Is CAD for Me?

For an introduction to 3D CAD concepts, try MEW 217 as a starter.

Yes you can learn!

Being very familiar with 2D CAD can be as much of a barrier as a help when starting to use 3D CAD. The order of doing things is very different.

Edited By JasonB on 04/06/2019 16:06:11

Edited By JasonB on 04/06/2019 16:06:20

Just before I retired I bought a CNC Mill in order to give me a challenge. In order to use the mill properly I had to educate myself on how to draw 3-D having had no experience other than a 3 month course in Auto Cad many years ago which I used infrequently over the years.

I also had learn how to use the mill and generate G-codes etc. It took me the best part of two years to get to the stage where I can now draw most components I intend to machine, and to generate the G-code necessary.

I do not see age as a barrier to learning stuff like this, a lot depends on how keen you are to do so and how long you are prepared to persevere.

Edited By JasonB on 04/06/2019 16:06:37

+1 for Paul McWhorter **LINK**. The problem I used to have with various 3D CAD programs was determining where I was in the drawing space. Fusion 360 made that very straightforward and I've gotten by but just watching the first couple of Mr McWhorter's videos has taught me loads about making the whole process easier.

Rod

Edited By JasonB on 04/06/2019 16:06:56

It seems natural to create in 3D as real world parts are inherently 3D. As SoD says there are also big advantages to using 3D models and in particular assemblies. The drawings for my traction engines are poor so I'm modelling parts in 3D and creating assemblies before machining:

I'm lazy and want to make parts that are correct, and fit together first time.

There seems to be a hang up about creating isometric views. I'd agree they're difficult to draw from scratch. But 3D packages should automate the production of 2D drawings, including isometric views. I always add an isometric view in the top left of my 2D drawings as a sanity check to check the relative orientation of features. With a 3D package it is also simple to add sections to a 2D drawing. No way I'd ever go back to 2D software.

Of course for very simple parts I just do a paper sketch of the part with key dimensions and don't worry about modelling. Before using 3D CAD I normally "design" the part in my head, possibly with some paper sketches as well.

Andrew

Edited By JasonB on 04/06/2019 16:07:12

Thank you - some interesting points there.

I'll explain the Why etc:

My primary but not sole project is a 4"-scale steam-wagon built only from a few surviving trade reviews and their photographs, so entailing a lot of designing, trial and error, unforeseeable problems and so on. From peripheral contact with CAD at work, and its increasing use in Model Engineer, I realised if I could understand it to a reasonable level its advantages potentially mean it being very useful to me.

'

The advantages I saw even before trying it for myself were in:

1 Generating the drawing's elements (geometrical figures) rapidly and accurately,

2 Ready editing - e.g. to alter diameters or fastenings.

3 Ready copying of one element as where necessary, thus avoiding having to draw each individually.

4 Automatic dimension calculating, with the ease of working in 'thous' rather than vulgar fractions, to suit the machine-tools. Or indeed, in millimetres, though at the time all my machines used real inches. I didn't yet know dimensioning can be fraught with its own difficulties.

Additionally, I knew 3D drawing (I didn't yet know is called " modelling " would allow the three first above, plus

5 Ready viewing of the subject from all sides

6 Building assemblies to verify e.g. clearances, mounting-points, pipe-runs, controls.

I even thought one day I could be rid of the massive A0 drawing-board dominating my dining-room!

I am familiar with manual drawing, but I realised CAD introduces a vastly greater set of its own skills between idea and printed drawing.

'

So I did not go into buying TurboCAD blind. My employer's drawing-office uses Solidworks but I gambled on sufficient similarity, and all these packages do essentially the same things for my more modest home-workshop needs, even if their controls and tools differ markedly.

I knew it would be a lot to learn; but not how hard it is.

'

Of points raised by others:

Age? I don't see that as a problem, assuming undiminished memory and mental capacity. Ability to learn is what counts. A friend gained a PhD in a geological subject as a retirement hobby - I had the pleasure of participating in many of his field-work expeditions to Norway. Yes, he is far brighter and more able to learn than me, but he's also a decade older than me and I am not aiming even for a degree.

'

Need? All right perhaps want more than need, for a hobby, but I saw CAD as a potentially valuable adjunct, or tool, to my hobby's real aim - creating real things.

'

View of CAD, as SillyOldDuffer describes? None of us are mind-readers! Apart from my preferring you talk to than about me when discussing how I think, I don't agree with your analysis anyway. For: -

I formed and retain a clear idea of wants, as above.

I knew you need think of CAD very differently from manual draughting.

I knew I don't need CAM files or brochure-quality pictures; though careful colouring can clarify an assembly-drawing: even 19C draughtsmen used tints and textures.

I realised 3D modelling manipulates objects, but they are still pictorial representations of geometrical figures, so that does not " blow my fuse ", thank you! I find it easy to see them as isometric even if they are not strictly so.

I don't think TurboCAD allows animation, but that does not worry me.

'

My real problem is the lack of any clear, printed literature on a) CAD concepts generally, and b) specific to the individual make of CAD. I started using computers, at work, at about the change from MS-DOS to MS WIN3, learnt some very basic BASIC, and have used 'Word' and 'Excel' extensively since, both professionally as a lab assistant, and supporting my hobbies of model-engineering, caving and geology. I wrote a book mss. on an Amstrad PCW. So though I find advanced software difficult to learn, I am no novice! However, all those were backed by genuine manuals. CAD is not: its publishers rely on formal teaching (unavailable to me) and videos (available but not helpful).

I did help myself greatly though, by indexing TurboCAD's God-awful pdf " manual " properly. I discovered that very unusually for a pdf file, I could copy its contents page and via 'Word', form a 2-column 'Excel' version sorted alphabetically. So I can use its print to select the page in the still-on-line original, which you can't search by words.

+++

Finally, on unwanted icons, I did not know the thing was there until after posting, and there seems no way on this forum to edit a posted comment.

Edited By JasonB on 04/06/2019 16:07:33

Andrew Johnston:

Your point and its sample drawing didn't appear until I'd posted the above.

I wonder though, and have thought this of other contributors who swear by what I see as the long way round ... did you start learning engineering-drawing with isometric-first CAD?

If so, I understand how you'd visualise your designs that way.

I didn't, so have no problems with orthographic drawings and maps. Further, I see viewing a solid item one face at a time as just as natural as seeing it in 3D. We do that all the time, knowing the sides recede from view somewhere even if we can't see them. Viewed across the room, my long-case clock is obviously 3D because I see it at an angle; but face-on, the crowded shelves of books form a two-, not three-, dimensional image.

Consequently, is needing a 3D model to ensure components fit together without fouling others, an instinctive result of being taught to work what I think the long and hard way round? You can still assess fits etc. in orthographic drawings. It's correct shapes and dimensions that matter, not views!

Converting a 3D model to the necessary 2D workshop drawings means advanced knowledge of your particular brand of software. It's probably there in TurboCAD but like forming 3D assemblies, hidden. I can't speak for Fusion, Alibre etc but TC's maze of solid-generation types and snap methods, co-ordinate systems and work-plane types, all of different properties; makes 3D assembly-drawing a baffling mass of unspecified combinations of which only one will work in each situation.

So it's really a matter of how you've been taught, or have taught yourself, technical drawing. If you have been introduced to 3D CAD by someone who can explain it properly, or can learn this extremely difficult subject yourself, and right from the start, then I can understand you'd work that way.

If you are used to seeing 3D items in 2D, you have not missed much if you cannot learn 3D modelling. It is disappointing intellectually to spend hours on the challenge only to find it impossible; but it's not essential to putting on paper a design whose physical reality will fit together and work as intended.

You can't though, work backwards, turning a multi-part orthographic drawing into an isometric assembly. CAD is not designed that way.

Edited By JasonB on 04/06/2019 16:07:51

Quoting from Nigel above...

You can still assess fits etc. in orthographic drawings. It's correct shapes and dimensions that matter, not views!

You can - but not in a single click with clashes higlighted automatically. It's also easy in 3D to check for interference in multiple positions of a mechanism.

Converting a 3D model to the necessary 2D workshop drawings means advanced knowledge of your particular brand of software.

In most 3D CAD, to get from 3D model to 2D drawings takes just a few mouse clicks (hardly 'advanced'. Making sure all dimensions are arranged nicely and to preferred format can take much longer, I agree.

You can't though, work backwards, turning a multi-part orthographic drawing into an isometric assembly. CAD is not designed that way.

You can! It isn't preferred, nor best practice, but it's perfectly possible (except that assemblies are not isometric - views can be isometric). I will sometimes use this approach if someone has already created 2D CAD drawings, but a 3D model is needed for other purposes (though it can be faster in some cases to start again from scratch).

Don't worry Nigel - I know you won't be persuaded. There's nothing wrong at all with sticking to 2D should you prefer that.

The odd thing is that most 3D CAD modelling starts with 2D sketches anyway.

Edited By JasonB on 04/06/2019 16:08:13

That damned winky again ! Only noticed now it's too late to edit.

Edited By JasonB on 04/06/2019 16:08:35

Sorry Nigel but you weren't my target audience. My comments were aimed at anyone reading your well-written CAD-negative posts who might be accepting them as valid insights. If your observations were correct then 3D CAD would be impossible to learn without professional training. That's not true.

I'm certain you're not daft or stupid - just the opposite. I reckon you're a clever chap who's been super-glued to the wrong end of the stick by circumstances. I think you're mistaken about why you happen to find learning CAD so difficult and I'm concerned your critique will put people off who would otherwise have a go themselves.

The Alibre offer is a glorious opportunity if you want to learn CAD modelling and need assistance. It's supported by a series in a print magazine that you can study. Any questions you have can be put to an Alibre expert on the forum. There are more than a few forum members who can help as well. Everything else you've tried doing to get into CAD has failed. Why not try a new approach?

Dave

Nigel, you have the procedure back to front: the 3d models are not pretty afterthoughts, but are what enable you to produce renderings, analysis or 2d drawings. To use 3d CAD, you have to get past the 2d drawing first mindset.

I think this is what makes it easier for those of us who don't have 2d training. Although I can read a 2d drawing, learning how to make them is not a good use of my time when 3d is more intuitive for ME.

+1

Thank you Jason, for the move. I admit it was rather hogging things. The Tea-room would need a Carnforth-size urn and whole collection of Rachmaninov!

+++

Thank you Dave for the clarification.

Thank you too for your compliment - but I was always a slow learner in anything, up to random levels I cannot exceed.

I've just sent the geological article and its 3 TC-drawn diagrams for peer-review, so some good's come of it! One is a simple 2D cross-section. The other two are 3D extruded sections, coloured even; but might be printed in grey-scale at the editor's judgement.

My other 3D TurboCAD efforts are all engineering-related, but not useable. Most are pure exercises but include an unfinished design for an (x, y, angle) jig-table for the bench-drill, of far wider range but lower profile than those cross-vices.

'

I take your point about a new start, but I stayed with TurboCAD not just by price. Alibre is so different I thought it illogical to start all over again after managing to overcome TC sufficiently to produce fairly simple, if rather rough, orthographic drawings I can use.

Also, Alibre looked no easier than TurboCAD despite its support article in MEW; though it seemed friendlier than Fusion, which I'd tried cold and without any support.

I went from manual draughting in the 60’s to 2D CAD in the 80’s and finally 3D CAD in the 90’s. I found the best way to produce a 3D model was to start with a block and take shapes away, as you would machine a piece. In the real world I found that modelling from nothing, adding shapes together often created difficult or impossible machining.

+2

Andrew

I'm afraid I don't understand what isometric-first CAD means? Nor do I understand why isometric drawings seem to be promiment in this thread? To me an isometric projection is a nice to have on a 2D drawing, but is by no means essential.

I'm not a professional draughtsman and am largely self-taught. As a kid I started creating 2D technical drawings at home beginning with various widths of propelling pencils and then moving on to Rotring pens (birthday present). All drawings were done on a small drawing board (another birthday present). No computers in sight; the concept of a PC didn't even exist. During my post-school training, and at university, we did some technical drawing, but all 2D and all by hand on a drawing board.

When I started working in motor racing I bought my own simple 2D CAD package that ran under DOS. Can't remember what it was called, but I rang rings around the draughtsman at work who was still using pencil and paper.

Later, when self-employed, I bought a cut down version of 2D AutoCAD. It was ok but not really anything more than an electronic drawing board. In order to design tightly packed electronics I needed something much more capable. That was when I switched to 3D CAD. I didn't really have a problem learning it, at least for the basics. I've never had a problem knowing what I want to design, but it can sometimes be an issue getting the CAD package to create the model the way I want it.

While I use isometric views on 2D drawings generated by the 3D package I don't remember ever using them when drawing by hand or with 2D packages.

I'd agree with Nicholas, using 3D CAD is natural, not a long way round. Parts are inherently 3D so it makes sense to create models in 3D and let the computer do the grunt work of creating the projections needed for a 2D drawing.

Andrew

Another + from me re forgetting about a 2D mind set when doing 3D CAD. The trick is to think "I want to create a part" rather than "where do I have to put that line to make sure I can project the next line bla bla bla....". For 3D If you want you can model a big block and carve it away with cut features just as you would machine it. Many other ways to start a model too. If you want to see 3D in action there are thousands of youtube video tutorials about getting started all the way to advanced techniques. I started on the draughting board but also trained in CAD at the same time, starting in 1981, the pioneer days of it. Very happy Solidworks user these days after many years of ProEngineer CAD.

I struggled with TurboCAD for ages, Alibre was a breath of fresh air until they massively increased the price.

Fusion is currently free, more capable, and mostly easier to use.

One of the best things they did was to name the first thing you do to model a part a sketch. This suggests that it's just a step, not a finished item.

Then there's the top down approach to assemblies; it makes a great deal of sense to create a piston in the bore it's going to run in, than on a completely separate sheet.

As you work with the program, there are lots of useful features that save a great deal of time: using an existing part to create the matching features on an ajoining one; sectional views wherever you want them; moving parts to develop linkages etc etc

I had no training in this, besides watching some of the videos and reading forums

Edited By Nicholas Wheeler 1 on 04/06/2019 22:21:50

I can see you find it natural to draw something in 3D because it is a 3D object, but I rarely have much difficulty seeing a 2D representation of it.

Maybe it's what we are used to, but no-one yet has really explained why necessarily-orthogonal workshop drawings have to me made from a pictorial original. Especially as the 3D relies heavily on extruded 2D figures.

Instead, they all say what they prefer, have been taught, or what the CAD publishers intend.

I see 3D CAD's advantages for very complex shapes or assemblies, for the expert; but I still do not know why it is intrinsically wrong to bypass the process.

'

For example, a simple hollow cylinder like a bearing bush.

In 2D (any method) it's two concentric circles and a rectangle.

In manual isometric it is 3 or 4 plotted ellipses and two lines.

In CAD isometric, the computer plots one of several complicated forms: umpteen-sided polygons of individual facets that result from extruding circles, stock "primitive" cylinders modified to size, a solid cylinder subtracted from another... All with specific properties and reactions to further operations; no offered clue which to select. You are assumed to understand it from the start.

By the time I've drawn it in 3D I could have drawn it in 2D and (if I'd managed to print it), made it!

'

Experts naturally find hard things easy; and assume it is easy for anyone. How and why? Why is my approach, my natural and only option, wrong?

In 3d CAD it's two concentric circles and an extrude; quicker than it took to type that sentence. And I would have 'drawn' it in the hole it fits, using those dimensions for both the diameter and length so that any changes, that might be necessary later, cascade through all of the parts.

I don't care how the program does it, in the same way that I don't care how the graphite got in the pencil.

And once again, what the hell is 3d isometric?

You only have to learn a technique once. Then it becomes a massive time and effort saver.

Edited By Nicholas Wheeler 1 on 04/06/2019 22:56:12

Edited By Nicholas Wheeler 1 on 04/06/2019 22:57:43

Anyone here got any experience of Linux CAD packages - FreeCAD for example? I've installed it, and some of the supporting literature, but it looks completely baffling (where are the nursery slopes?). FreeCAD is also available for Windoze and Muck operating systems.