Hi folks,

Alansanmateo and I have been discussing 'route planning,' an approach to efficient driving that seeks the best way to higher mileage. He agreed to let me share the analysis so folks can see the method and methodology.

BACKGROUND

Our hybrids don't preform very well during the warm-up. In the case of the Prius, the mixture is very rich to heat up the catalytic converter and in some models, they require the coolant to be at 70C before fuel-efficient, hybrid operation begins. During this 5-10 minute interval, our hybrids run like ordinary gas-only cars with poor performance. But if we can minimize the engine load during warm-up, we can achieve higher overall performance.

Initial Warm-Up

In the following chart, commute end points are at the lower left and upper right and that is the direction of travel we'll ldiscuss. You'll notice there are two maps, a Google map used to identify streets and a topographical map used to identify flat versus sloped areas:


To map the different routes, I used the Web to get a Google map and a topographical map. I loaded these into a PowerPoint presentation and scaled them to approximately the same dimensions. Then I put them in the "Slide Master" layer. This allows use of the drawing tools without accidently grabbing the maps. BTW, the Google map showed the optimum route to use the freeway around the neighborhoods, ~4.8 miles.

We want to minimize the speed and load on the hybrid in the first 5 minutes. So we select a route that would have slow, neighborhood speeds, 25 mph or less, and relatively flat. So early use of the freeway is avoided because of too much energy needed while the hybrid is in an inefficient, warm-up mode.

Instead, we cut into another neighborhood on the other side of the freeway. In the other neighborhood, we try to avoid descending into a "hollow" because of the heavy load needed to climb out of it. Running parallel to the gradient lines minimizes the up/down loads on the car.

Cruise

Having warmed up, there are two choices, taking the freeway or cut through another neighborhood. It depends upon the type of hybrid. In the case of the Prius, there is a critical speed, 42 mph, that separates two control laws. Below 42 mph, the Prius (and Toyota systems), will cycle between ICE power and electric power. This gives the best possible mileage in a warmed-up Prius. Other hybrids like the IMA Hondas and belt assisted units don't have a documented critical speed and can take whatever route, including the freeway, that gets them there easiest. They still do better at slow speeds but the Prius has a lot more regenerative braking capacity and excels at stop-and-go, slow speeds.

In our example, we turn right onto a cross-town street and then take a left to once again cut through a neighborhood. This allows slower speeds and generally less traffic although there may be speed bumps and other things to avoid. The map doesn't show these features.

Cool-down

The last mile or so of a trip is when the electric mode should be exploited as much as possible. Lower speeds allow the electric systems to provide more power and the gas engine to remain mostly off. There is no advantage to trying to 'save charge' in our NiMH batteries because of the "self-discharge."

CONCLUSIONS

The real proof is in testing alternate routes but 'cold' testing is generally limited. Just because you spend a Saturday testing the different routes in an already warmed up hybrid, remember that the real "warm-up" is different. Also, the maps don't show speed-bumps or other traffic hazards and sometimes, the maps are wrong. So the different routes have to be tested and the best morning route may not be the best evening route.

Our hybrids excel at slow speeds and often the shortest route is also a slower route. In this case, Google recommended taking the freeway, a high-speed route, around the neighborhoods. But the direct path not only increases hybrid MPG but it also reduces the distance and usually eliminates any time penalty.

Bob Wilson

 

it will be interesting to try this and see if our mpg improves

then we can work on the route back home!! (UP the hill)

 

I'd probably take 31st back to Hillsdale and hop the freeway to De Anza. The reason is 31st starts fairly level, good for warm-up. When you reach Hillsdale, you're pretty well into cruise mode. Then take the freeway home for a quick hop and cruise-down to the driveway. Looks fairly straight forward . . . depending upon what 31st looks like.

GOOD LUCK!
Bob Wilson

 

Bob, just curious, will you get a chance to do any testing on how engine load effects warm-up time? Engines working at higher load warm up more quickly than engines that are essentially idling; I wonder if there's any balance point where reducing load during warmup just extends the amount of time the engine isn't warm beyond the benefit of reduced fuel consumption at lighter load.

(did that question make any sense?)

cheers --
doug

 

Hi Doug,
Every morning, I've found that no form of loading increases the ICE warm-up fast enough to save a significant amount of fuel. I'm monitoring the air-flow and summed up by the MPG after 3 miles. It is as if there is a maximum rate of heat flow from the cylinders through the walls into the rest of the engine. As the power levels go up, the heat not absorbed by the walls and exhaust valves just pass out the exhaust pipe that much faster.

What I typically see is in "D", the early air-flow will be typically 7+ g/sec. But if I put the car in "N", the air-flow will drop to 4 g/sec. Yet the total warm-up time doesn't change by a significant amount. But you have suggested an interesting experiment.

What I can do is two warm-up tests in the driveway. I'll block the wheels and use the parking brake. In one test, I'll put the car in "N" and monitor the rate of warm-up. In the next test, I'll leave the car in "D" and monitor the same warm-up.

I have some temperature studies from last winter that suggests the warm-up profile is somewhat load independent:


The rate of warm-up on three separate trips appears to be pretty close to identical.

BTW, I have some temperature studies already:
http://hiwaay.net/~bzwilson/prius/pri_warmup.html
http://hiwaay.net/~bzwilson/prius/pri_temp.html

None of these tests has been as rigorous as the proposed driveway test using "N" and "D" (or "P"). It will fill in a missing nitch in the data.

BTW, my earliest clue came from an informal observation of a low mileage Prius owners. He lives at the bottom of a hill and every morning has to climb the hill, right off, cold and at a pretty good clip.

Bob Wilson

 

I live about a block away from the expressway on ramp that I use to get to work. So I go about 5 miles imediately at 55mph then get off that expressway and take surface streets for about 5 miles @35mph at which point I enter another expressway at 55mph for the last 10 miles of my commute.

My question is, would I get better overall FE if I skip the first 5 miles of expressway and take surface streets @ 35-40mph for the first 10 miles?

On the return trip I could do similar, take surface streets at 35mph for a few miles before entering the expressway.

Or should I just get a block heater and have the car warmed up already for the morning commute ?

 

From an MPG number stand point, yes, but it could result in a lot longer commute and that would be bad. You need to 'run the test' and by all means, don't let a frustrating, fuel saving route lead to a bad plan.

There is solid evidence that the block heater is effective. I'm still getting some numbers but our Japanese friends and our friend who recently did 1,000 miles on one tank in Virgina Beach also report good results.

I just started since Sunday July 22, running with both a block heater and transaxle pan heater (I'm a belt and suspenders engineer.) I'm seeing results that range from good, 52->55 MPG, to outstanding, 52->61 MPG. I suspect there are somethings to learn about how to optimize the warm-up when you have a block heater. This morning's result, 52->61 MPG, involved an early, mid-speed of 45 mph, higher than my normal 38 mph. Based on my limited testing this week and reports of others, there is real merit in getting a block heater and using it year round.

Bob Wilson

 

This morning, for the second time, I came in at 60+ MPG in my normal, 52-53 MPG morning commute. The biggest changes:

• using block and transaxle heaters
• using 45 mph earlier in the commute instead of <25 mph for first mile

I recorded the data and will analyze it over the weekend. But this preliminary result strongly suggests that if you have a coolant thermos like the 2004-current Prius or a block heater in your NHW11, you can and may need to use a higher speed earlier in the commute. Again, I need to see what the data shows but I suspect the higher, early commute speed leads to more efficient battery charging and travel.

I have one other test in progress, using Type WS instead of Type T-IV in my NHW11 transaxle. The NHW20 uses Type WS and it has a lower viscosity that Toyota claims in one paper reduces transaxle losses. This should also help but given the use of a transaxle oil pan heater, it is impossible to quantify this, yet. Right now, I'm watching the wear pattern to make sure it doesn't 'eat' the transaxle.'

BTW, Alansanmateo and family are performing the test and we've been discussing some of the preliminary data. I'm hoping he'll add them to this thread.

Bob Wilson

 

It's too bad but the way the expressway is designed i think some of the on ramps aren't too safe. If the on ramp goes too short and there isn't sufficient room to move over (and I have felt crowded out on it several times), I'm just not doing it unless I have to. Too bad since I have a Corolla and it would be to my advantage (no doubt) to do as much highway driving as possible. Otoh, I could rethink the other way as i think the off ramps are ok.

--des

 

Bob, can you shed some light into where you got the terra maps from? I have been looking for something like this in my area to help plan my trips.