How to draw aircraft

[CJGibson]

Re: How to determine a 3view from a picture.

Chances are if it's in the Science Museum, we've got a GA already. Adrian Mann and I have a menagerie of exotica.

If you don't want to put it up on the forum, I'm sure a PM to the usual suspects for this (me, Paul, Jens) can only help.

Chris

 

[steelpillow]

Re: How to determine a 3view from a picture.

The aircraft is not in the museum, only its engine is. FYI it has been called the "James monoplane" by a couple of historians, but the museum's source documents do not use that name: at least one calls it the "Levis-Belmont" monoplane. I don't know how to link to another thread here, but if you search you should find where I posted about it.

The technical file they keep these documents in also includes two photographs of the plane, one is the best print I have seen of a published photo (despite a crease right across it), and the other is a photo I had never seen before and may well be unpublished. I have not yet sought their permission for anything other than private copies for research, so I am reluctant to post them here.

Before I start sending copies via PM, how secure is that system from prying eyes? I don't want to breach the Museum's trust, even by accident. Might it be better if I send then via external email?

Ultimately, I will ask the Museum if their prints can be properly scanned - I doubt the original negatives will ever be found. But can you work with lesser quality? The only digital copy publicly available is a low-resolution jpeg of the published photo. I took a handheld photo of the museum print which is better resolution but I think that it is probably slightly distorted - the upper border is gently curved. I have photocopies of both the museum images but no scanner. Would handheld photos of those photocopies do, or would any slight distortion mess up the software results? Sorry to be getting so technical/paranoid.

 

[CJGibson]

Re: How to determine a 3view from a picture.

Is it this?

http://www.google.com/patents/USD97364

Chris

 

[steelpillow]

Re: How to determine a 3view from a picture.

Nope, it's not American. It's this one: http://www.secretprojects.co.uk/forum/index.php/topic,22544.msg228130.html

 

[Jemiba]

Re: How to determine a 3view from a picture.

... I'd love to be able to get at some of that software that people have mentioned here and put it through its paces.

Not to discourage you a mite, but at least that "magical software" is at least to me unknown. There are tools, that will
help you to reconstruct a 3-view from photos and if done properly, results will be reasonable.
But to my best knowledge, there's no way just to measure some points on your photos, punch them into your software
and get a 3-view.
There's certainly no problem with slightly distorted photos/photocopies, especially not, if the kind of distortion is obvious
by, say bent picture frames, or maybe by the squeezed wheels of the landing gear.
And the rest may be easy, if the prerequisites mentioned here http://www.secretprojects.co.uk/forum/index.php/topic,22628.msg229148.html#msg229148
are fulfilled. If your photos weren't taken with a calibrated camera, well, sorry ....
But it's not the end of the street, you still can produce quite good results. And for your comfort, many drawingns, that
can be found in our source publications were made the same way !

 

[Richard N]

Re: How to determine a 3view from a picture.

Take a look at the SketchUp Match Photo tool: http://help.sketchup.com/en/article/94920


A video of the SketchUp Match Photo tool from when SketchUp was part of Google: https://www.youtube.com/watch?v=fhzpqf7fywM


The SketchUp home page: http://www.sketchup.com/

 

[Jemiba]

Re: How to determine a 3view from a picture.

Yes, there are several tools like this one, I think. I didn't try SketchUp, but ImageModeler with good results
for examples, as shown in the video: Buildings, with lots of straight and parallel lines, typical for that kind of
objects, as long, as you don't try to adapt it to a house designed by Friedensreich Hundertwasser ....
(Photo via http://unterkunft-reise.com/magdeburg-sehenswurdigkeiten/hundertwasser-haus/ )
For aircraft, it's much harder, because span of wing and tailplane often are the only lines, that can be determined
with reasonable accuracy.

 

[carsinamerica]

Re: How to determine a 3view from a picture.

I remember visiting that page when this topic was first posted two years ago. I'm regretting that I didn't make an MHT copy of it at the time, since even the Wayback Machine can't find it now.

What I'd really like is a technique to approximate the wingspan of an aircraft from a drawing. All I know is the length. It's 275' 2" in length (fuselage) and I'm assuming that the overall length to the rear of the tailplane is 281' 10". Any ideas? Close would be good enough.

 

[steelpillow]

Re: How to determine a 3view from a picture.

I remember visiting that page when this topic was first posted two years ago. I'm regretting that I didn't make an MHT copy of it at the time, since even the Wayback Machine can't find it now.

What I'd really like is a technique to approximate the wingspan of an aircraft from a drawing. All I know is the length. It's 275' 2" in length (fuselage) and I'm assuming that the overall length to the rear of the tailplane is 281' 10". Any ideas? Close would be good enough.

A rough eyeball estimate suggests the span to be approx. 6.8 times the spacing between a pair of inner and outer engines. This is likely to be common across the range. Check a scale drawing of another model. Caution: nobody else's eyeball will agree with mine, but by extending the leading edges until they meet in the middle, you will be able to measure the distances on a print and compute the ratio more accurately.

 

[carsinamerica]

Re: How to determine a 3view from a picture.

A rough eyeball estimate suggests the span to be approx. 6.8 times the spacing between a pair of inner and outer engines. This is likely to be common across the range. Check a scale drawing of another model. Caution: nobody else's eyeball will agree with mine, but by extending the leading edges until they meet in the middle, you will be able to measure the distances on a print and compute the ratio more accurately.

Thanks. The only problem is that the pictured model is a re-winged "747-500" sketch, so the spacing between inboard and outboard engines won't be the same as the rest of the 747 range. The engines themselves don't resemble any other design in the 747 family, either.

 

[steelpillow]

Re: How to determine a 3view from a picture.

Ah.

I just checked the perspective in the image and it is inaccurately drawn - lines joining the inner pair of engines, the wingtips and the tail plane tips do not converge on a single vanishing point. The craft as drawn would be wonky, even if we could scale it.

Estimating the span of the tail plane is about your only hope. Then measure the ratio of the two spans. The perspective is minimal enough to ignore here, just use a ruler and divide the wing by the tail. I'm guessing it's around 2.7. Multiply that by the actual tail span to get the wing span. Or, is the tail stretched by an unknown amount as well?

 

[Richard N]

Re: How to determine a 3view from a picture.

Take a look at Autodesk 123D Catch. It isn't exactly the answer to what was asked here, but it shows how much can be done from just a set of photos.


http://www.123dapp.com/catch


https://www.youtube.com/watch?v=OxsmnDKO7D0