Honda Insight Tail Extension Project

[Patrick]

Originally Posted by 3-Wheeler

I think that there may be a rendering problem with the airfoil shape from 70% to 100% length. I made several larger circles (shown in forth picture with blue line) that hugged the airfoil from 70% t0 90%, then veered away from the airfoil from 90% to 100%.

The next picture shows how shallow the airfoil shape is in this region. If anyone has any ideas, please let me know.

The airfoil is not a continuous arc. It flattens toward the end.

Continuous arcs were tried as airfoils, but found to be inferior because the pressure recovery at the rear is not ideal and separation occurs.

The way that most people copy a shape like this is to take measurements of the "drop" at different stations along the airfoil, then transfer those measurements to the pattern. Use a flexible "stringer," such as a long, thin piece of wood to connect the "dots" and define the final shape.

[aerohead]

Originally Posted by 3-Wheeler

Since the tail extension build is getting about two feet in length now, I decided it was time to ensure that the master template I made from AeroHeads Streamlining Airfoil was correct in shape.

To do this, I cropped his streamlining airfoil to a smaller size and placed it in a CAD program for analysis.

I drew several large diameter circles, as you can see in picture one.

For the first 60% or so, the circles seem to be a good match for the curvature of the streamlining airfoil. This is the portion of the curve that matters most to me since my tail extension will not be longer than this.



Here is a closeup of the two red circles as they fit above and below the streamlining airfoil.

It shows that my constant master template does indeed match this shape as I did my best to get this shape into a constant curvature.

This same master template is now ready to create the curvature of the remaining two feet or so of tail extension.



I think that there may be a rendering problem with the airfoil shape from 70% to 100% length. I made several larger circles (shown in forth picture with blue line) that hugged the airfoil from 70% t0 90%, then veered away from the airfoil from 90% to 100%.

The next picture shows how shallow the airfoil shape is in this region. If anyone has any ideas, please let me know.



This picture shows how the 90% to 100% line looks distorted in shape. Notice how far away the blue line is from the airfoil at 90%, but then suddenly gets much closer at 100%?

I tried extracting data from the airfoil and placing values in a spreadsheet, and that's when I noticed there was something fishy with the airfoil rendering. The angular numbers did not make sense from 60%, on.




Jim.

Jim,I did a graphical analysis of the airfoil curvature and it is defined by 4-separate and distinct radii,sweeping through 4-different degrees of rotation about each axis.Here's what I came up with :
(1) R= 3.35-H @ 12.5-degrees rotation
(2) R= 5.294-H @ 12-degrees rotation
(3) R= 6.588-H @ 7-degrees rotation
(4) R= 7.96-H @ 9-degrees rotation

This was actually for Mair's 22-degree boat tail but the 'Template' is close to the same solution.just smoother.
As with the template,H is measured from the road to the top of the max roofline camber point.
Hope this helps,Phil

[3-Wheeler]

Originally Posted by Otto

Got pics of that motorcycle fairing?

Hi Otto,

Sorry your post ended up at the bottom of the page, and I just noticed it yesterday!!

Yes, there are pictures listed here:

Picasa Web Albums - jsmosher - Motorcycle-Fa...

I had posted the album some time ago, but unfortunately forgot to make the posting "public" in the settings, so it did not show up.

Hope this helps.

Jim.

[3-Wheeler]

Originally Posted by aerohead

Jim,I did a graphical analysis of the airfoil curvature and it is defined by 4-separate and distinct radii,sweeping through 4-different degrees of rotation about each axis.Here's what I came up with :
(1) R= 3.35-H @ 12.5-degrees rotation
(2) R= 5.294-H @ 12-degrees rotation
(3) R= 6.588-H @ 7-degrees rotation
(4) R= 7.96-H @ 9-degrees rotation

This was actually for Mair's 22-degree boat tail but the 'Template' is close to the same solution.just smoother.
As with the template,H is measured from the road to the top of the max roofline camber point.
Hope this helps,Phil

Hi Phil,

I'm trying to understand your geometric description above, and spent about an hour using the CAD to analyze your numerics. Here's what I came up with so far, but I am making some assumptions:

1) Not knowing the exact shape of ideal airfoils, one assumption I am making that that all extended radii used to duplicate the airfoil shape are referenced to the maximum camber of the wing, or the thickest point of the airfoil.

2) This means that all radii center point originate from a tangental line draw through the maximum cordal thickness.

3) Being that I am not designing my own wing shapes, the assumptions above may not be true.

The first graphic shows four large circles draw over the AeroHead Streamlining Template.



The second graphic shows that the transition from the 3.35-H radius to the 5.294-H radius is incredibly short, and shown in location by the short black line. Your pictorial that you are working from must be much more detailed than the one I have to work with, because I could not make out that much detail.

The 5.294-H radius curve fits the pictorial quite well, and for quite some distance.



This graphic shows that the transition to the third larger radius occurs close to the 60% position on the streamlining template.

Notice that the point of intersection between the two lines, there is a small
"dip". This transition should be smoother to match the pictorial.



And here is the intersect point for the third and fourth radii. Again there is a dip at the point of intersection.

One way I can think of to avoid these dips is to create a airfoil shape where the wing curving surface is created with a constantly enlarging radius, or at least broken into smaller sections so that these dips would not be an issue.



Based on the above, it still appears that my constant radius Master Template made for the Insight would provide a good match to the AeroHeads Streamlining Template out to about the 60% point.

Because the car is currently jacked up I can't measure the height of the roof line exactly, but estimating with a tape measure, I got about 50 inches.

The 60% point on the Streamlining template is therefore about 103 inches back from this point, and this ends up about 42 or so inches behind the hatchback spoiler trailing edge.

Assuming the Master Template has the proper curvature in it, the tail extension should work fine by creating the remainder of the extension distance utilizing the same curvature it has up to the two-foot point, and extending out another 1-1/2 or so.

Jim.

[3-Wheeler]

This shows the entire tail extension after removing from the car, and placed on the ground.

Getting the entire section past the muffler pipe took a while.



Here is an inside view of the extension after sanding the lower portion smooth. As usual, lots of sanding dust! It gets interesting on a windy day, and you get a face full!!



One of the neighbors was just starring at the extension in this picture. He could finally get a good look at what all the construction was about.

I asked him "what do you think?", and he said, "it's really hot today!".



Fiber glass cloth was just wetted-out in this picture. You can see how smooth this section is.



This detailed picture shows something that *could* work into a problem in a high stress area of the build, depending on how high the stresses are; air pockets!!

The foam was sanded smooth in this area, but lacked attention to small divots in the foam. When wetting out the glass, it became immediately apparent that this area was not smooth enough for 100% adhesion.

Trying to work extra resin in this area is very time consuming, and usually not very effective when trying to wet out the entire area while the resin is still free-flowing.



Now the extension is tipped on it's side and glass placed in various areas before wetting out.



Inevitable low spots are again filled in with spackling. These low spots are very easy to create when a foam panel is glued a little off location.

Sometimes, this low spot does show up readily. For example when the surrounding area is sanded to create the slight curvature of the panels, that's when a low spot becomes more apparent. It's all part of making compound shapes in foam.



Here is another low spot. They seem to be quite popular in this project!!



Glass work done several days earlier is now fully cured.



This picture shows the rear hatch resting on the extension. From here, it's relatively easy to tell that only three inches are required for the hatch to clear the extension when opening.

But maybe 3 inches is not enough? What about room for the hinges to the support the smaller hatch?



An executive decision was made, and I used 5 inches. The black marker shows the anticipated cutting path.

I was quite tentative at this point, as any mistake could be tough to fix, since there is a nice layer of glass on the back side in this area.



There it is!! The cut was carefully made and the hatch actually cleared in about 98% of the area.

Some fine sanding was needed near the outer edges of the cut, for the hatch to stop rubbing when opening.

The fine sanding was confined to an area on the *inside* edge of the foam, so the sanding will not show in the area of the seam to an outside viewer, which is very nice indeed.



Visual progress has slowed up quite a bit lately as the hatch latching mechanism still has to be worked out and implemented before I continue with extending the tail further. This is the point where a lot of *eye-ball* engineering is taking place, and is less apparent in the build.

Jim.

[aerohead]

Originally Posted by 3-Wheeler

Hi Phil,

I'm trying to understand your geometric description above, and spent about an hour using the CAD to analyze your numerics. Here's what I came up with so far, but I am making some assumptions:

1) Not knowing the exact shape of ideal airfoils, one assumption I am making that that all extended radii used to duplicate the airfoil shape are referenced to the maximum camber of the wing, or the thickest point of the airfoil.

2) This means that all radii center point originate from a tangental line draw through the maximum cordal thickness.

3) Being that I am not designing my own wing shapes, the assumptions above may not be true.

The first graphic shows four large circles draw over the AeroHead Streamlining Template.



The second graphic shows that the transition from the 3.35-H radius to the 5.294-H radius is incredibly short, and shown in location by the short black line. Your pictorial that you are working from must be much more detailed than the one I have to work with, because I could not make out that much detail.

The 5.294-H radius curve fits the pictorial quite well, and for quite some distance.



This graphic shows that the transition to the third larger radius occurs close to the 60% position on the streamlining template.

Notice that the point of intersection between the two lines, there is a small
"dip". This transition should be smoother to match the pictorial.



And here is the intersect point for the third and fourth radii. Again there is a dip at the point of intersection.

One way I can think of to avoid these dips is to create a airfoil shape where the wing curving surface is created with a constantly enlarging radius, or at least broken into smaller sections so that these dips would not be an issue.



Based on the above, it still appears that my constant radius Master Template made for the Insight would provide a good match to the AeroHeads Streamlining Template out to about the 60% point.

Because the car is currently jacked up I can't measure the height of the roof line exactly, but estimating with a tape measure, I got about 50 inches.

The 60% point on the Streamlining template is therefore about 103 inches back from this point, and this ends up about 42 or so inches behind the hatchback spoiler trailing edge.

Assuming the Master Template has the proper curvature in it, the tail extension should work fine by creating the remainder of the extension distance utilizing the same curvature it has up to the two-foot point, and extending out another 1-1/2 or so.

Jim.

Hey Jim,just caught your post.
The relationships I came up with are from Mair's boat tail research.His table appears in Hucho's 2nd Edition as Fig.4.41,after reference 4.40.It also appears in an SAE Paper from around 1969 when he first published his results.
I figured that his Doctorate trumped my B.S.,and we'd be able to bet on his work.
It turns out that if you remove the cylinder portion of his windtunnel model,substituting the 1.3 d nose section onto the tailcone,you end up with a body-of-revolution of L/D = 2.5,just like the 'Template.'
My numbers are realized at 'small' scale,so it's not surprising that there's some 'wiggle' in the coordinates.
And the axis of rotation is exactly like your doing,merely moving further and further away from the point of max camber ( max thickness for a wing section ).
I think that as far as DIY'ing,if one just follows the 'trend' line of the 'Template' curve,you'll end up with fully-attached high-quality flow to wherever one decides to do the truncation.
When I build,I literally bend a length of 1-inch Schedule-40 PVC pipe across some bulkhead stations to define the points in between,allowing the pipe to do the curve-smoothing.It's kind of a Clarence 'Kelly' Johnson thing.
Stainless steel would be better,as it has superior uniform elasticity during bending deformation.
If I were building an experimental aircraft,I'd sweat every detail.
I am learning to have the 'discipline' you demonstrate on your projects but oftentimes,time constraints and immoveable deadlines force me to wam,bam,thank-you-ma'm things together,which led to the 'Template',as it was meant to help expedite the build with a high degree of confidence of road success.
Hope this has helped,I'll await your response,happy modding!,Phil.

[3-Wheeler]

Originally Posted by aerohead

Hey Jim,just caught your post.
The relationships I came up with are from Mair's boat tail research.His table appears in Hucho's 2nd Edition as Fig.4.41,after reference 4.40.It also appears in an SAE Paper from around 1969 when he first published his results.
I figured that his Doctorate trumped my B.S.,and we'd be able to bet on his work.
It turns out that if you remove the cylinder portion of his windtunnel model,substituting the 1.3 d nose section onto the tailcone,you end up with a body-of-revolution of L/D = 2.5,just like the 'Template'.....

Hi Phil,

Thanks for the response, and clearing up several things for me.

I wrote the previous post just to have you look it over and make sure I was applying the correct concept to approximating a wing shape, with regard to the Template.

Jim.

[BigSherm]

Originally Posted by 3-Wheeler

About 16 years ago, I made a composite foam/fiberglass full coverage motorcycle fairing.

At the time, Craig Vetter was in full swing with his WindJammer fairings, which weighed close to 40 lbs.

My foam/fiberglass model, which actually covered more of the bike, including riders legs, weighed only 13 lbs.

The tail extension on the Insight should weigh less than 30 lbs.

Jim.

Hey Jim,

Please tell uys more about this!

Thanks,
Scott

[3-Wheeler]

Originally Posted by BigSherm

Hey Jim,

Please tell uys more about this!

Thanks,
Scott

Hi Scott,

Check here:

Picasa Web Albums - jsmosher - Motorcycle-Fa...

Jim.

[silverinsight2]

Visual progress has slowed up quite a bit lately as the hatch latching mechanism still has to be worked out and implemented before I continue with extending the tail further. This is the point where a lot of *eye-ball* engineering is taking place, and is less apparent in the build.

Will the interior of the tailcone be available for light storage?