PnG: revisiting SAE 2009-01-1322

SAE paper 2009-01-1322, "Vehicle Inertia Impact on Fuel Consumption of Conventional and Hybrid Electric Vehicles Using Acceleration and Coast Driving Strategy", Jeongwoo Lee, Douglas J. Nelson of Virginia Tech and Henning Lohse-Bush of Argonne National Laboratory is the first formal paper that tested a 2004 Prius, NHW20, in Pulse and Glide versus the equivalent constant speed. One of the problems with 'I use pulse and glide' is the absence of a control . . . what happens when the car is held at the equivalent block-to-block time? This paper is the first I've seen that attempts to answer that question.

The problem shows up when I tried to analyze pulse and glide using a 2003 Prius, NHW11 and got substantially lower numbers:

	Column 1	Column 2
0	[th]low[th]high[th]MPG[th]constant[th]MPG[th]model
1	[tr][td2]30 mph[td2]40 mph[td2]151.5 MPG[td2]35 mph[td2]81.0 MPG[td]NHW20 @10 s. (Table 4	SAE 2009-01-1322)
2	[tr][td2]30 mph[td2]40 mph[td2]99.9 MPG[td2]35 mph[td2]80.9 MPG[td]NHW20 @20 s.(Table 4	SAE 2009-01-1322)
3	[tr][td2]25 mph[td2]43 mph[td2]87.8 MPG[td2]33 mph[td2]79.0 MPG[td]NHW11 @17 s.(Wilson PnG Test)

The constant speed values were pretty close, 79.0 vs. 81.0 MPG, no problem. But the pulse-and-glide metrics were off quite a bit. Now I didn't have the same model NHW20 as the authors of the SAE paper but we do have a 2010 Prius.

The paper has a set of weight adjusted, roll-down coefficients for the NHW20. In my case, I'm using the EPA roll-down coefficients to plot the drag-power curves for the NHW11, ZVW30, and their NHW20 constants:


So let's start by looking at the power required for constant speed, NHW11 and NHW20:

	Column 1
0	[th]mph[th]HP[th]MPG[th]model
1	[tr][td2]33.5 mph[td2]4.35[td2]79.0[td]NHW11
2	[tr][td2]35.0 mph[td2]4.2[td2]81.0[td]NHW20

Source: EPA_rolldown_2010.xlsx

If we look at HP as a stand-in for rate of fuel consumption (i.e., both the NHW11 and NHW20 have similar 1.5L engines), we find:

• NHW11 @33.5 mph - 79.0 MPG due to higher vehicle drag
• NHW20 @35.0 mph - 81.0 MPG due to lower vehicle drag

With lower drag power requirements, we would expect the ZVW30 to achieve:

• better MPG at 35 mph

Some preliminary results on the test track last night suggest ~85 MPG at 35 mph. The only open question is what will be the pulse and glide results relative to the equivalent constant speed.

My test track is 6 miles long from:

• 6000 Bailey Cove Rd SE, Huntsville, AL
• 320 Green Cove Rd SE, Huntsville, AL

Starting at 6000 Bailey Cove, drive the protocol towards the Green Cove end. At the Green Cove end, the terrain raises so I can shift into "N" at a given point and ballistics coast up the grade until the speed decays to 15 mph. Then do a "U" turn and repeat the route back to 6000 Bailey Cove. Between 2:00 AM and 5:00 AM, this route is all but traffic free and the stop lights default to green.

Now one of the problems with the SAE paper is they observed traction battery SOC changes depending upon the protocol yet had no way to factor it out of their results. Also, they didn't discuss how they approached vehicle warm-up. So in my case, I'll start the test at the 320 Green Cove Rd end driving the expected protocol. By the time I reach 6000 Bailey Cove, the car should be at the 'steady state' SOC charge and warm-up state for the protocol. Make a quick "U" turn and run the route with the given protocol.

The mileage will be recorded using the vehicle tripmeter display. Since the tripmeter MPG only goes to 99.9 MPG, if the MPG exceeds 100, we won't have useful data. However, we can preload the tripmeter and record the START and STOP values. Then use these values to calculate the difference.

The route will be GPS recorded using both the dashcam and a Garmin nuvi GPS. I will try to use a miniVCI to record vehicle metrics BUT it keeps asking to continue recording every two minutes or so . . . a distraction.

Bob Wilson

 

Actually with 10,000 posts here and probably as many in my Yahoo Group 'Prius Technical Stuff', Ecomodder.com, GreenHybrid.com, Yahoo Group "Prius_Owners", and the USENET group, I think I have published already. <grins>

I would rather publish where folks can get useful information and not have to pay the SAE fees. Also, I appreciate having a dialog and too often the SAE papers look a little too 'canned.' Posts here are subject to revision and correction but SAE papers (like other formal papers) are often 'frozen in time'.

Now if I bring anything useful to the party it is an emphasis on engineering disciplines and testing. Every now and then we see folks who 'do the heavy lifting' and do their own, credible experiments. These are the folks whose company I most appreciate . . . and I try to emulate.

But the other day I was looking for something from an earlier study and noticed I might have enough material for a small book:

• "A Prius Owner Guidebook" - then I laughed at myself

What does make sense is to follow good Prius Friend Hobbit's example and organize my scattered papers into a Web-based book. Something with a table of contents, index, and even spell checked!

Bob Wilson

 

I've had an epiphany about the paper while planning my testing. Looking at Table 4:

	Column 1	Column 2
0	[th]test[th]parameter[th]PnG[th]Cruising
1	[tr][td]30-40 mph	10 s. acceleration[td]FE (mpg)[td2]151.5[td2]81.0 @35 mph
2	[tr][td] (above) [td]Initial SOC[td2]52.0%[td2]56.0
3	[tr][td] (above) [td]Final SOC[td2]50.5%[td2]56.0
4	[tr][td]30-40 mph	20 s. acceleration[td]FE (mpg)[td2]99.9[td2]80.9 @35 mph
5	[tr][td] (above) [td]Initial SOC[td2]52.5%[td2]56.0
6	[tr][td] (above) [td]Final SOC[td2]52.0%[td2]56.0

Source: SAE 2009-01-1322

The clue comes from this hill climb data of the NHW11:

Notice that the amount of gas needed to climb the 8% grade hill goes down at speeds above 65 mph because the NHW11 Prius 'borrows' energy from the traction battery. That is exactly what happened in the SAE paper to the NHW20:

• accelerating 30-40 mph in 10 seconds, 1.0 mph/sec., is borrowing energy from the traction battery, 1.5%
• accelerating 30-40 mph in 20 seconds, 0.5 mph/sec., borrowed 0.5%, one third the energy

In effect, the SAE paper accelerates fast enough that the extra energy comes from the traction battery just as accelerating up an 8% grade hill faster than 65 mph draws upon traction battery power and gives the illusion of lower fuel consumption. Notice that the mileage improvement is proportional to the drain of traction battery SOC.

Now one other difference is in the acceleration profiles:

• NHW20 in SAE test uses constant fuel flow, higher initial acceleration that tapers off
  • requires additional instrumentation to replicate
  • keeps engine at one BSFC point, the equivalent energy level
• NHW11 cruise control uses constant acceleration rate, lower initial fuel flow and higher at end
  • reproducible without additional instrumentation
  • ending power may push the engine into fuel inefficient BSFC

I checked my earlier data and only recorded the traction battery current, not the SOC. But let's focus on the ZVW30 test.

Obviously, we'll need to record the traction battery SOC. But if there is a change, how do we quantify the effect on the MPG?

1. Do some sort of test that lets us measure the fuel consumed per % SOC. Perhaps the warmed-up car could have the traction battery partially drained to say 45%. Then put the car in "P" and record the fuel consumed as the traction battery is charged.
2. Use EPA values for eMPG. The math could get a little 'hinky'.
3. Run the test with an initial low MPG and see what the ending SOC becomes. Re-run the test with a normal SOC and see what the ending SOC becomes. Split the difference and drain the traction battery to that level and re-run the test again.

Thoughts?

Bob Wilson

 

Ok, start with these:

• Development of Electric Motors for the TOYOTA Hybrid Vehicle "PRIUS", Kazuaki Shingo, Kaoru Kubo, Toshiaki Katus, Yuji Hata
• Development of New-Generation Hybrid System THS II - Drastic Improvement of Power Performance and Fuel Economy (SAE 2004-01-0064) Koichiro Muta, Makoto Yamazaki, Junji Tokieda
• Model Year 2010 (Gen 3) Toyota Prius Level-1 Testing Report (ANL/ES/RP-67317) Eric Rask, Michael Duoba, Henning Lohse-Busch, Daniel Bocci

These will give good high-level views of the historical and current Prius. Then here is a short list of papers:
[pre]
0471354619_tire_dynamics.pdf
120761.pdf
121097.pdf
121119.pdf
121559.pdf
121813.pdf
122356.pdf
2000-01-c042.pdf
2002-01-1087.pdf
2009 Toyota Prius Base - Consumer Reports.pdf
2009 Toyota Prius touring - Consumer Reports.pdf
31306.pdf
890029.pdf
Anti-Hybrid_Analysis.pdf
Argonne_Lab_2.pdf
Battery_Hybrid15.pdf
Battery_Thermal.pdf
Col_Hybridreport.pdf
David_triprecords.pdf
Detailed_Miller_W04-1.pdf
Dougs_test.pdf
FreedomCAR_prius121813-1.pdf
FreedomCar_122430.pdf
Georgetown_factory_description.pdf
Isulation Diagnosis.pdf
MPG_Secrets_Whitepaper.pdf
Matsunaga electric inverter.pdf
Metal_hydride_cells_having_improved_cycl.pdf
Method_of_activation_treatment_of_Ni_MH_.pdf
NHW11_system.pdf
PNAS-2011-plug-in_report.pdf
Pace_patent_appeal_06-1610.pdf
Planet drive explain#27520D.pdf
PnG_resolver_1579n2e.pdf
Prismatic_Format_NiMH_for_Trans_and_Stationary_Mkts.pdf
Prius+as+Back-up+Power.pdf
Prius_User-Guide.pdf
Prius_die_de_wereld_#26E8E2.pdf
SAE_2004-01-0064.pdf
Scanner_UserManual-1.pdf
Series1000Brochure.pdf
Squib_tester_master_#263CFB.pdf
SuperCar.pdf
TechCompareInsightVsPrius.pdf
Tires and Passenger Vehicle Fuel Economy.pdf
Toyota_SUV_motor.pdf
aging_life.pdf
battery_temp_2a_2002_01_1962.pdf
chris-dragon-dash.pdf
design_Chris_Mi_handout.pdf
edta_vancouver_05_co#24D801.pdf
erl7_1_014003_HEV_ICEV.pdf
fact_sheet5828.pdf
fact_sheet8976.pdf
hay_in_a_prius.pdf
hydraulic_Hydrid brochure.pdf
mini_scanner_UserManual.pdf
multi5_set_defaults.pdf
prius-e85.pdf
roll_down_testing_paper.pdf
sae_gov_ind_may_2005_hevs.pdf
tire_duleep.pdf
tire_rolling_resistance (trans research).pdf
tireinflationpressure.pdf
toyot.pdf
transaxle_thermo_122586.pdf
www-coastaletech-com_keys-htm_cjtvlqko.pdf
www-coastaletech-com_transmitters-htm_3s4ses2g.pdf
www-slate-com_toolbar-aspx_rsp55fqr.pdf
[/pre]

Bob Wilson

 

Try Mr. Google?

Bob Wilson