With thanks to those who helped me solve why I could not copy areas of one drawing to another, and the source drawing also freezing, I have a more detailed question.

The same [Not Responding] problem arose briefly yesterday, with my large general-arrangement drawing.

It has several layers in different colours, but overall is only 4 orthographic elevations with little detail. I have been careful not to create the hidden stacks that crippled the other drawing.

It is not a rendered 3D picture – mathematically I'd have thought my wagon GA simpler than Jason's 3D image of his engine irrespective of different make of software; but it may hold more elements.

'

So my question…

Is there a definite limit to the size and complexity of a drawing that TurboCAD and the computer can handle?

Or is the problem caused by keeping all the Layers visible when working on any one part of the drawing?

Or something else again?

(If just drawing size, I don't understand how it copes with producing the beautiful pictures in the TC Users' Forum.)

'

= = =

Incidentally, I said I would find making a 3D image of the wagon, impossible.

Well, that's not just a guess. I did try it.

I drew the two long chassis rails, parallel, but without the cross-section tapers of their hot-rolled steel material.

Next, the three cross-members shortened from copies of the above.

Plus one temporarily copied to the front of the chassis for locating further components. That was to simplify the co-ordinate calculations as the image's screen co-ordinates are random, and the 3D sizing and moving tools often have unexpected and sometimes irreversible effects.

Then the two short, parallel channels for the narrowed front section.

That was it!

My barrier was the converging section linking the two parallel sections.

The physical, full lengths of the channels were fairly easily shaped using a rail-bender to chalk-lines on the floor, and tape-measure. The drawing's convergent parts lie at angles to the X and Y axes and though very easy to draw manually, I found them too difficult in CAD.

Thank you.

I spent some time this morning doing a little housekeeping on the computer, but whether it will make much difference, I don't know.

It hadn't occurred to me the computer had been taken over in the background by some automatic application or other.

I'll finish and post this then see what Task Manager says is happening. It shouldn't be much especially if I minimise this screen.

Jason –

I don't recollect previous problems as severe as they were this time. TurboCAD sometimes does need to pause and think about what I've asked it to do, but usually for only a second or two, not ten minutes just to let me re-size a selected object.

Looking at your drawing there, I think I can see how I can do that: draw the plan view of the vertical web with its two bends, and turn it into a "solid" wall (an Extrusion). Then do the same with the flange, extruding that but by only 5mm, copy these figures as necessary and add them together. I recall previously making "angle-iron" by joining two extrusions edge to edge then "Add" – ing them. Not sure how I would cut the front ends back to that long bevel though.

Strange – Despite that "house-keeping" I mentioned above I still had what should have been a straightforward drawing freeze, my vehicle's GA, for a worrying but short time. It is very easy in TurboCAD to produce all manner of hidden, surplus fragments so there might be some in there.

By experiment I ran Task Manager's performance graphs while opening that GA, probably my largest drawing.

There were expectedly two big CPU peaks as the programme then drawing opened, but it registered very low memory use, and when I highlighted an object and moved it, the graphs barely flickered.

I thought the size of the drawing – 4 orthographc elevations – might be too big but it seems <400kB.

Jason, Nicholas –

Thank you for the hints, but I'm afraid I cannot produce what you say is easy, that pair of cranked chassis rails. I could draw them as Extruded sections if they were straight and parallel to an axis, but not at angles.

Jason –

I think you are using Alibre and perhaps non-axial figures are easier in that. You are right – the rails are bent only in plan (I hope the real ones are, too!).

'

Nicholas –

What did you use there?

TurboCAD Deluxe 19 has no "sweep" command, according to its manual. The TC Users' Forum sometimes refers to "rail sweep", but I don't know what that does.

I don't think it allows dimensioning in full 3D as you have, only on elevation views. TC has associative dimensions that follow object changes, but I do not know if reciprocally.

The exercise proved I cannot draw in TC complicated items like those cranked channels from extrusions of the cross-section. Instead, I would have to work from the component's plan. The real metal is 50 X 25 X 5mm, but I used inches to suit the model itself, rounded to 2 X 1 X 0.2 inches. SO:

– 1) A thickness-extruded, 8-sided Polyline; full length, 1" wide X 0.2" deep.

– 2) A copy moved to one side, dropped it 1.8" (the Reference Points are in mid-metal). These are the flanges

– 3) A second copy, moved aside, Exploded edited and re-united as a Polyline 0.2" wide X 1.6" deep (the web).

– 4) Assemble the flanges to the web.

– 5) Copy the whole caboodle, slide the copy away, invert it, calculate from a plan-view dimension the spacing adjustment.

– 6) Calculate moving both rails to put the nearest corner as a datum on round-numbered co-ordinates: probably entailing some geometrical construction, because the combined Reference Point is somewhere in the middle..

This omits the flange tapers and roots as merely needless complexity. Even so, it proved too difficult and I realised I cannot draw the rest of the vehicle, not in 3D.

'

Forming that long bevel at the front needed:

– Extruding a long bar from a triangle, swinging it horizontal, Assembling it to the rails.

– Copying it, moving the copy aside.

– Subtracting the first prism from one rail – this also deletes the prism.

– Returning the copy for the second rail's subtraction.

(TurboCAD's version of "Cut" is "Object Trim", working only in 2D.)

Similarly I don't think it's possible to Extrude a solid to another, only extend lines to them. Normally to ensure things meet in 2D I draw them over-length and Object Trim them back. In 3D I need calculate their right lengths.

I calculate co-ordinates and relative moves from points on the drawings because I cannot understand the extraordinarily complicated Work-plane and Co-ordinate systems. I don't know if this complexity is just in TurboCAD or 3D CAD generally.

'

Sorry, but the concepts and methods you show are new to me, don't apply to the CAD version I have, and are more advanced than my level. I used to think I could learn 3D CAD!

I am impressed, Nicholas – and envy your skill! – but I do know TurboCAD lets you create similar models if you know how. I don't know though if my edition at least, gives 3D dimensioning.

I would not care trying to machine those parts from those model drawings, though, even with each part on a separate sheet – as industrial practice. They would be are far harder to read with the certainty of identifying the dimensions correctly. You know where they are only because you drew it as well. The isometric image is useful in the corner of the sheet, but the orthogonal elevation is far easier for making the thing.

The adjective "traditional" is meaningless, especially if you mean it derisively as appears. There is little in mechanical-engineering tools and methods that is not "traditional" in some form or other, because notwithstanding technical developments and innovations, the basics work! The principle of the lathe appears as old as Ancient Egypt, so is the lathe "traditional" therefore unworthy?

'

I may be wrong but I am sure I have seen references to animation in the Fusion users' forum.

'

I did not say my methods are as IMSI intends – I know they are not the proper, efficient TurboCAD way!

TurboCAD does not expect you to have calculate anything, but I don't know how to use it to avoid the sums.

TurboCAD is perfectly capable of drawings like your engine, and just as efficiently; but I don't know it well enough to let it do that for me.

I briefly tried Fusion, alone, and Alibre via the first couple of instalments in the MEW series. I bought a subscription to MEW, already subscribing to ME, but missed two editions so that stopped progress; then I realised changing from one CAD package to another even though comparable, would be a mistake.

'

Those other makes must have equivalents to TurboCAD's 'Work-planes', under a different name and perhaps more direct. Your crankshaft drawing bears that out. I think early in that MEW Alibre series, it highlighted a surface to develop a shape on it – that is one among several types of what TC calls a 'work-plane'.

'Viewport' is only TurboCAD's method of transferring the raw drawing to the printer page-size, but all CAD systems must have similar, under different names, perhaps simpler. They need to, for setting images onto any of a huge variety of both American and international-standard sizes. (I think there must be two dozen in the printer-setting menus.)

'

There is though a fundamental point beyond how well I can learn such systems.

TurboCAD lets you choose between orthographic elevations or isometric art-work. The former matter in the workshop, and I expect TC can develop them from 3D models but that to me is the very long way round. Isometric drawings have always had a place in engineering-drawing though, if not in the machine-shop then for, e.g., illustrating instruction-manuals. To me, draughting either manual or CAD is an engineering tool, but 3D CAD seems presented as more of an academic art-exercise. Fusion and Alibre appear not to give that choice so drawing a simple object to be made physically, is subsumed into a long and hard exercise that prioritises the drawing over the object.

It's like my lathe. I don't own it to admire it and swarf. I have it for making objects that can really only be made properly by turning. Hence, I bought a CAD package to help me towards making physical objects, not beautiful 3D pictures of them; but I did want the isometric function as well, as potentially useful occasionally. I was simply not prepared for just how difficult it is to make 3D CAD pictures.

'

I looked again last night at that 3D outline of my wagon; but by the only methods I know, knowing them not as TurboCAD is best used..

I managed the chassis frame, spaced other parts along it from a dummy cross-bar on the front, gave it rather Flintstone-esque wheels. I progressed to the bunkers but a mistake in hollowing them from their generating solids, removed their floors. The floor is a long plate extending back as the footplate and seat mounting; but I was totally unable to return to the method that worked for the bunker walls, to generate it. (It is a difficult shape, narrowing forwards to a corner with a large radius – but a possible solution has just occurred to me.)

At least I don't have a telly to distract me!