Old School Drawing Exercises and 2D CAD

Fair effort, the curve is fuller than mine, I give it a B+, same mark I give my attempt at scaling and aligning our two drawings!

Here's how my book does it:

1. Find the origin of the blue 1.5" radius outer semicircle by drawing two 1.5R arcs from points A and B
2. Draw the blue semicircle and then several perpendiculars to it from the 60° line AB
3. Then draw connected horizontal lines of the same length, the ends of which define the required curve.

A CAD package should include a spline curve tool to span all the construction ends and draw a clean curve. The easiest way to join them manually is with a French curve tweaked to blend nicely by eye. Not as easy as it sounds to draw really smooth curves by hand, my attempts usually have obvious joins! And see Georgineer's post about being marked down for blending his curves too well in a City & Guilds exam!

Comment on the book method - unlike bending a spline, or picking a random French curve out of the box, it generates accurate reproducible dimensions for the maker. Whether or not the result more artistic than, say, Jason's fuller example is a matter of opinion. Might just be fashion, but some curves, like a Spitfire's wing, definitely look better  to most people. I've no idea why!

Dave

Edited By SillyOldDuffer on 04/07/2020 11:59:29

.

I believe that's because they were laid-out full size, and were actual spline curves

... inherently beautiful because they are ‘natural’

MichaelG.

.

As an aside ... I have just found [but not yet studied] this:

https://www.aerosociety.com/media/4843/the-spitfire-wing-planform-a-suggestion.pdf

Edited By Michael Gilligan on 04/07/2020 12:20:34

Here's a test that might be of interest. The yellow spot indicates a tangent.

I did it a bit differently.

Drew out the shape with a diagonal lien inplace of teh curve and extruded that.

Added an additional plane at an angle, in this case 45deg

Drew the semicircle on the new plane and positioned it so the curve passes through the two points where angle meets vertical

Cut extrude the semicircle and the end view is what I posted, playing with the angle of the plane will alter the severity of the curve produced.

I like it, and there's a parallel between creating a new plane and my book's tilt a semicircle method.

As expected I had trouble using the book approach with FreeCAD's Part Design Sketcher. The lack of Layers to hide construction detail makes it clumsy, and the need to Constrain everything is time-consuming. It can be done, but I made an early mistake in this model that takes a lot of backtracking to fix. The top dimension has floated to 0.751" because the sketcher honours an incorrect constraint, and it can't be forced to 0.750" without undoing several steps. Lots of constraint clutter in red already with many lines still to be added:

FreeCAD's sketcher makes a poor general purpose 2D drwaing aid. Of course it's not meant to be used as one! It has an easy to use Spline Tool. The draughtsman just adds a spline curve and shapes it to look pretty. The whole moulding can be drawn in a few minutes provided Old School exactitude isn't demanded.

and seconds later:

One more small step and my 3D printer could be churning a moulding out while I enjoy a glass of wine!

Dave

Edited By SillyOldDuffer on 04/07/2020 15:13:31

Certainly is, and I've awarded myself a 'C-, could do better'.

My version:

Not bad at first glance, but there are a few problems. First my R50 curve doesn't match Gary's, nor does my blend. Differences emerge when our drawings are overlaid:

Thought logical at the time I've not positioned the origin of my 50mm radius curve correctly: it's too high. A second mistake pops out when the junction between R3 and R50 is zoomed into, the curves don't meet.

On a hand drawing, the two curves failing to join wouldn't matter much; a human reading the drawing might not notice the gap or care. Fatal to 3D CAD and CAM though. That tiny gap is a serious error because the geometry described by the broken drawing is impossible. A 3D object can't be created from it. Drawings that look perfect are a common cause of 3D beginner frustration.

Not sure why my 50mm radius is wrong, I need to have another think. Gary can bask in glory while I retreat in confusion...

Dave

Edited By SillyOldDuffer on 05/07/2020 15:56:12

You should not need to place the origin of the 50mm radius. You just need it to be tangental to the angled line and intersect with the end of the 48mm vertical line. once that is do radius the junction.

Edited By JasonB on 05/07/2020 16:27:52

Thanks Jason - I've been overthinking it!

Dave

A new challenge for 2D Draughtsmen is to produce First or Third Angle drawings of the sharp end of this tin-tack:

Dimensions: tack is 10mm in diameter and 30mm long. The tip has 3-facets at 60°.

The hard part of this one is correctly developing the curves where the flat facets meet the round body of the nail.

I think it's an example of an object easier to make in the real world than it is to model in 3D, and considerably quicker to model in 3D than it is to formally draw in 2D. Hardest of all on a drawing board rather than CAD because of the multitude of construction lines that have to be erased to finish the drawing neatly.

Trainee Draughtsmen were set many exercises like this and took years to become fully proficient. Old time engineering required immense skills and patience, it makes their work even more impressive.

Dave

I was never a draughtsman professionally but did take GCE A-Level Technical Drawing, and Interpenetrations of which I think that tack would be an example, were one of the topics covered.

It was not the most difficult topic though - I award that prize to something called "Lines in Space". Indeed, our teacher warned us if we didn't grasp the principles from the start we would be very unlikely to learn it! Despite sounding like an early Pink Floyd track, this method plots the shortest line between two lines crossing one above the other in a 3D space. It is fairly easy to visualise a solid, 3D object shown on a 2D sheet - until you venture into such realms as the innards of a Burrell compound cylinder casting, perhaps - but a line from one (x,y,z) to another (x,y,z) is a pretty nebulous entity. Now add another equally arbitrary 3D-sloping line, then try to imagine where they are at their closest, and how to draw them to establish that link accurately....

I have no idea if CAD could give you the answer. Even if you knew the 4 co-ordinate trios the link-line would be fiendishly difficult to calculate mathematically, if possible at all. I have an idea too, we did not actually have the lines' start and end co-ordinates, but I forget how they were described. You'd need a third line and all crossing at a single point to use simultaneous equations with 3 unknowns, which is difficult and laborious algebra - here we start with only 2 lines unsociably distanced!

'

I don't think the industrial draughts-people many years back would necessarily have had to erase all the construction lines though, because the workshop prints were taken off ink tracings of the pencil drawings. ("Tracer" was our Mam's occupation, in fact!) That erasing would have been necessary in later years when drawing straight onto draughting-film (tracing-paper with a posh name). I worked as materials-cutter for several years for a printing-machine manufacturer, not long before CAD started to develop professionally, and all the original drawings there were on tracing-paper from which the shop prints used by the machinists and I, were photo-copied.

Here's another challenge for you. All transitions are tangents, and the top line is horizontal.

Nigel, I think I did the shortest distance from a point to a plane and between two lines in space in A level maths. Also if I remember correctly the formula of the line of intersection of two planes. So much stuff has been pushed out of memory by time and new stuff to remember.

Martin C

Garry, don't you have anything more taxing as that one is quite easy. Just establish the 3 points draw some concentric circles then a few larger circles and a straight line before going to work with the tangent icon. then just trim away the excess.

The only unknown are the two 9mm circles which being shaded may be a recess or projection the extent of either being unknown.

I think it is called Buffer Overflow, have been blaming that for a few years now !
DaveD

Sorry, no.

CAD simplifies things like this, but it's rather challenging with a manual drawing board.

With construction lines, done in different colours to highlight the intersections.

CAD makes it easy to draw the curve along the construction, but I don't see any way QCAD might automate working out the intersections.

Gary's July challenge is easy in QCAD because the software does tangents at the click of a button, whoo hoo. Much harder with a Drawing Board I think.

Dave

Why? you know the distance from the centres of the smaller circles to the bigger ones, so just strike arcs and draw circles tangential. (I'm so old I have used a spring bow pen and Indian ink to draw on linen sheets. Only to modify old drawings to be sure)

I wouldn't have done it like Jason, just draw the 3 smaller circles and the horizontal line and use the fillet tool

No really, just a case of swinging a few arcs

Fillet won't do it in alibre as you can't select two complete circles

Edited By JasonB on 21/07/2020 17:12:46

Apologies in advance for being a churl, but I've highlighted my learned colleagues drawings in blue were I think the curves look like tangents, but are eyeballed approximations or mistaken.

The mistake, line actually tangent with the lower construction circle rather than the target:

By eye, zooming in suggests the lines don't actually meet:

My painful effort shows the construction lines needed for accuracy. It isn't quite right either - one of my dimensions is bodged to 99.9941 rather than 100 in order to get the curves to meet correctly. Although done in QCAD I only used drawing board methods - circles - to position the curves. Promise!

Although a close approximation can be sketched without too much bother, I still think this is tricky to draw geometrically, ie with tangents accurately at 90° to a radius. I might have made heavy weather of it. Any Old School professional draughtsmen out there?

Dave

Edited By SillyOldDuffer on 21/07/2020 22:32:19