Alibre – A First Attempt

[Nigel Graham 2Participant @nigelgraham2]

Thank you for your encouraging comments!

David –

I recall the first exercise I tried did use the anchor tool, but I don't think the MEW one (the scribing-block does. Unless I failed to spot it. It still locked the clamp in place on the screen though.

Ady –

You can set your own "homework" yes, but with care not to be too ambitious. Both of my attempts used moves I'd already seen in the set exercises. Though having made something simple, like a pipe-flange or a flywheel, say, you can at least experiment with it.

[David JuppParticipant @davidjupp51506]

Nigel,

The Anchor is a quick way to fix a part in space, but it gives little control. Typically all you can do to define the position is to deliberately insert at the workspace origin, or at some other inserted point. Orientation will be whatever it was when the part was first modelled.

The MEW exercise showed the alternative of showing the reference geometry of an inserted part, then constraining that to the reference geometry of the assembly workspace. This takes longer, but gives full control – you can alter orientation of the inserted part, position it wherever you wish, set it at an angle….

Both methods are valid, and for the first part in an assembly are largely interchangeable, unless you want to set a particular orientation for the part.

I only brought it up because you'd mentioned problems constraining, but were not very clear on exactly what. Failure to fix the first part in an assembly is quite a common issue for beginners and can give really weird behaviours.

[JasonBModerator @jasonb]

Posted by Nigel Graham 2 on 01/05/2023 13:42:24:

The text also refers to a 'Flip' command under Assembly Constraints, to turn individual parts the right way up, but wherever it really is, is not there!

Good progress

The flip command was there a couple of days ago as I pointed it out to you and you used it. You may have it layed out slightly different in Atom but same two red/blue arrows

[Nigel Graham 2Participant @nigelgraham2]

Well, I certainly make Alibre Atom behave in very weird ways instead of what I'm led to expect. When that happens, all I can do is scrap the whole drawing so far and start all over again without having the slightest idea why I failed. Which is very off-putting if well into a complicated Part.

Following instructions is all very well provided you can make your results match theirs, but the text's author assumes you can understand the instructions and their effects fully, and never make mistakes.

So even if you manage to copy the image as intended, you still do not really know why, nor what can go wrong and how to put it right. If you don't, you have no idea what you have done wrong, especially if the result is an error-message written in computer-speak.

It's like trying to learn turning from the tool-geometry diagrams in a Sandvik catalogue.

[JasonBModerator @jasonb]

Why scrap the whole thing and start again, only a couple of days ago the UNDO tabs were mentioned, if it goes wrong first thing to do is hit that UNDO tab and you just go back one step.

on an assembly you also have close rather than accept if the part move where you did not want it to and in 2D sketching you have OK or Cancel on a lot of the boxes, click cancel if it is not looking right

This is why DavidJ's offer should be taken up as he will talk you through things if they don't come up as expected, he will understand those error messages

Edited By JasonB on 02/05/2023 09:55:50

[David JuppParticipant @davidjupp51506]

Also your saved files (with the 'errors&#39 may well point to exactly what went wrong. Much more instructive to talk through a file that exhibits some issue, and look at what may have happened / how to fix it. You'll learn more than from getting it right at a first attempt.

I've looked at the MEW exercise again just yesterday. At some steps, I had to stop and read it again carefully as the wording perhaps isn't what I'd have used myself (though my wording may have been just as ambiguous in its own way). I sometimes had to turn the workspace slightly to be able to select and edge. I sometimes found that I had to hide a part to be able to see something that the text told me to constrain in an assembly.

So yes there are potential barriers, which can be very frustrating if you don't realise how to get around them.

In addition to Jason's comment about using Undo, you can also delete or suppress individual features in the Design Explorer (on left side of workspace). You can also go back and edit sketches or features you previously completed.

[Nigel Graham 2Participant @nigelgraham2]

I know the purpose of the Undo / Redo arrows – as in almost any Windows-hosted software – but I'd met situations where I seem to have gone too far for them. I don't know if they have a maximum number of steps by the application or the electronics, but I'd made mistakes whose effects only appeared much later.

The Flip arrows on the assembly menu did nothing on the one occasion I tried them, so I must have been using them wrongly.

Though I did discover the Assembly tools themselves can rotate a part to the right orientation when being added to the fixed one, with a rather wonderful visual effect as it operates. At least, right for the mathematics: it is aligning surfaces and does not know if I have made them the right size and the right way round. A mistake there makes the added part sink into the host, not lie on it….. I found.

I have found how to correct simple drawing errors like a wrong size, rather than Alibre-operating mistakes as such; by deleting the appropriate Sketch and making a new one. I did that with the scribing-block to "re-drill" the tapered hole in the base to parallel, after finding I cannot taper the column.

[JasonBModerator @jasonb]

Think of assembly much like having a box with all your parts rattling about inside. As you take them out the box they could be any way up. It is not until you assemble them that shafts go into holes and faces mate. CAD assembly is no different.

[David JuppParticipant @davidjupp51506]

Nigel,

If something in a sketch is wrong size, don't delete the sketch – just edit the dimension that is incorrect.

One of the biggest advantages of 3D CAD is how easy it is to make changes with throwing work away.

You realise after completion of model and having made your 2D drawing, that (for example) a hole should have been a different size. Edit the dimension in the sketch that defined that hole, exit the sketch, regenerate part to last feature and Save.

Next time you open the 2D drawing of the part, it will update automatically to the new hole size.

[Ady1Participant @ady1]

Sometime you have to "force" an item into its position by using the constraints in two stages

First stage you get it into a basic ok position

Then you DELETE those constraints and it will stay in its new position/direction

Then apply the constraints you wanted originally

It can be a bit fiddly, but it works eventually

Just another skill you have to master Nigel, practice practice practice

[David JuppParticipant @davidjupp51506]

Ady,

Much simpler to hold down the Shift key whilst dragging the part in assembly – that will roll/rotate the part freely in the assembly. Combined with a standard drag to position it, you should be able to get the part close enough that applying assembly constraints becomes straightforward.

Sometimes the order you apply constraints can be important (that is often down to experience or trial & error) – occasionally I have to temporarily suppress one constraint to allow the application of another, then un-suppress it again.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Nigel Graham 2 on 02/05/2023 11:16:44:

…

I have found how to correct simple drawing errors like a wrong size, rather than Alibre-operating mistakes as such; by deleting the appropriate Sketch and making a new one. …

Well, as with using a sledgehammer to crack a nut, that might be dangerous. I can't speak for Alibre, which I've never used, but the usual procedure is to edit sketches, not to delete them.

By analogy, say we're building a house and a crack appears in the attic. The mistake is traced to a pillar in the cellar, designed and installed much earlier. In a model, the pillar can be moved and resized within reason by altering the sketch and this won't disturb whatever else in the model depends on it.

Deleting an early sketch is the CAD equivalent of blasting out a real pillar with dynamite. Might work, but there's a high risk the whole structure will be compromised, with further work failing unexpectedly because of internal faults caused by not sorting the pillar out safely.

Better to learn how to edit sketches than torpedo them.

Dave

[Nigel Graham 2Participant @nigelgraham2]

Thankyou – I can see the point there. It may be just as applicable to Alibre as any other CAD system of similar type.

In the first exercise I tried, I missed an important step so when the thing would not assemble properly. The Parts were legitimate in computing ways but I had put two features in the wrong place; one by a small dimension error, the other by the missed step.

Trying to correct that, I deleted these features and redraw them because, if I understood the error messages aright, I was trying to overlap the new sketches with the original ones. I don't know, certainly didn't then, if you can simply move them.

……

That building analogy has its own for me!

I spent an hour this evening modifying one of my steam-wagon's tank fittings, already soldered in place, so I can raise the tank for better clearance above the rear axle for the water-pipe going forwards. I had made the tank, oh, I don't know, maybe 15 years ago….

[lee websterParticipant @leewebster72680]

Do I sense a lightbulb moment? It sounds as though you are getting to grips with Alibre.

A friend and I nearly bought a full size road going steam lorry. He had an Austin 10, Mk2 Jag, 1948 Bedford 2 or 3 tonner, not sure which, and a 1938 Morris commercial. I had my Austin 7. We never did buy the lorry, but my near neighbour here in Cornwall has a couple os steam traction engines. He took one to Trevithick day a few days ago. I heard it chugging up the road on his return.

Keep us informed please Nigel. Your progress is very interesting.

[Nigel Graham 2Participant @nigelgraham2]

How on Earth did the designers of those fine old vehicles manage with loose Tee-Squares, 2H pencils and a book of log. tables?

Light-bulb? I'm afraid the bulb's blown…..

I have just spent some two hours going back over that scribing-block exercise to work out why and where the Clamp ends up on its side, making the Assembly process even harder. So I set out to re-draw that one part from new. It's the orange block in this picture, with the hole and keyhole-shaped slot; and is on its side there: the column goes through the keyhole.

I followed the words very carefully but it all went to pot, even worse than first time.

I had enormous problems with various tools' selections taking many clicks to respond. The Escape key often took several presses to clear, and the Undo arrow suggested big stacks of hidden moves by taking many presses to turn to grey.

.

To cut the slot you extrude a sketched rectangle through it. That bit was easy but the Rectangle tool just would not respond at all. I had to construct it by separate lines away off the block itself then struggle to chivvy it into the right place without knowing how to move sketch objects properly.

Eventually I managed it, though somehow missed radiusing the outer ends. The photo is of my first attempt, not this second one.

'

Then tried to Assemble it as the exercise text tells you. It just went haywire.

.

Could the mouse I've just bought be faulty? I don't think it is. A dodgy switch might account for tools not responding to being clicked on, but other mouse moves work. Nor would it account for the Escape key not always responding immediately. I could test this by reverting to the previous device, though it has no proper scroll facility.

'

Progress? Interesting? One way to describe it….

Aye – Alibre is easier than other CAD makes I've tried, and I was making "progress" at first; 2 steps forwards, 1 back.

It's now 2 forwards, 3 back.

Any complicated software is a long uphill struggle for me; but I expected the footpath to reach the summit eventually, not fade away in the foot of the scree-slope.

[Grindstone CowboyParticipant @grindstonecowboy]

I don't think you've mentioned the spec of your computer, but from what you say about slow responses et al, it would seem to be short on processing power, RAM or both. Or you have an awful lot of stuff running in the background (intentionally or not) which is using resources.

Rob

[David JuppParticipant @davidjupp51506]

Nigel,

No need to re-model those parts, they look fine. A part being 'on its side' is not a big issue, you can still apply constraints that will position it 'correctly'.

Either there is something amiss on your system, or there's some small step you are missing out or not quite grasping.

Offer of help still stands – have you read my comments about how to rotate/roll parts in assembly space ?

I've checked your computer spec – it isn't great, but should cope with models like this. If it is being sluggish to update screen, try turning off Ambient Occlusion on the View tab of the Ribbon in 3D workspaces – that will significantly offload the demand on the graphics adapter. There are other settings that can be tweaked to help further if needed.

[lee websterParticipant @leewebster72680]

When things get "slow to respond" as it were, I run the CCleaner (free) software that will remove gremlins. A few weeks back, I noticed a significant reduction in speed and ran CCleaner. It found over 20,000 trackers. The computer was a lot faster after removal. The delay you describe sounds like either the computer spec, or trackers. My cad computer runs things just fine. It isn't high spec, but it has a separate graphics card, and no internet connection. I installed the latest version of Solidedge on the computer I am using now. An old Win10 all-in-one Acer with 8gb of mem and no dedicated graphics card. Running SE or Designspark is slow. I could make a cup of between clicking the mouse and it doing something. A little HP laptop I was given last year also runs Win10, has no graphics card, but it does have a solid state drive. So, much, much faster than the poor Acer, but not as fast as my cad computer which also has a SSD with Win11. I don't think the version of Windows matters too much, so long as it is 64 bit.

A bit off topic but refering to your comment about how things were designed in years past. I know the man who restores and rebuilds the ancient machinery for The National Trust here in Cornwall. When I went to his workshop last year he showed me what I will call a con-rod in my ignorance of correct steam engine terminology. It's the rod that runs alongside the actual con-rod and I believe it fits on an eccentric big-end wise. The rod was made in the late 1700s, and the little end had been hammered out, bent over a mandrill to form the eye, and then hammer welded back onto the main rod. The were no visible signs of any of this work. And the engineer restoring the rod, and others who had examined it, didn't think it could be replicated today. A skill long gone.

[Nigel Graham 2Participant @nigelgraham2]

Most likely I'm just not grasping most steps, not just some small one.

This Forum and the two magazines shows so many others have tried 3D CAD have mastered it to advanced levels, and probably quite rapidly; regarding it as simply a "machine tool" like their lathes and milling-machines.

I knew from the start several years ago now, I would take a long time to learn it to anything useful for model-engineering level; simply by my naturally low learning speeds and capacities, specific to subject.

Eventually I doubted I would reach more than a basic level, perhaps rough images of very simple, single parts.

I now doubt I can learn it at all.

'

I would be surprised if my computer cannot handle Alibre Atom as it can run the far more complicated TurboCAD, which has at least two mathematically different classes of "solid" and very powerful rendering routines. Though I noticed that in Alibre, closing the file, and re-opening it, sometimes cleared the jam.

I wonder if I am jamming it by repeated button-presses stacking up a lot of hidden data. Or does the trial edition and its exercises run on-line so susceptible to broadband speed variations?

The PC is a Dell, factory rebuilt Desktop.

2.3GHz processor, x64

8Gb RAM (7.77 useable)

[David JuppParticipant @davidjupp51506]

Nigel – if you'd only accept some help we could sort out what the issues are. At least you'd then be in a better position to decide if you want to go any further…

As I've already said, your computer spec is not great on the graphics side, but should cope with these simple models (disabling Ambient Occlusion, Ground Shadow, Ground Reflection, Model Shadow reduces significantly the work the graphics adapter has to do and hence will make things smoother).

[Ady1Participant @ady1]

Posted by Nigel Graham 2 on 03/05/2023 10:09:38:

Though I noticed that in Alibre, closing the file, and re-opening it, sometimes cleared the jam.

I wonder if I am jamming it by repeated button-presses stacking up a lot of hidden data.

Going back and forward causes issues and I do what you do, save the file and reopen it to clear the cache

[David JuppParticipant @davidjupp51506]

Nigel,

The trial version does NOT run 'on-line'. All Alibre software is installed locally and runs locally. The only 'on-line' aspect is a check of licence status – this also feeds back some basic information about your installed version of the software and your computer hardware (which can be very helpful in diagnosing problems).

If you don't believe that, simply disconnect your computer from the internet – Atom3D will still run.

[blowlampParticipant @blowlamp]

Nigel.

If you won't accept help, then you are beyond help.

At this point, what do you suggest that someone could say to you, that will help you move forward?

Martin.

[JasonBModerator @jasonb]

Nigel as I said yesterday the way items first appear on the assembly screen has little effect on how they are assembled.

You can see here that the column was on its side but as soon as I applied a constraint to the hole in the base and the column it jumped to the right way up.

Same with the clamps, I purposely made two files one with it the right way up and another of it on it's side. For each I apply a single constraint to the column and the column hole in the clamp as they are what I want to line up and in both cases they go straight to where I want them, the right way up one just move sideways but the other both sideways and rotates itself to be the right way up.

No different to you picking up the parts from a pile on the bench and manipulating them the right way round so they slip onto the column. Hopefully you would not say that you can't assemble the pile of parts if they were actually on your bench just because one were laying on it's side rather than on it's bottom.

[lee websterParticipant @leewebster72680]

JasonB,

Does Atom 3D do what you show? That is a very neat way of aligning objects.

[Author]

Posts