Drag Coefficient, and other parameters wanted Options

[astronomerdave]

Does anyone know where I can find this information (for '72 914/4) ? . . .

• drag coefficient
• vehicle height, width
• gear ratios

Thanks,
--Dave

[jd74914]

Why don't you take a picture, print it on graph paper, count the squares, and then figure out the apsect ratio of the squares? I'm not sure how else you would calculate something like that without building a full CAD model and taking a crosssection.

[Richard Casto]

Why don't you take a picture, print it on graph paper, count the squares, and then figure out the apsect ratio of the squares? I'm not sure how else you would calculate something like that without building a full CAD model and taking a crosssection.

I know very little about Cd CdA values, but I agree that you could do this via digital photo. Use a quality lens (to avoid distortions), get back a bit and zoom in to flatten the perspective, include a reference measurement (such as a yard stick, etc.) in front of the car. Take the photo and overlay a digital grid over the image that matches up to your reference measurement and count squares. The more granular the grid, the more accurate your results. If you want to go crazy, count pixels!

I think there are computer programs that will measure the area of an irregular shape in an image file. I think Adobe even has one. There might be cheaper or free software that does that.

If you want to do it manually, I can see getting a few large poster boards, marking them out with a grid, put them on the back of your garage, use a laser pointer attached to something to keep it perpendicular to the car and then trace the outline of the car. You would need a helper to trace the "shadow" you create with the laser pointer as you move it around the outline of the car on the poster boards. When done, count squares. If you park the car between to parallel walls, with the car perfectly perpendicular to the walls, you should be able to make this work. You could create a jig that holds the pointer perpendicular to a wall and then just slide the jig around on the wall as you trace the outline.

[johannes]

Why don't you take a picture, print it on graph paper, count the squares, and then figure out the apsect ratio of the squares? I'm not sure how else you would calculate something like that without building a full CAD model and taking a crosssection.

I know very little about Cd CdA values, but I agree that you could do this via digital photo. Use a quality lens (to avoid distortions), get back a bit and zoom in to flatten the perspective, include a reference measurement (such as a yard stick, etc.) in front of the car. Take the photo and overlay a digital grid over the image that matches up to your reference measurement and count squares. The more granular the grid, the more accurate your results. If you want to go crazy, count pixels!

I think there are computer programs that will measure the area of an irregular shape in an image file. I think Adobe even has one. There might be cheaper or free software that does that.

If you want to do it manually, I can see getting a few large poster boards, marking them out with a grid, put them on the back of your garage, use a laser pointer attached to something to keep it perpendicular to the car and then trace the outline of the car. You would need a helper to trace the "shadow" you create with the laser pointer as you move it around the outline of the car on the poster boards. When done, count squares. If you park the car between to parallel walls, with the car perfectly perpendicular to the walls, you should be able to make this work. You could create a jig that holds the pointer perpendicular to a wall and then just slide the jig around on the wall as you trace the outline.

Yes Sketchup does this... This software is mostly focused on architecture and architects need surface figures.

Your method with the squares is way to complicated. Usualy, you decompose the shape in basic geometry like rectangles, triangles, disc and you can approach the surface very quikly.

[Richard Casto]

Your method with the squares is way to complicated. Usualy, you decompose the shape in basic geometry like rectangles, triangles, disc and you can approach the surface very quikly.

Sorry, yes, I assumed whoever would do this would realize that you do not need to count every granular grid square individually (1, 2, 3...). Large areas can be resolved quickly by converting into shapes. As you say, boxes, triangles and portions of circles alone would get you pretty close. Clearly its the area around the edges that require more attention and how close your shapes matches defines your accuracy.

And clearly you have shown a cheap way to do this via computer. This assumes your image you are starting with is accurate. I have seen that image before and I assume it is sourced from a Porsche document of some sort. The question is if that is an accurate rendering, or an artists rendering that just happens to look right. If I was taking the time to do this, I would create an accurate image via photo just to be sure. That also lets you handle different car configurations (GT flares, wider tires, different air dams, etc.)

It should be relatively trivial to create an outline of the car via a photo. I have used Sketchup very little, so I can't say if it will, but I have used other tools to create vector diagrams by tracing an outline of a photo which could then be imported into Sketchup.