5th generation civic hatchback - improving aerodynamics

[diesel_john]

i have had a tough time isolating how much difference one change makes in the mileage. every part is so dependent on whats ahead of it and whats behind it in the air stream. i now have the truncated back, full belly pan except under the engine oil pan, single side skirts, and front splitter. gains are around +5 mpg. but other versions with full egg shaped front with air dam, full front and rear fendercovers ending at the rear bumper, double side skirts and no fastback gained as much as +9mpg.

IMHO the most important traits in the back are that all surfaces (top, bottom,sides) end at the same time with the same amount of pressure against them. And as the air leaves the trailng edge that it flows off exactly parallel to the directions of the vehicle, not curving inward towards a low pressure area. imho tapering the trailing edge only works best when the rear comes to a point. if the yarn tufts blow straight back you got it.

[Big Dave]

Lookin' good, AndrewJ.

Some just talk the talk, but you are walkin' the walk.

[boxchain]

Awesome project! If I keep seeing projects like these, I may start aeromodding myself!

Originally Posted by AndrewJ

After having it on the car, I've decided that I'm going to try to find a hinge to put on the front part of it tomorrow. A hinge will greatly facilitate tire removal if (when) I get a flat.

Save yourself a lot of trouble (and weight), where you drilled out holes in those brackets, take a dremel or equiv and cut a channel out of one end of the hole:
| |
|C
| |

Now you can remove them with 1/2 turn of the screw, lift and the bracket should pop off and you won't lose the screws

[AndrewJ]

So, I just did my first long trip with Harold yesterday, I drove from McKinleyville to San Fransisco, a distance of about 275 miles. It wasn't a big deal until I got to the mountanous portion of the trip. After an elevation gain of 1800 feet in 20 miles or so, the car started acting kinda funny.

The temp gauge was showing normal temps, but the motor wasn't making enough power to climb the steeper grades. The only thing that seemed to help was kicking the heater into high gear, sucking some heat out of the engine bay.

I think I'll be cutting out a vent in the airdam for the return trip. That'll give me an opportunity to rig up a variable open-close vent using a manual choke cable. I've kinda been wanting to try that anyway.

But before I do that, does anyone have any ideas as to what would cause the motor to loose power in that situation? The coolant temps weren't abnormally high, but due to the high load and low speed it's not unreasonable to think that under-hood temps were higher than normal.

Oh, and by losing power I mean that I was unable to keep up a "reasonable" speed (50mph) even after downshifting into 4th. By downshifting to 3rd I could keep about 42mph.
I can keep 50mph on my "calibrated" test hill all the time (it's a similar grade to the hills I had trouble with), though that's the only hill on a 10 mile drive.

[SVOboy]

I have the same problem with my car in the hillz. I assumed it was just running the wrong ecu, 70 hp, and mehbe being a tooth off on the timing, but I guess this is another thing for me to check out,

[Daox]

If I had to take a shot, maybe your engine has a knock sensor and its retarding the ignition due to increased engine temperatures. Once it gets too hot you'll get pinging which the knock sensor will sense and thus retard timing to avoid engine damage. I don't know if your engine has this. I guess its possibly built in to the coolant temp sensor's programming too. Theres really no way to tell for sure what your coolant temp is without some sort of gauge. I think we all know that the hot/cold gauge on your dash isn't remotely accurate.

[TomO]

Originally Posted by AndrewJ

After an elevation gain of 1800 feet in 20 miles or so, the car started acting kinda funny.

That much of an elevation change in such a short time while the car was running will make it not run right.

Our Civics use a speed density system to calculate airflow into the motor. The elevation change was enough that the pressure decrease skewed the fueling and timing.

Next time you make that trip, when that starts to happen....pull over, shut off the car, turn the key to the run position and wait until the check engine light goes off, then start the car.

During that test period the car checks the voltages on all sensors. The voltage that it reads on the MAP sensor is deemed to be the atmospheric pressure for the outside of the car. Then the computer has a lookup table based on atmospheric pressures for fueling and ignition advance.

This process used to be printed in the owners manual for fourth generation civics, but for some reason I don't see it too often in the US manuals.

Hope that helps.

[diesel_john]

good call TomO.

key off, key on, count 1,2,3 key start.

K-cars do the same thing.

[AndrewJ]

So on my little San Fransisco misadventure, I cut a hole in my car. It's been that way for a week, and I decided today to do something about it.

On the return from San Fran, at a random gas station as I was about to turn over the wheel to my wife and get some much needed sleep in the passenger sleep, I cut a hole in the airdam right in front of the radiator, so she wouldn't have to wake me up for any power-loss or overheating issues.

So, a few days ago I stopped in the auto parts store and grabbed a 12' "Manual Choke Cable"

Step 1. Find a place to mount the choke cable knob on the dashboard. Lucky me, I have about 1000 unused dummy switch covers in my dash.






Now route the choke cable under the dash to the nearest pre-existing (if at all possible) grommet in the firewall.




Route the cable through the engine bay, around the radiator, to the airdam. Rig up a few other peices with whatever plastic scraps and spare hardware you have laying around.




Test that it opens and closes to your satisfaction, and enjoy the peace of mind knowing that your car will be much harder to overheat.

[SVOboy]

I'm extremely disappointed that these aren't pictures of an installed MID.