will not go over 20 mph in EV mode - does everyone agree with this?

 

I don't think that is true but I can't check for sure until I get home Saturday evening. I'm fairly certain that my ScanGauge has displayed RPM=0 and GPH=0.04 when the EV light is on and the car is moving at high speed, say in the 50-70MPH range. When I press down on the accelerator to increase the load, the EV light goes out, I can hear and feel the gas engine kick in, and RPM and GPH increase. So in this sense I think the car was in "EV mode".

The previous discussion had to do with whether the gas engine is spinning while the EV light is on and if small amounts of fuel are being injected into it.

If by "EV mode" you mean a mode like "ECO" and "POWER" you can select from the dash, the Gen 3 had an EV button but when I pressed that I often got a message saying that the car could not go into EV under present conditions. So I ended up ignoring it, leaving the Gen 3 in "ECO", and manipulating the accelerator to switch off the gas engine when I could. That has carried over to how I drive the Gen 4, so unfortunately I can't remember if it even has an "EV mode".