Efficient driving for a 1,000 mile tank

INTRODUCTION

This is a personal log describing a marathon, driving experiment . . . a place to draft a work in progress. It is not meant as instruction or brow-beating or even a 'bowling trophy'. Rather, just an experiment in progress.

Background

A Prius owner since October 2005, I've been impressed by the Japanese 1,000 mile tank club. Spread over weeks, they have incorporated efficient driving into their daily commute. It is a practical approach that has a lot to recommend.

Goal

Achieve a 1,000+ mile tank with my wife's 2010 Prius

OBJECTIVES

1. Prepare the car
2. Instrumentation
3. Route
4. Weather
5. Driver training

Prepare The Car

By happy accident, one of the original tires suffered a non-repairable puncture. This led to replacing them with Sumitomo T4, 195/65R15. These are low-rolling resistance tires with a maximum sidewall pressure of 51 psi. Bought new in the first year, they now have 30,000 miles and have not been rotated (using European practice.)

Using GPS and mile markers, these tires have less than a 1% error in the indicated speed and odometer values. The original tires that were off by about 3%. Using a 12V powered air pump, I made sure the tires are fully inflated plus 5-10% to handle any leakage.

The transaxle oil has been changed twice, at 5,000 and 15,000 miles. The first change 'flushed' the early manufacturing debris and testing projected a 15,000 mile service life. The second change did not have enough in the sample bottle for testing (thanks Toyota Service Center). So the third change is coming due.

Past transaxle oil tests suggests the oil 'shears down' the viscosity, in effect making it an even lower viscosity lubricant except for any wear debris. At 40,000 miles, nearly 25,000 service miles, it is due but also in a 'good state.'

The engine oil was changed about 3,000 miles again. Like the transmission oil, as it wears, its viscosity goes down reducing internal friction.

Have some quality music, in my case, iPad with all my tunes and an iPhone with Internet radio. Both have and play podcasts but music runs long enough to make marathon driving tolerable.

Instrumentation

The car has a dashcam that has taken the GPS power. However, the dashcam has a recording GPS that allows tracking the route taken. It also exports the data in "kml" format that Google Earth uses to map a route.

A ScanGaugeII is configured to display:

• engine RPM - key to knowing if the engine if ON or OFF
• gallons per hour - the rate of fuel burn, ~0.60 GPH is normal engine warm-up and an optimum fuel burn for level roads. During a climb, a 2x fuel burn, ~1.20 GPH, is a very efficient rate to convert engine power to kinetic or potential energy
• engine coolant temperature - lets us know if the engine is in a fuel-efficient range and capable of normal hybrid operation (i.e., engine off when not needed).
• instant MPG - unlike the vehicle MPG, the four digits provides faster feedback and covers the area above 99.9, the maximum of the stock indicator

Route Planning

It is important to have a commuting and urban routes that support driving at 25 mph without leading to 'road rage.'

• multiple lanes in the same direction - this allows changing lanes so other traffic can pass on either the right or left
• unpopular routes - most rush hour traffic is going to funnel through very heavily used routes. Instead, choose routes through neighborhoods or from neighborhoods which are often posted at lower speeds, 25 mph is perfect, 30 mph is OK, faster routes MUST be multi-lane.

As flat as possible with preferences for:

• long, gentle descents - with a minimum of stops or known light timing for "just in time green"
• modest, upgrades - especially to a traffic light or stop sign so gravity slows the car without having to use brakes
• right-turn Yields - your friend as it preserves momentum but often 'right on red' depending upon local police can also be treated as a "right-turn Yield"
• avoid police patrolled areas - a slow moving car often indicates a potential intoxicated or under the influence driver. No reason to 'wave a red flag' in the face of public safety officers.

A commuting route that includes a trick so it can take more than 20 minutes to reach work. The transaxle is the slowest to warm-up, taking nearly 30 minutes. As it warms, the transaxle drag significantly goes down. So scout out a 'lap' path near the commuting route that can add 10-20 minutes at 25 mph without stops. For example, my rapid commute to work takes 15 minutes at just under 50 MPG. Ordinary commuting takes 20 minutes at nearly 55 MPG. Marathon commuting takes 40-60 minutes at over 99.9 MPG.

Neighborhood route posted at 25 mph for at least the first mile. Do not worry about stop signs as much as being able to use "N" while rolling to let the engine warm while rolling. However, 40-45 mph, commuting routes can also be used provided it is not rush hour.

Weather

Dry, the temperature ranges:

• 90F or higher - brutal to driver unless dry, sun and heat makes it difficult to drive efficiently
• 80-85F - ideal, perfect temperature if acclimated
• 70-79F - good enough
• 60-69F - still good but not perfect, radiator inlet block helps
• 55-59F - marginal, must be especially skilled, and radiator inlet block
• 50F or lower - no, the air is too dense, the tires too stiff, and heat loss leads to cooler, more viscous lubricants

Driver Training

Speed control is handled 95% of the time using cruise control. This frees the driver to pay attention to traffic and get their head 'out of the cockpit.' It also allows much better control of the driving conditions so the results are reproducible:

• 26-27 mph normal speed setting - the cruise control will cut off at 23 mph so 3-4 mph headroom is needed to prevent it cutting off on upgrades
• use "UP/DOWN" trim - if the car is too fast or slow, use a brief, 1-2 second tap to increase or decrease the speed by 1 mph
• use "UP" to resume last set speed - once the cruise control is optimally set, use "UP" to resume speed

Use "N" when:

• descending gentle grade AND the power indicator shows energy flow to the battery AND engine is off - this converts potential energy into distance traveled
• approaching a known or likely stop AND engine is off - at the earliest opportunity

Use 1-2 second, "N", "D", and "N" when:

• approaching a stop and engine is running - this will get out of cruise control, the "N"; the "D" for 1-2 seconds, lets the engine auto-stop, and; "N" resumes a ballistics (aka., momentum) roll to the stop

Manual accelerate at ~1.20 GPH or ~3/4 of the power bar. This is a very efficient engine operating range. Otherwise accelerate with traffic when in a clump of cars.

When in a clump of cars, look for a parking lot to pull-in. Return to the road when the traffic clears.

Monitor the rear and side-view mirrors. If traffic is overtaking, you should change lanes to let them pass in whatever lane they are in.

Use "emergency flash" to let following traffic that you are driving slow so they can pass without a problem.

MORE COMING

There are more technical details to follow and I'll add them to this thread. Feel free to ask questions along the way, I'm a great believer in dialog. I'm already planning to post traffic management.

Progress


I'll clean this up later tonight but as of July 3, 2013, ~11:30 AM, after arriving at work:

• 12.2 gallons usable fuel @90MPG -> ~1,098 miles >> than potential 1% error

Bob Wilson

​

 

Something I learned about several years ago from one of the EU members, don't rotate the tires front-and-back every year or so.

Initially the heavily loaded, front pair wear first. When that happens, the rear tires are moved forward and a new pair are bought for the rear. So instead of buying four tires at one time, only a rear pair are bought, cutting the tire replacement cost in half. However, I'm interested in the tire wear patterns.

Abnormal, tire wear patterns stay with the wheel. Rotating tires on a chronological or mileage schedule 'averages' this wear which makes identification of a problem wheel more difficult. But I want the wear patterns so if there is a toe or camber issue, I can confirm it with a wheel alignment and add shims as needed.

I first learned about this practice from an EU member and matches an experiment I'm performing with our 2003 Prius. The front tires are ~6% larger diameter than the stock tires, notice the difference in the side walls in the photo:

The tires are described here and consist of:

• Front - Sumitomo T4 195/70R14, 849 rev/mile vs 902 rev/mile stock, improved straight-line stability, probably due to gyroscopic forces. The 6% over-drive effect means the transaxle turns slower, lower stirring losses at equivalent speeds and distances. Also using Type WS transaxle oil instead of Type T-IV, the original.
• Rear - Sumitomo T4 175/70R14, 886 rev/mile vs 902 rev/mile stock

I am thinking about trying the same experiment with our 2010 Prius, using a larger diameter front tire to achieve an over-drive effect. Sure it will reduce the maximum acceleration rate . . . as if a Prius is a sprint car. I'm looking at it for fuel efficiency, an "experiment." I can afford to risk ~$100 on an experimental pair of tires. If it works, life is good. If it fails, I have no problem with a more conservative replacement pair.

Back to the 1,000 tank, I had been using one trip meter to measure some local mileage events. Sad to say, I accidently reset the tank trip meter and will have to restart. <darn> So I'm changing the trip meter protocol, leaving both A and B alone without any resets. I still need the power bar for driving clues but there is no reason to risk a reset on either until flash.

Bob Wilson

 

A little disgusted with myself, I 'fat finger' wiped out my earlier trip meter record and had to start again yesterday afternoon, July 7. So this is my progress:

July 8, 2013:

 

July 9, 2013


The adventure continues:

• Leaving for work at 5:00 AM - makes it a lot easier to travel at cruise control speed of 25-26 mph. Less traffic and better behaved lights. Later, leaving before the rush hour again makes it easy to use cruise control at slow speeds.
• Are pickup drivers "Type A"? - looking at their driving behavior, most are in a hurry to get somewhere and not really too concerned about anything but being first. Sure some have small businesses or are in service work and I understand that but the utility vans don't seem to be quite so 'frantic' in traffic.
• Driving via the mirrors - taking routes that have two lanes has worked well but I'm finding my lane changes are more to let over taking traffic not have to change lanes. Sometimes, I have to hit my emergency flashers and I'm also using other 'outs.' For example, with a big slug of traffic behind, I'll duck into a parking lot that has two exits to let them pass. Also, I've used a mock left-turn lane to wait until following traffic clears and then pulled back in.

Bob Wilson

 

Marathon driving is not about getting somewhere as much as going as far as possible on the minimum amount of gas. But I can't stand such waste so I've incorporated it into my commuting and ordinary activities. I'll post more details but here is the executive summary:

• ~15 minutes, 13 miles - gotta get into work NOW, ~48 MPG
• ~20 minutes, 10 miles - normal going into work, ~52 MPG
• ~60 minutes, 20+ miles - marathon commuting

It takes a solid 20 minutes for the transmission to warm-up so the car can achieve maximum efficiency. So as soon as I arrive at work, it is finally ready to turn in some impressive numbers . . . and then I park it and it cools off. In marathon driving, I pass by work and proceed to a loop that I had already mapped many years ago when doing some MPG vs mph testing:



I'll post more details later but I travel clockwise around the loop. This minimizes having to change speed due to traffic coming into and leaving the loop.

I have uploaded the Google Earth, kml file with a ".txt" suffix to support uploading. Once you download it, take off the ".txt" and drop the "1000mi.kml" onto Google Earth.

Bob Wilson

ps. A Southern apology for NASCAR which turns in the opposite direction.

 

I think that was only true for the North American, NHW11 and NHW20s, the 1.5L versions, that have a fuel bladder in the tank. I've measured 12.1-12.2 gallons in the tank of our ZVW30, 1.8L Prius.

Bob Wilson

 

Here is the dashcam trace and vertical profile:

In this example, I had gone into work late. When I left the office, I drove to the loop mentioned earlier. From 5-20 minutes, I was doin' laps that give 99.9 MPG. The driving protocol:

• lowest point - set cruise control to 26-27 mph
• climb to highest point - feet flat on the floor, the car efficiently gains altitude
• crest of hill - as car crests the top, the engine unloads and rpm goes to zero, shift into "N"
• roll down the hill in "N" - the accelerates to 30-31 mph before reaching the bottom and slowing down
• shift into "D" and "resume" on cruise control - the loop repeats

Other than initially reaching 26-27 mph to set the cruise control, the foot never touches the accelerator. All power operations are by hand:

• shift into "N" when engine stops cresting the hill. I use a Scanguage II for precision but it can be done with the stock instrumentation.
• shift into "D" and hit cruise control "resume" when speed decays to ~26 mph.

This is the same protocol used away from the loop, when driving from point A to point B. The foot seldom touches the accelerator or brake except as needed for safe driving.

Bob Wilson

 

Hi Walter,

Thanks for the comments but there are some points unique to this experiment:

• 12.1 gallons measured - from six, out-of-gas, experiments, using my normal, second-click, stop. I did not try to 'burp' the tank nor did I switch to pure gas, still using E10.
• HSI display is accurate - thanks to the early replacement with Sumitomo T4s, our HSI is less than 1% off.

Since I'm using cruise control to manage speed and power, we have to live with the 23 mph minimum speed and at least a 2 mph 'head room' to handle speed differences of upgrades. Although 25 mph mostly works, using 26-27 mph provides enough to handle even the shortest, upgrades at bridges over creeks and the occasional slope up to a light in Huntsville, AL.

• Sumitomo T4 - rated 51 psi max sidewall and inflated using 12V pump at start of experiment
• "N" is my standard operating procedure - used historically only during the catalytic converter warm-up, for this experiment, using the Scangauge II "gallons per hour" to select the best mode along with monitoring the engine rpm
• route planning - fortunately, I don't have to use gravel roads and Huntsville AL is historical 'river bottom'

Not this one because it is insignificant and handled by traction battery energy during the initial catalytic converter warm-up. However, I do 'back-in' park at the end of each trip so the 'cold soak' and catalytic converter warm-up preserves momentum.

NO and I have the dashcam to prove it. This is efficient driving and that means respecting stop signs and lights. However, I do select routes that have more "Yield" and "right-turn on red" options. But it is much easier by commuting outside of the Huntsville AL 'rush minute.' Even so, there are times:

• pull into center left-turn lane at uncontrolled intersection to a full stop - lets following traffic pass
• put into a parking lot with two exits - lets following traffic pass
• use "emergency flashers" to indicate to following traffic 'I am slow'
• change lanes so over coming traffic does not have to change lanes - driving with all mirrors
• use 'alley' routes for low traffic, low speed, direct paths - WATCH FOR STAFF, but avoids being in traffic

I also have my 'trick', the 99.9 MPG loop that I can 'do laps.' About two miles away from work, each loop adds a mile but at 99.9 MPG. So in 10 laps, I've averaged in 99.9 MPG with whatever the 10 miles to work took. I'm finding keeping the indicated 90 MPG fairly easy. Best of all, 'small bites' on each trip so I'm not in 'butt busting mode.' After all, I have obligations to my wife and our two dogs. <grins>

Marathon driving is not for everyone but our Japanese friends, especially Ken@Japan, gave us a power clue with the 1,000 mile tank club. Having spent a year on Okinawa Japan, I remember the roads and speeds. But my own warm-up testing verified that complete warm-up takes 20 minutes, my ordinary morning commute time. So adding 40 minutes of 99.9 MPG laps with careful attention to warm-up, it is possible, not trivial but possible.

Bob Wilson

 

July 10, 2013


Roughly 3 gallons of the 12.1 gallon tank burned, about 25%. So we're looking at:

• 1,091.4 mi = 90.2 MPG * 12.1 G :: expected ending miles
• ~49.6 hrs seat-time = 1091.4 / 22 mph
• ~16 days calender time = 4 days * 4 quarters :: expected time to complete (not counting weather delays)

This is incorporated into my daily commute and other chores around town. I don't have to take vacation or time off and my wife and dogs still get to see me.

Bob Wilson

 

I'll see how it works out but from memory, I think it is about three bars down out of ten.

As a general rule, I like to use trip meter 'miles' and 'MPG' (my tires have less than a 1% error) to calculate the amount of gas remaining. This is important if you plan to do fuel-exhaustion testing. You kinda need to know when to put the spare can in the back and start monitoring the 'energy flow' display.

Bob Wilson

 

July 11, 2013


Thanks to a thunder storm and traffic, I've lost MPG. So I'll do more 'laps' to bring it back to 90 MPG for tomorrow.

So after a long meeting, I went to the lap to 'calm down' . . . one hour in the loop:

	MPG	miles	fuel	time
1	89.3	364.1	4.08	2:29 PM
2	88.5	331.3	3.74	8:45 AM
3	loop - 96.5	32.8	0.34	1 hour in loop: 24 min. transit

Source: Prius display

The loop takes 12 min. to reach and leave in mixed traffic. As mentioned before, it takes an honest 20 minutes to completely warm the transmission and I haven't done today's loop. This was just make-up for yesterday's drive when time ran out as the thunder storm approached.

The reason this is important is this segmented marathon drive is a process, a procedure, a way to achieve high MPG. It is not a practical driving style . . . boring beyond belief without good music. Even so, 1-2 hour segments are a whole lot 'less bad' than trying some soul crushing, butt-busting stunt . . . and this is, like our Japanese friends, a solo effort . . . although the knowledge came from brilliant Prius pioneers.

FYI, I just got home:


Bob Wilson