Designing a tail extension for the first-gen Insight

[aerohead]

Originally Posted by RobertSmalls

Ah, yes, there it is.

Here's what happens when you follow the template literally:


Here's the same, with the bottom curving upward along a spline that crosses a 7° construction line about 2/3 of the way along its length.



The rear truncation is just slightly taller than a US license plate. Looks like you will have room on both sides for reverse lights and a license plate light.

Also, as you can see, I was a little sloppy where the tail attaches to the car, leaving a gap.

EDIT: I should add, this model is 4.89m long, 95cm longer than stock. That's much better than the 5.42m that I predicted in post #1 of this thread.

Robert,I think you'll like that section chop at 70% of the effective aft-body.I've spent a lot of time recently going over Hoerner's fuselage work,also The Jaray,Kamm,Lay,and Morelli CNR banana car phantom tail stuff.
I'm not ready to present that in a thread yet,but I can tell you categorically,that if you can extend the body to 70% of the template you're basically "there".
The region where the license plate and lights will be will be within this phantom flow with almost immeasurable drag penalty compared to the full length tail.

[3-Wheeler]

Originally Posted by jimepting

... I been looking at the rear of my own Insight and rolling some of this over in a very rough fashion. From a practical viewpoint, it looks like one would want to remove the rear bumper facade, the tail lights, and the hatch minispoiler to get a "clean" starting point and attachment point...

Yes, that's exactly what I have in mind as well. For sure, remove the rear plastic bumper, remove the tail lights, which will then reveal areas on the body work for attachment points of the extension.

As far as the hatch spoiler, I might just want to leave that intact on my car. I would like all the mods to be easily reversible if need be.

Jim.

[3-Wheeler]

EDIT @ 7:39pm The following calculations are assuming the Aero Template is resting on the ground plane, and not lined up with the under carriage.

OK, Robert got me thinking about the accuracy of my first attempt at duplicating the curvature of the AeroHead Aero Template.

So it was time for another take, and this time with a little more precision.

You recall the first graph, which showed the Aero template and Honda Insight. Looks pretty consistent in shape, right?



So let's calculate the change in Y-data, or delta-Y and then graph that relation. Now how consistent does the graph look?



It turns out that picking data points from the Insight photograph posted earlier in this thread, is not so easy to do. So how do we fix the template data?

I graphed the change in curvature of the template data, and found out that the change in angle gave a constant ramp/rate change, and so the plot below shows a curve that was generated from zero all the way to 22 degrees over our length of 183.5 inches, which duplicates the length of the graphic by Metro.



Next, using some trigonometry, I reverse calculated the effective change in the Y-Axis Offset to mimic the ideal curvature change, as seen below. This was done by altering the X-Axis data in steps of 0.100 inches, all the way out to 183.5 inches. This amounts to spreadsheet with about 1800 data points! This gives us very good accuracy.



Lastly, Robert will ask for the R^2 value of this ideal curve, and I shall show him the following error graph. That shows that the error is less than 0.001 inch quite easily. More than adequate for our purposes of making a card board template.



Almost forgot to include the new polynomial that generated the above curve...

y = 0.00149991110386985 * x^2 - 0.0000513899089202675 * x - 0.000105687318828214

The equation gives us the ideal AeroHead Template shape with very low error, of less than 0.001 inch.

Jim.

[3-Wheeler]

Originally Posted by RobertSmalls

OK, now this might sound like a dumb question...

Is the intent of the AeroHead Aero Template to be aligned with the underside of the car

OR

should it be set on the ground plane (aka Metro)?



The placement of this template changes the effective length of the tail greatly.

Thanks, Jim.

[RobertSmalls]

The two are almost the same.

I aligned the ground plane of the template with the bottom of the model's wheels. I would have aligned the line at the top of the template with the top of the roof, but that would result in a discontinuity in slope where the hatch ends, due to Honda's slightly sharper curvature. The similarity in ground clearance is coincidence.

Here's what it looks like if your tail has a flat bottom:

The sides converge before too long, so there's no need to extend your curves beyond 60 inches or so.

[orange4boy]

Nice work guys.

The template is a good guide, but are you going to tuft test a prototype first?

[3-Wheeler]

With the theoretical airflow template curvature worked out, now it's time to attempt to determine the shape of the tail extension bodywork.

To do this, some measurements from the car are taken. But how does one measure angles on the car? What reference point does one use?

One way I could think of to solve this, was to attach a "rod" to a step ladder and place the step ladder directly behind the car in the following manner:

1) Position the car to that it faces straight on the driveway.
a) this was verified by opening the driver door and aligning the step with the edge of the driveway. I had to move the car back and forth manually to accomplish this.
2) Position the car so that the tail was right over a perpendicular slab edge in the driveway.
3) Use this edge to align the step ladder also perpendicular to the back of the car
4) Align an aluminum beam such that is parallel with the ground
5) Use this reference beam to ensure that all photographs taken can use the beam as a reference point for angular measurement
6) Any picture taken, must have the beam also in the picture to allow measurement from "straight" of the surface of the trailing edge of the car.

This first picture shows the step ladder, along with parallel aluminum beam, reaching over the tail of the car. This is our movable data point for later measuring surface angles of the rear portion of the body.



This is the measured angle at position #1



Position #2



Position #3



Position #4



Position #5



Position #6



Position #7



Position #8



Position #9



Position #10



Position #11



Position #12



Lastly, this shows the locations where each angle measurement was made.

Robert had mentioned that his tuft testing showed some of the yarn not adhering to the glass at the outer edges. This makes sense, as you can see that the effective angle of departure in these locations is actually steeper than in the middle portion of the hatchback.

Also note that no attempt was made to "smooth" the transitions between various angles. These angles all can reveal the measurement uncertainty of my process. When sliding the step ladder along the ground, it is quite easy to get the legs slightly twisted, and thus adversely affect the angular measurements on the side of the car.

The author attempted to minimize this effect by sliding the ladder carefully, although it's easy to see there was still some non-parallelism of the ladder legs present in these measurements especially for measurements #1, #2, and #3.

EDIT @ 8:53pm After checking my numbers, it seems that the horizontal angles are steep by about 2.5 degrees, and this is most likely due to the step ladder resting on the sidewalk portion of the driveway, which has a slightly different angle than the slab that the car sat on. Whoops!! I have adjusted the angles with this in mind.



Jim.

[RobertSmalls]

Originally Posted by orange4boy

Nice work guys.

The template is a good guide, but are you going to tuft test a prototype first?

That would be a good idea. I'm hoping to test the CAD model, but a good CFD simulation is amazingly complex. Hopefully, I'll be able to figure it out.



I plan to install the tail, and leave it on permanently. So I was thinking about what it would be like to live with a tail.

How will you open the hatch? A tail will bury the lock cylinder and hatch release switch. You don't need the lock cylinder as long as your 12V battery has juice, and the hatch release switch can be replaced with a momentary switch located anywhere. You will still need to pull up on the hatch while pressing the momentary switch. Maybe move the switch to a handle just below the spoiler, and cut a recess into the tail big enough for three fingers to access it.

You will still be able to stand beside the car and load and unload cargo. You lose the ability to unload from behind, though. That will make the lower cargo basket much less acccessible, and getting at the spare tire will require climbing. That's okay, I only do that twice a year, during tire rotations.

I'd like to add a cargo basket, i.e. a "trunk", inside the tail. It could be entirely outside the car, with its own door. Or if you're feeling ambitious, you could make its lid part of the hatch, and cut away the barrier between the car and the new cargo basket.

The new cargo basket should be strong enough for 100lbs of cargo, shifting around as you drive. That would require a much stronger tail than one that is merely an aerodynamic device, but it would be nice to have in a vehicle that has so little cargo space.

[orange4boy]

We love the CFD sims. Looking forward to that. Hopefully Trebuchet can give you some pointers. Is he still with us?

[RobertSmalls]

Originally Posted by 3-Wheeler

Very meticulous, Jim.

So if Honda's curvature is a little too sharp in positions #8,9,10, would the correct response be to make the tail extension a little flatter in those areas?

Jim, I'd encourage you to do some quick tuft testing, taping a few dozen 4" strands of yarn to your glass and watching it in the rear view mirror. I found it very educational.