P&G or steady state driving?

I hope this doesn't come out wrong, but really? 21mph? Is that the posted speed limit?

 

Hi VaBeach...,

I have only a few places where traffic will tolerate 21 mph steady state. And one way on my commute I am on the highway at 53 mph, and the other way pulse-and-glide, with lots of lights. In this weather (80 to 90 F), I am in the high 60's consistantly for the whole tank. I had another 70.1 mpg tank recently.

On a pulse and glide portion I was up just above 70 mpg on the tank, after just starting the tank, and at the end of each glide the result was a few tenths higher. Indicating the actual P&G mileage was well above 70 mpg. And this was only a 40/35 mph P&G. In another instance, I had a really poor tank going, and in one run pulled it up significantly. When I got to my destination I took the time to figure out the mileage for that pulse and glide and it was 77 mpg. The main reason was allot of lucky light timing. That resulted in the trip actually being 10 minutes faster than normal, or 27.6 mph average , instead of 23.4 mph average.

 

Hey, stranger, where ya been?? Good to hear from you again.

I don't do much steady state driving these days, though that's in large part due to road and traffic conditions. I'm frequently accelerating and decelerating anyway for red lights, stop signs, and traffic, so I'm P&Ging simply to do it all in the most efficient way possible. Where there is an opportunity to choose between P&G and steady-state, I choose the former because of the abundance of anecdotal evidence pointing to it as the better method. Two examples, the first from last summer and the second from this summer:

http://priuschat.com/forums/gen-ii-...-trips-oh-how-i-love-summer-fuel-mileage.html

http://priuschat.com/forums/gen-ii-prius-fuel-economy/63215-97-mpg-commute-aiming-100-a.html

I'm currently pushing for my first 80 MPG tank, though it might be a struggle with some of my non-commute trips generally in the upper 60s or low 70s. It's currently at about 79 MPG, 300+ miles into it, 1 pip gone.

Few have done controlled tests comparing the two methods. Bob Wilson did so in his Gen I and got comparable results with the two methods, IIRC, using cruise control both as his steady-state method and his method of pulsing. We can't necessarily extrapolate those results to the Gen II. Besides the drive system differences, CC used as an acceleration tool in the Gen II easily pushes ICE RPM into what I consider inefficiently high ranges. I've wanted for some time to do a test, but for a variety of reasons (mostly related to life) I've not done so.

 

Things are well. Not married yet; we're in no hurry.

As for the Gen III's fuel economy capabilities, ordinary drivers seem to be getting the expected 10% improvement over the Gen II. For those that push the envelope, the envelope doesn't seem to have moved much, if any. Wayne Gerdes' parents bought one and he has had some opportunity to drive it. His comment in post #39 of this CleanMPG thread:

"... those that are pushing high FE with their –II’s would be disappointed in the –III imho."
​

Kinda makes me glad I didn't pull the trigger when I had the chance to a few months ago.

 

Hi Vabeach...,

I am running 46/44 PSI, so not much tire pumping. Everything else you mention but route selection I have done without success. A long time ago (like 30 years) I learned how to anticipate traffic and time lights, not for hypermiling purposes, but to get around stoping and all the other drivers at lights. Its the same skill set, just applied differently.

What made the difference was the combination of route selection, and the MYCANSCAN teaching me SHM, and how to keep the engine in its efficient zone during accelleration. Knowing what route to pick was actually a result of watching the temps and trip time on the MYCANSCAN. The temp gauges showed that I was warming the car up over and over on some routes. Rather than having it warm all the way up and stay there, even if that was a slow warmup. And the trip timer pointed out how one route while slow speed limit route, got me to the same place actually faster than the other routes. So, in actuality, the main improvement can all be laid down to MYCANSCAN.

 

Chris,

Don't forget that all of Bob Wilsons steady state in the Gen III are done with only the in car displays and no additional instruments. His Classic runs had the Mini-Scanner to instrument the tests. Take his Gen III results with a grain of salt. My own runs show P&G will be better than steady in most cases in a Gen III. Exception being where you live, nice and flat.

Wayne

 

Hi Va...,

Ken has said there is a guy in Japan that does his pulses at 2300 rpm, but is only going like 25 mph max, and gets 124 mpg consistently over his "practice" route.

The problem I would have here is that its not 95 F and sunny, so driving along at a constant speed of 25 mph, my engine would never get hot enough for the hybrid system to get to stage 4 for the entire length of my trip.

I had a bit of that problem today. It was not sunny, so the road was not hot (this was before the rain) and the car would not auto stop the engine, until I was 1/2 way home. Based on this I have a new hypothesis - the radiated heat from the roadway is effecting some temperature sensor - most likely the Catalytic Converter sensors - and keeping the car from going into full hybrid mode for a long time. Even so, I was able to start the day at 68.7 mpg, and end at 68.7 mpg. No 70's for this tank.

 

Hey Wayne,

My wife offered to let me use her ZVW30 for my daily commute so I'll have a chance to check the vehicle calibration. It will take about 30 days and then we'll know.

My methodology will be to fill the car twice a week, Friday evenings and Sunday evenings. This will let me separate city driving from weekend and see if the calibration is a function of factors other than the vehicle.

I'm always interested in reading direct comparisons of P&G to the equivalent steady-speed. If you have some data, trot it out with the methodology used and maybe well all get a clue. That is why I share my results and enjoyed reading and commenting on the SAE paper, 2009-01-1332. More the point, if you have measured by other means the steady state mph vs MPH, trot it out.

The best criticism of facts and data is a better set of facts and data ... that is what science and engineering is about. Do the experiment and report the results and leave the unhelpful comments like "a grain of salt" somewhere else.

Bob Wilson

 

It costs $15 but you might enjoy reading SAE 2009-01-1322, "Vehicle Inertia Impact on Fuel Consumption of Conventional and Hybrid Electric Vehicles Using Acceleration and Coast Driving Strategy," Lee, Nelson and Loshse-Busch. In one protocol, they claimed a 24-90% performance improvement. However, their testing was based upon using a dynamometer with calculated aerodynamic drag. But they also included this:

"There are some known safety issues using the PnG driving strategy on the road; traffic considerations, unstable braking performance, and unstable power steering system if the engine is off, so this driving technique should only be used on a closed course under controlled conditions." pp. 2
​

Their test article was an NHW20 and I only have an NHW11. Still, it is a good read and there is one interesting table in their paper:

Table 4. Summary of Results of 2004 Toyota Prius

	Column 1	Column 2	Column 3	Column 4
0	Protocol	PnG MPG	Cruising MPG
1	30-40 mph	10 seconds of acceleration (NHW20)	151.5	81.0 @35
2	30-40 mph	20 seconds of acceleration (NHW20)	99.9	80.9 @35
3	. . . krousdb with an NHW20 . . .
4	30-40 mph	'n' seconds of acceleration (NHW20)	106	93-94 @35
5	. . . my data with an NHW11 . . .
6	25-43 mph	15 seconds of acceleration** (NHW11)	87.8	79.0 @33

pp 6.

** Using cruise control resume function.

I noticed their data showed a decrease in SOC during their dynamometer-based test. Mine consisted of multiple runs in the field, a lightly traveled road, and mixing five runs along the track and each PnG run included six cycles per run or 30 PnG cycles in total.

It would be fascinating if someone with an NHW20 could replicate their protocol and report from the field the same data they reported on the dynamometer. I have my wife's ZVW30 so I can replicate the test for the 30-40 mph range although I'm also running a series of calibration runs that will take about 30 days to complete.

Bob Wilson

 

Before the marathon, in May 2005, I drove from Pittsburgh, PA to Huntsville, AL. It took 2 days via backroads and the whole time I had cruise control set at 35mph. The trip lasted about 950 miles and returned 85.7 MPG. This was before I discovered pulse and glide.

Later that spring I was doing more experimentation in the Myrtle Beach SC area and repeatedly recorded 93-94 MPG runs on a 14 mile long test course, at 35 MPH with cruise control.

Back in Pittsburgh, around early summer 2005 I found the marathon test course. That's when I discovered P&G. I would pulse to 40 MPH and glide down to 30, pulse to 40 and repeat. The average speed was about 34 mph and recorded well over 99.9MPG. But I had no way to measure it. S I ended the test by pulling out on the interstate and held the instantaneous at about 40 MPG. Then I waited for the display to drop to 99.8, noted the time spent at 40 MPG instantaneous and calculated that the P&G yielded 104MPG.

About that time I commented on one of the Prius yahoo groups that I needed something that would measure over 99.9 MPG. That got a lot of laughs and skepticism. Then Ken messaged me about the SuperMID and sent me one just before the Marathon.

But I digress. To sum up what I am trying to say, there is some benefit to P&G over steady state. The best apples to apples comparison would be the 35MPH steady state average that I achieved at Myrtle Beach, 93.6MPG, compared to the 34 MPH average speed using P&G, where I measured 104 MPG. I guess you could call it 10%. But personally, given the choice, I would choose steady state every time. It is a lot less work and yields great results. I have also tried P&G vs steady state at highway speeds, in in that case, I could find no measureable difference.

 

I updated the table ... does this seem about right?

Table 4. Summary of Results of 2004 Toyota Prius (added % improvement rjw)

	Column 1	Column 2	Column 3
0	[th]Protocol [th] PnG MPG [th] Cruising MPG [th] %
1	[tr][td]. . . Lee	Nelson	Loshse-Busch . . . [td][td]
2	[tr][td]30-40 mph	10 seconds of acceleration (NHW20) [td] 151.5 [td]81.0 @35 [td] 87%
3	[tr][td]30-40 mph	20 seconds of acceleration (NHW20) [td] 99.9 [td]80.9 @35 [td] 24%
4	[tr][td]. . . krousdb with an NHW20 . . . [td][td]
5	[tr][td]30-40 mph	'n' seconds of acceleration (NHW20) [td] 104 [td] 93.6 @35 [td] 11%
6	[tr][td]. . . bwilson4web with an NHW11 . . . [td] [td]
7	[tr][td]25-43 mph	15 seconds of acceleration** (NHW11) [td] 87.8 [td] 79.0 @33 [td] 11%

pp 6, SAE 2009-01-1322, "Vehicle Inertia Impact on Fuel Consumption of Conventional and Hybrid Electric Vehicles Using Acceleration and Coast Driving Strategy," Lee, Nelson and Loshse-Busch.

** Using cruise control resume function.

Bob Wilson

 

Yes, and I would guess 15 seconds acceleration from 30-40. It's interesting that we both show 11% improvement. I have a difficult time believing the 151.5MPG. We tried a variety of accel rates during the marathon, including 10 seconds and even less than that. We found that the accel rate had little or no correllation with FE. What we did notice was that the FE went up as we extended the glides down to 25 MPH. Near the end, Wayne was dropping below 20MPH. That was on his 120 MPG segment.

 

My speed range was limited to the cruise control speed range ... key to a reproducible protocol.

I've monitored some of the 1,000 mile club tanks and it seems a stretch too. Now the paper mentioned their state of charge decreased during some of the PnG segments but we don't know how long these PnG test sessions were. I'm already running into problems trying to measure fuel consumption at speeds under ~35 mph because of the new battery in my wife's 2010 has the original Ahr capacity. I have a circular track posted at 25 mph so I have to run multiple laps until the system stabilizes. But I don't have a good track for the range from 25 mph to ~40-45 mph.

I didn't have the details but the log pretty well showed what was happening:


Thank you for the details.

Bob Wilson

 

Hi Va...,

Why not drive the car until it reaches 6 bars in that test protocol. The 6 bar level is what the car wants to do. That way, you will truely see the P&G mileage, not the P/Charge&G mileage. When I am driving, I am trying to use the uphill pulses to get the charge back to 6 bars. Because with all the starts-from-stopped I do, I invaribly get down below 6, even though I back out of the throttle as quickly as possible to drop the battery bar on my starts-from-stopped. I think the Prius will shut the engine off much more readily whe SOC is 6 bars.

 

Hi Chris,

Let me suggest one modification. Achieve steady state in each protocol, which may take a couple of miles and then start the measured segment. This eliminates trying to use the battery SOC indicator. But the real goal is to eliminate the transients from starting either over or undercharged. It is the steady state performance between the two driving styles that is of interest.

Also, to match the SAE paper table:

• min-max speed
• average speed
• acceleration time (given vehicle weight, lets us know to power setting used)

This will make it easy to add your entry to the table for an apples-to-apples comparison.

Thanks for performing the experiment. Controlled tests take time and effort and I for one always appreciate such efforts. It is how we learn what works and how well.

Bob Wilson