Testing aerodynamic modifications

I've made a few aerodynamic modifications to the Prius this summer that I thought I would report here along with the methods I used for testing their effectiveness, in case anyone is interested and wants to try for themselves. First, here's what the car looked like this spring:



As it stood then, I had made the following changes:
-lowered ~35mm (Tein H-tech springs)
-partial cooling air intake block (aluminum sheet behind stock grill)
-rear wiper, washer nozzle, antenna removed
-exterior mirrors removed
-fairings from a C-max added in front of wheel openings, and front wheel air dams enlarged with outdoor rubber mat
-external air curtain ducts added




I was not happy with the front air curtain ducts, as they made the car less stable. I had modeled them on the ducts of the Hyundai Ioniq, which are angled downward from horizontal, and I think this was producing an upward force on the ducts themselves. So, I redesigned them and made a new set to test. Version 2.0 are taller, have horizontal vanes inside, and have a 12mm wide outlet, down from 15mm in v1.0, against a 30mm wide inlet--so the speed of the outlet flow has been increased from the first version.



After fitting them, the difference was immediately apparent; the car was noticeably more stable, which I theorize is because the yawing moment has been decreased as a result of the faster-flowing air down the sides of the body. (More on this in a bit).

What effect did this have on drag? Well, I tested them by going out to a flat section of road, 3 miles long, in the middle of the night when winds were low, temperature was constant, and traffic was nonexistent. I measured fuel economy on the Scangauge with no ducts, front ducts, and front and rear ducts over multiple runs in both directions (to average out the effects of wind). Result: with front and rear ducts fitted, fuel economy improved 1.2% at 55 mph.

Next, I tried an experiment to reduce intake air temperature, since Toyota saw fit to redesign the intake tube from the 3rd gen to 4th gen with a pickup in front of the radiator, taking in ambient air rather than warmer engine bay air that has passed through the AC condenser and radiators. I fitted a NACA duct in the front bumper cover:



2.5" tubing connects it to the air filter housing, with a secondary inlet at the low point of the tubing that also serves as a drain point for water ingestion (my first road trip with the new intake I ended up driving through a series of thunderstorm downpours; the filter and box were completely dry afterward when I stopped to check them).



I measured IAT on the Scangauge with no tube, tubing run to the stock inlet location behind the headlight, and tubing run to the new inlet and found, at 90 degrees Fahrenheit outside temperature:
-no tube: IAT 12-15 Fahrenheit degrees above ambient
-stock location: IAT 10-12 Fahrenheit degrees above ambient
-bumper inlet: IAT 6-8 Fahrenheit degrees above ambient

Finally, a couple weeks ago I purchased a manometer and pitot tube, and fabricated a pressure sensor disk, in order to measure panel pressures on the road.



I thought I would see how effective a 21mm tall lip spoiler is as my initial test, but first I taped yarn tufts to the back window to see the airflow pattern with no spoiler, the spoiler at the top of the window, and the spoiler at the back edge of the factory spoiler:


no spoiler


roof spoiler


rear spoiler

As you can immediately see, the roof spoiler causes flow to separate over the entire window--not good for lift or drag! What would pressure readings show?

I took readings at 4 centerline locations on the roof, window, and base and found the following differential pressures from ambient (all values in Pa):


no spoiler


roof spoiler


rear spoiler

As expected, the roof spoiler decreased pressure over the window substantially, but with a pressure increase on the roof ahead of the spoiler; the spoiler in the rear location had an ambiguous effect on rear window pressure but measurably increased pressure on the roof, a phenomenon I did not expect.

To further investigate, I measured pressures in ten locations on the stock spoiler, window and roof outside of the centerline, with and without the lip spoiler at the rear and found:


no spoiler


spoiler


change in pressures from fitting a lip spoiler

Surprise! A substantial and consistent increase in pressure simply from fitting a 21mm lip spoiler, all the way up to the middle of the roof.

So, I fitted the lip spoiler permanently to drive for a few weeks and see how the car feels and if there is any measurable impact on fuel economy, which would indicate higher or lower drag:



Next up, I will measure pressures on the side of the car with and without the air curtain ducts I wrote about previously; I expect to find lower pressure (higher speed) with them fitted. Then, I will be measuring pressure on the rear diffuser with and without diffuser vanes to see what effect, if any, they have.

All these techniques are covered in an inexpensive book by Julian Edgar, former editor of Autospeed, called Modifying the Aerodynamics of Your Road Car; I highly recommend it.

I'll reply to this thread as I complete more tests.

 

Well, manufacturers have to optimize under a variety of constraints: regulations governing car design, crash test performance, looks, cost. As modifiers of personal vehicles, we don't have to operate under those constraints.

 

Funny you should ask that--I ran the 2018 Green Grand Prix like this:



And prior to that, I built a temporary corrugated plastic fairing for a trip to Connecticut and back:



However, I found that these did not make as much difference as most people think they will. At the GGP the car got 62.8 mpg with the tail, and 59.2 mpg the next year without (in much worse conditions, with standing slush on the track). Plus, it was awful to park in Chicago and New York and limited hatch access. So now I'm more interested in reducing drag and lift, measuring changes, and doing it without extending the footprint of the car.

 

Oh, for sure more optimization is possible. At this point, however, I'll leave that for someone else to test unless I suddenly find myself with a lot more time on my hands.

My reasoning was, it mattered enough for Toyota to consider the extra expense engineering and then manufacturing an over-the-radiator ambient air intake justified. It may be that cooler intake charge produces slightly more torque and allows the engine to run at a lower RPM (which was one of the reasons engine size increased from 1.5L to 1.8L from 2nd to 3rd gen). It might be possible to test by measuring mass air flow and fuel economy in different configurations.

 

Not yet! But I am thinking of covering some of the center exhaust tunnel with aluminum sheet. The rest of the underside is pretty well covered.

 

There's always this route:



But I never understood why, in a post-apocalyptic wasteland where oil is scarce and controlled by capricious baronets, all the cars have huge, gas-guzzling engines.

 

I did a lot more pulse-and-glide on my previous car, a Civic with a manual transmission and an engine kill switch. I'll make sure to coast up to stoplights and such, but on the highway anymore I just drive at a steady speed. I don't get too close to trucks, but their wakes are huge; you can get an idea how long they are if you look at one on a wet road where water spray visualizes the wake. Even a fair distance back you might see some benefit from following a truck.

Unfortunately, there is no standard in the US for low rolling resistance tires (Europe gives a letter grade compared to a reference tire, IIRC). I think I have Continental TrueContacts right now, about 45,000 miles in; I keep them at max cold pressure on the sidewall, 44 psi.

That would be nice, but I've never really looked into what it would take--time or cost.

I don't have any experience with the "Red Bullet" solenoid or larger throttle bodies, so I can't comment on them. I did put a larger BBK throttle body on my Ram SRT-10 when I had it; it was one of the first ten made that we had a group buy on. That changed throttle response drastically. But, that was a vehicle with mechanical throttle linkage, not the electronically-controlled throttle that most cars have now.

 

You're welcome! You're right, I haven't done anything to the underside aside from a catalytic converter cover plate. I do have some cheap eBay diffuser vanes that I'm going to throw on and measure pressures under the rear of the car to see if it makes any difference.

I'm considering hood vents; a couple of years ago I used a Magnahelic to measure the pressure differential between air in the engine bay and on top of the hood in several locations and found consistently higher pressure under the hood. Venting the hood might relieve some of that pressure, improve airflow through the radiators, and possibly reduce lift. This fall or next spring I'd like to get some second-hand ride height sensors so I can measure lift directly.

 

Yeah, I might see if I can come up with a more elegant attachment in future. These have rivnut inserts in the bumper cover and machine screws so I can remove them easily for testing.