ZVW11 experiment
 

TEST A - 53.6 MPG @70 mph
http://hiwaay.net/~bzwilson/prius/shield_A_010.jpg

TEST B - 51.6 MPG @70 mph
http://hiwaay.net/~bzwilson/prius/shield_B_010.jpg

TEST C - 53.3 MPG @70 mph
http://hiwaay.net/~bzwilson/prius/shield_C_010.jpg

Protocol and test conditions

• 57F (14C), dry pavement, humid
• two runs, West and East, south wind at ~10 mph
• I-565 Jordan Lane Huntsville AL to Mooresviille AL
• cruise control maintained speed, GPS calibrated
• +13 miles each leg, added together for total performance
• 2010 Prius (VZW30), 8,500 miles ODO, OEM tires, 45 psi
• ICE and transmission oil changed at 5,000 miles
• Shell 87, E10
• Warmed up at least 20 minutes before first run
• Disposable, cardboard shields held with duct tape

Bob Wilson

Thanks Bob,

I've been wondering about that as I have the "invisible" lower grille block in my 2003. I imagine the results would be similar.

So that's close to a 5% improvement just from a flush grille block? Nice.

I was curious about the relative efficiency of front flush versus rear a rear block. In aviation, we use cowl flaps located at the air outlet to control cooling air through an engine compartment and at high speed, reduce drag. But most of those air inlets were fairly normal or flat relative to the incoming air. Our Prius has a 'bend' and I wondered if excess air might dip-in and bleed out the lower edges and take away energy. However, the measured difference 0.3 MPG versus 2.0 MPG is pretty much 'in the noise.'

From a design standpoint, this means a set of sliding, black, rear panels will work. Right now, I'm thinking of these control laws:

• ICE < 85C - drives to closed position
• ICE > 95C - drives toward open position
• PID control - to smooth operation and partial opening
• curved and feathered lip - to reduce noise at high speed

In the summer, I'll have low drag in the morning commute. In the late afternoon commute home, the vent opens to avoid overheating the ICE. I'm not too worried about ice freezing the slide as it is difficult to overheat even with a full block at 32F/0C.

Bob Wilson

Thx Bob, can you add instrumentation to your conditions? i.e. was it toyota stock or sg or ?? Also can you note if you checked the state of charge before and after each test?

Dave.

Very nice Bob! Thanks for the testing. :)

Hi Dave,
The purpose of the 20 minute warm-up at speeds of 45 mph and higher is to normalize the charge and eliminate warm-up issues. I simply use the built-in display to handle the distance, mileage and total speed. The GPS is used to check the actual versus indicated speed. The +1 MPH error on the speedometer (the car reads higher than actual) remains but the trip meter speed is dead on accurate and agree with the GPS.

Future experiments:

• rear wheel well covers (aka., Insight style)
• front air dam and side skirts

Duct tape and cardboard allows a lot of flexibility to run inexpensive experiments and get useful data. Once we have quantified the effects, one system test and then look at fabrication of a longer lasting, weather proof solution.

Bob Wilson

Is it safe to say the battery did not make any unusual contributions or detractions from any of the test?

I suppose this may be viewed like asking if you were dragging the brakes on test 2, but verifying the state of charge before and after the test on a hybrid would help eliminate the possibility of the battery deciding to intervene, even without the operators knowledge. There can be lots of variables in a computer program (and variables in the folks who write them).

Yes:

1. pre-test warm-up - by keeping the car at or above 46 mph, the battery soon reaches its nominal SOC. In this case, the warm-up area was at high speed, the Parkway at +50 mph, and the approach to the test runs on level ground, Governor's drive exit onto I-565.
2. After 10 miles, the MPG does not significantly change - another key element is the vehicle achieves a steady state on a fairly level run. The local variations have been 'averaged' out of the final numbers.
3. Proportionality between three West and three East runs - West: 53.0, 52.3, 52.9 and East: 54.2, 50.9, 53.7. Although the exact ratios vary, the order remains the same: A>C>B ... suggesting a valid data sample.

However, I agree a better procedure would be more randomized trials along the lines of an "A" "B" "A" series. Serendipity, the vehicle baseline runs without an air block were sandwiched between the two experimental configurations. Since I plan to do some wheel well cover tests and air dam and side skirts, I'll see if some of the combinational tests can mix it up a little better.

BTW, the results are consistent with what I had expected. The rear block should have some flow of air in that spills out the lips on either side. This should result in a loss of momentum and higher drag. The front, flush block would keep the air stream displaced but at velocity. The unblocked air inlet should cause significant energy loss as the air that enters is brought to near vehicle speeds and then dumped overboard.

I was lucky on Sunday. It was overcast which kept the temperatures fairly constant during the day. The last run was just at sunset and the temperatures were still what I'd seen earlier that afternoon.

Bob Wilson