P&G 45-60 mph

Recently, someone claimed higher mileage using pulse and glide with a speed range of 45 to 60 mph than cruise control managed 55 mph. This seemed counter intuitive since as Ken@Japan pointed out, the threshold between engine off, hybrid operation is 46 mph. P&G assumes the engine goes off, no fuel burn, so a range of speeds 45-60 mph would ordinarily result in little or no performance improvement. Regardless, it is a testable hypothesis.

Early this morning, I drove to a new test route, highway 72/20, and the GPS reports a 9.7 mile segment between traffic lights and with a speed limit of 65 mph between these two points:

34.619052 N 87.051419 W
34.649002 N 87.213184 W
altitude range: 174-194 m
​

The temperature ranged 64-68F. This was just a survey to find a stretch to do some informal, field testing:

• 45-60 mph - average speed 53 mph, 59.5 MPG
• 60-60 mph - average speed 60 mph, 58.7 MPG

So obviously, I need to go back and retest with the temperatures above 80F and use 53 mph for the cruise control, equivalent speed of the 45-60 mph P&G. Also, I need to do runs in both directions to normalize any wind effects. In order to generate reproducible results, I use cruise control to resume speed, the pulse, and "N" to coast.

But this begs the question of the poll.

Bob Wilson

 

In fact the SAE paper and I both agree on this principle ... it needs to be done away from traffic. My early morning tests were between 4:30AM and 5:30AM and the P&G testing abandoned as soon as early morning commuting traffic showed up ... not even rush hour but the leading edge.

Bob Wilson

 

I'll be shifting into "N". Can you define "Stealth" say using the power display?

In the initial test, I'll use "N" but if there is an agreed upon definition of "stealth," I'm willing to give it a try once I've also measured the equivalent, steady speed.

I agree on the first but haven't measured the second.

Bob Wilson

 

So it needs hills?

The reason I ask is the route is 9.7 mi. (15.52 km) with a maximum altitude change of 20 m.

These are GPS reported values so realize each minor altitude unit, 1 meter, is 1/1000 of the horizontal units, 1 km. Drawn to true scale, this looks about as flat as the Tennessee River. I'm pretty sure only the very end is steep enough that the Prius would roll on its own.

So what is the velocity profile or model for Pulse and Stealth on river bottom land?

Bob Wilson

 

I'm with Ken on Pulse and Glide and it is fairly easy to test on this flat route using cruise control acceleration and "N" glide. Once that is 'off the table,' there may be merit in looking again at "Pulse and Stealth."

Speculation but if we can get a good BSFC chart and map the transition between normal and energy re-circulation (aka. 'heretical mode'), it may be there is an operational line, actually a type of hysteresis loop that might be exploited:

• acceleration - high BSFC, normal mode, pumping up the kinetic energy
• stealth - highest BSFC, energy re-circulation, extending the duration by making up any energy deficit from a fraction of the kinetic energy

My NHW11 has a Graham scanner and I have historical data. It may be as simple as mapping out BSFC data in normal and energy re-circulate mode and seeing if a closed, operational curve might provide a high efficiency, high speed advantage. For example:

Acceleration along the upper blue dots followed by extended operation in the red dot area at peak BSFC. "Steath" might be as simple as having an energy re-circulate indicator and driving a profile based upon it and ICE rpm. But we don't have enough information about the ZVW30 to map out these states and I haven't seen energy re-circulate in the SAE papers, yet.

What I'll do is run the classical Pulse and Glide experiment Saturday morning with my wife's ZVW30. I'll try to add a battery current reading and record the data with a digital recorder including:

• speed (mph)
• instanious MPG
• ECO power bar
• traction battery current

This may give us insight to refine Ken@Japan's power bar map ... a rough approximation of power bar to wheel efficiency.

Bob Wilson

 

Preliminary Results (more coming)

Protocol

• Pulse - at 45 mph (72 kph), the gear is shifted into "D" and cruise control "resume" activated to accelerate to 60 mph (96 kph)
• Glide - upon reaching 60 mph (96 kph) the car is shifted into "N"
• No air conditioner but windows on one side rolled down
• Running start into each measured segment

Route Alabama 72/20 - 14 miles (22.4 km)

• Eastern end - mile marker 63 just west of Rocket Road
• Western end - mile marker 49 just east of County Road 150

	Column 1	Column 2	Column 3	Column 4
0	End	Latitude	Longitude	Altitude
1	East	34 37' 24.48" N	87 04' 53.74" W	195 m
2	West	34 41' 03.33" N	87 18' 20.47" W	196 m

​

This divided, four lane highway is posted 65 mph (104 kph.) Whether running pulse and glide 45-60 mph (72-96 kph) or the equivalent 53 mph (85 kph), we are always slower and in the right hand lane. With one exception, traffic flowed smoothly and passed without causing any backup or consternation. That one exception was the first P&G run that started late due to setup problems and had to end early due to unexpected, local congestion.

Results

• Temperature: 90F (32 C)
• Wind: 10-12 mph (16-19 kph,) winds from WSW or W

	Column 1	Column 2	Column 3	Column 4	Column 5	Column 6	Column 7
0	model	direction	method	MPG	miles	time	mph
1	NHW11	East	cruise	62.0	14	15:35.7	53.86
2	NHW11	East	P & G	60.5	12	13:32.9 *	53.14
3	NHW11	West	cruise	58.7	14	16:07.7	52.09
4	NHW11	West	P & G	55.0	14	15:42.3	53.49

​

I'll have to borrow my wife's car for the ZVW30 data. This may have to wait until Sunday.

(*) short segment, due to traffic, driving and recording screw-ups, this segment started late and came in 13:32.9 sec from mile 50 to 62.

Notes

There are other methods for handling what has been called pulse and glide or pulse and stealth or terrain driving. Understand that I have no problem with such claims except I don't have a way to replicate them. Without a quantified or mechanical or automated way to replicate these alternate approaches, test and validation remains a problem ... good luck! But the ZVW30 ECO power bar may provide a measurable power scale.

Ken@Japan has posted an ECO power bar with a scale. His first posting looked to be a 4 part scale although I think a 5 part scale might work even better. Regardless of the divisions, a scale on the ECO power bar would provide a way to define reproducible power levels. It would then be a trivial task to measure the acceleration and fuel burn of the different settings, a relative BSFC measurement. This would provide a quantifiable alternative to using cruise control for the 'pulse.' But let's address that in another thread.

Bob Wilson