INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Jobs

Real world v SolidWorks

Real world v SolidWorks

(OP)
Bare with me everyone, I know I'm as king a lot of questions....(and thanks to all that have taken the time to answer so far)

Background - I am new to modelling / CAD / drawing etc. I am transferring some old autocad files into a 3d model. The files represent a replica of an old swing frame - nothing smart, just some tubes, hinges and an apex joint.

The whole lot is at http://www.wilverley.com/Assembly.zip and the  the e-Drawings file is at

http://www.wilverley.com/Assembly.htm

I have two problems:

1) I have a folding brace (like the ones you get on baby push chairs) which connects to tubes which are fixed at one end to a joint. The brace locks the two tubes (legs) and when locked out fixes the distance of the two legs.

The problem is in 'the real world' the brace has an element of flex around the set screws where it is fixed and both faces of the stay that connect to the legs sit flush against the legs. Solidworks wont let me create a coincident mate as this would mean the other components would have to twist. It makes sense ofcourse but in the real world I am not forced to have a 5mm gap..... What should I do to emulate real life.

2) Before the legs were linked together by the brace they had movement in two ways

 a - rotation about the set screw where they are fixed to the apex joint hinge.
 b - where they are attached to the apex joint, the hinge they are attached to had perperdicular rotation about another set screw.

Once the two legs were linked by the stay I lost movement 'a' even if both legs were selected together.

Thats the end of my questions ! I can't find anything in the help files or on web searches etc.


Charles

RE: Real world v SolidWorks

Ask yourself: am I trying to create a motion simulation or am I just trying to get drawings done.  You may be doing a lot of work that will have no benefit.

That said, here's some ways to do it:
1) Create points to constrain joints.  This will allow you to have virtual ball joints that will help you keep from creating an overconstrained condition.
2) Create configurations of parts in bent positions.  If you have a part that needs to be flexed.  Create a configuration of the flexed part for use in the assembly.

-b

RE: Real world v SolidWorks

(OP)
b,

1) I'll take a look at constaint points - I have not come across them before - only mates in an assembly (as described in the help)

2) I do see what you mean.....the drawings have two purposes a) for construction and b) detailed instructions on how to errect the frame. It would therefore be very nice to freely manipulate the frame as it can be in real life. Are you saying I can't match the real world motion in solidworks.....drat, if thats the case.

Charles

RE: Real world v SolidWorks

Concerning your issue with the 5mm gap.  Make a plane on one of the parts (we'll call it part1), offset 5mm.  Mate this plane in your assembly coincident to the face (on part2) that you intended to mate the original face (from part1).

-Shaggy

RE: Real world v SolidWorks

Charles,

You may be able to match the motion, but it's probably more work than it's worth.  Solidworks is not realtime FEA, so it doesn't know how to deal with parts that deform unless you are explicit in their definition.  One way to do this (which I use for animations) is to use equations or assembly references to define parts that move or flex in response to the assembly.  Not worth the time unless you have a very specific need.

-b

RE: Real world v SolidWorks

(OP)
b,

I can definately say that equations would put be over the edge ( I doubt my maths and grey cells would cope ). I'm still unclear as to why the legs are locked in position.

Why is it that once the stay is in place both legs seem to forget that they can move - they did before. The stay just locks then into the same plane it shouldn't prevent them from being able to move with each other?

Charles

RE: Real world v SolidWorks

(OP)
Shaggy,

I see what you mean however the plane is not required as to achieve that result I can just make the hole on the leg and the hole on the stay coincident. I'm still left with the 5mm gap either way.

I like b's idea re ball joints, however that idea would mean the two legs would never want to move together as the stay would not be constraining them at all.

I can put an 's' bend in the stay to take up the 5 mm gap however in the real world its not required.

Charles

RE: Real world v SolidWorks

The legs don't become locked on my computer.

The ZIP file you posted doesn't contain the "Metric\washers\Flat Washer Narrow_AM" part so the mates connected with it are suppressed/broken. Maybe it has something to do with that.

cheers
Helpful SW websites  FAQ559-520
How to get answers to your SW questions  FAQ559-1091

RE: Real world v SolidWorks

(OP)
CorBlimeyLimey,

I am a limey!.... I have replaced the washer (it was part of the toolbox) and uploaded the lot to the same location. If you know of an easy way to replace a part with another but keep the mates I wouldn't mind knowing.

The legs are not constrained to move towards each other but with each other about the set screws which holds them to the apex joint when the stay is in place. If it however is OK on your machine then I will look a twit as it would seem I just can't use 'move component' !

Charles

RE: Real world v SolidWorks

What about making the flexing tab at at the end of the tube a separate piece and constraining it to flex like a hinge?

--
Hardie "Crashj" Johnson
SW 2005 SP 4.0 (reluctant to change)
Nvidia Quadro FX 1000
AMD Athalon 1.8 GHz 2 Gig RAM

RE: Real world v SolidWorks

(OP)
SnowCrash,

Thanks for the comment.

It is constrained like a hinge, well atleast its mated so and seems to move like I expect - note I am new to cad drawing in all aspects.

Movement in that direction is ok, its roation about the joint of the leg to that tab that is the issue. It would ofcourse only be possible to know that if you downloaded the assembly and not just the e-drawings file.

Charles

RE: Real world v SolidWorks

(OP)
SnowCrash,

Or maybe I misunderstand you. Do you meen the flattened part of the tube. It is a seperate part, but how do you mean to constrain it lige a hinge?

Charles

RE: Real world v SolidWorks

(OP)
Also, as an aside what is the standard way to hang something from something else. Are planes the best way?

A real life example would be: How would you model a ring hanging from a hook?

Charles

RE: Real world v SolidWorks

A few suggestions:
First suggestion:
I try to create a sub-assembly the way it would be created in "real life".  Your assembly "Assembly Folding Brace" has 10 components, 4 of them should be 1 sub-assembly:
Folding Brace Part B
Folding Brace Stop
Bead9
Bead10
This would result in less mates required at the assembly level (cleaner, easier to troubleshoot).
To create a sub-assembly, pick the 2 parts only (beads won't work), right-click menu, down arrows, then pick "Form Sub-assembly here".  
See screenshot: http://img119.imageshack.us/my.php?image=formnewsubassemblysb4.jpg

Name the sub-assembly and add the weld beads in the new sub-assembly.  Delete the unnecessary weld beads in the other sub-assembly.

Next suggestion:
I usually mate 2 parts to each other first without hardware.  The screws are then added later.  This way you can pick 2 parts/assemblies in the tree and easily find out how they are mated.  For example, in the top-level assembly you posted, CTRL+pick Assembly Apex and Assembly Leg, then pick the middle-tab in the task pane.  This is the "Property Manager".  The mates of the selected models will be highlighted.  In this particular case, there are no highlighted mates.  So you will have to go through all mates to see why your legs won't swing.  

As for your "real world", I would make another configuration, one that is static and locked for your drawing, another that isn't locked to test motion.  Actually it is a cheap kinematics, and you may need it to fold completely for packaging (what size box do I need).  On the top level assembly, under advanced mates pick the angle option.  This will keep your folding brace from folding/unfolding too much.  
Finally, for the gap, you would only need it for illustration purposes only, so make a configuration of Folding Brace Part A and put a slight offset too close the gap.  For shop drawings, show the flat part.

Flores
SW06 SP4.1

RE: Real world v SolidWorks

(OP)
scmadman,

Good points - I have done what you said re the sub assembly and an excellent point re the angle mate. It now doesn't fold back on itself  - I managed not to notice the advanced mates.

For the 5mm gap simply model it as it theoretically should be and be done with it. Agreed. I guess slop doesn't exist in computers! I have updated the 'Assembly - Side.SLDASM' at

http://www.wilverley.com/Assembly.zip

So all this now leaves is the rotation of the legs where they are fixed to the apex joint. They moved before the brace is added but not when it is in place which... just isn't right ....or are you saying that I need a static configuration and it can't be done.

Charles

RE: Real world v SolidWorks

The mates causing the "problem" are "Concentric31" and "Concentric34". These are constraining the legs to the Folding Brace such that the legs cannot be moved sideways.

If you could dimple or csk the holes in the leg & use a spherical faced washer, you could then use tangent mates between the sphere & the csk which should allow the sideways movement you need ... for the model. As you mentioned though, in real life, the retaining screw would not allow the rotation to occur without flexing of the brace.

cheers
Helpful SW websites  FAQ559-520
How to get answers to your SW questions  FAQ559-1091

RE: Real world v SolidWorks

By the way, when I posted earlier that the model worked OK for me, I did not realise you were trying to move the legs sidways.

One other suggestion, instead of making the leg an assy, it could have been created as a part. See compress a tube in solidworks  thread559-148035

cheers
Helpful SW websites  FAQ559-520
How to get answers to your SW questions  FAQ559-1091

RE: Real world v SolidWorks

(OP)
So, in effect make the attachment between the leg and a brace a ball joint ( as per bvanhiel, just in a different way ).

Now I may be over discussing the point here and I understand that the concentric mates are preventing one leg from moving away from the other but I should be able to move both legs at the same time as I can on the live frame since this will retain the alignment of the concentric mates?

Either way I'll try the ball joint idea.

Charles



RE: Real world v SolidWorks

(OP)
Grrrr, I can't get a ball joint to work....I'm having a bad week. I decided to try a reference point mating to a spherical surface as a test but the mate is not allowed. SW goes on about the distance being Xmm away.

I'll battle on. I tried to search for ball joints on this forum but the link to the example is dead....Grrr again.

Charles

RE: Real world v SolidWorks

You should be able to move the legs sideways in SW only if the legs are parallel, otherwise the sideways movement would be through a compound angle. You should be able to do that with configurations using angle mates, but not with a flexible assy where the parts are manually moved.

FYI, in the ball joint (joystick) example, the centres of the Ball & Socket were placed at the models origins and the origins were then used in a coincident mate. A plane perpendicular to the joystick handle was then used in a limit mate with a matching plane in the socket, to limit the amount of joystick "tilt".

cheers
Helpful SW websites  FAQ559-520
How to get answers to your SW questions  FAQ559-1091

RE: Real world v SolidWorks

Quote:

Or maybe I misunderstand you. Do you meen the flattened part of the tube. It is a seperate part, but how do you mean to constrain it lige a hinge? Charles
Yeah, I meant the flattened part of the tube could be a separate part. Constrain it to the tube with centerline plane to centerline, then one edge to the surface of the end of the tube. This would allow it to flex the way it actually does bend.
"Sort of"

--
Hardie "Crashj" Johnson
SW 2005 SP 4.0 (reluctant to change)
Nvidia Quadro FX 1000
AMD Athalon 1.8 GHz 2 Gig RAM

RE: Real world v SolidWorks

(OP)
CorBlimeyLimey,

Thanks for info re the ball joint. I have now made the legs parallel by aligning the sides of the parts that join the loft  - they were parallel and were parrallel as enforced by the stay before hand.

Still no luck - this is all very strange and I still can't see why the legs don't rotate with each other. If the hinged stay flexes with the movement of a leg then why don't the two legs rotate with each other since the stay is locking them together. The rotation does not pull the stay out of is mated position......Going very mad now...

For what it is worth I have updated

http://www.wilverley.com/Assembly.zip

'Assembly - Side' being the main file.

Charles



Charles

RE: Real world v SolidWorks

(OP)
SnowCrash,

But the tube doesn't actually bend in real life it rotates around the set screw that connects to the apex joint. This is hard to describe in words and I again could be misunderstanding you.

The tube can also go 'in and out' perpendicular to the rotation as the thing it attaches to is a hinge (of sorts).

I am very much a novice and I hope I'm not getting the wrong end of the stick and I do appreciate all comments made.

Charles

RE: Real world v SolidWorks

(OP)
I'm wondering if this is a bug in Solidworks. I'm continuing to try but logically Solidworks should understand the model.....

RE: Real world v SolidWorks

It is not possible to fold the legs because of the concentric mates on the legs and brace.   You would need a universal joint for the hinges to bend the legs back in SW.  You might be able to cheat that effect by adding a point to the holes and constraining the points to the temporary axis of the mating holes.  Also add distant mates to the points.  

It isn't a bug in SW, the only reason you can fold the legs back in real life is because of the gap between the screw holes and the screws.  

Flores
SW06 SP4.1

RE: Real world v SolidWorks

(OP)
Flores,

Are you saying that 0.2mm gap between the set screws and the holes which in real life is much less due to paint is allowing the legs to rotate around the hinges by 180 degrees?

I have a frame in front of me and the bolts you mention are torqued to a good amount and there is probably only 0.1 mm diameter difference between the bolts and the holes and probably less as some needed to be twisted throught the holes due to thick paint.

If both legs rotate around the the point where they are fixed to the apex joint (at the same time and with each other) how can the concentric mate be broken?

I can understand one leg can not move without the other as the concentric mates will be broken but both at the same time is fine. I'm not trying to labour the point but I just don't see why the legs can not rotate with each other around their respective apex fixing points.

Charles

RE: Real world v SolidWorks

I went upstream to the Assembly Apex Joint, and one of the problems is there.  Mater errors, not SW bug.  In order for the legs to fold forward the 2 parts "Hinge" would have to be parallel to each other.  Open Assembly Apex Joint.SLDASM and pick the 2 holes highlighted in the screenshot below; you end up with mate errors.

http://img412.imageshack.us/my.php?image=assemblyapexjointvv3.jpg

There isn't a fudge factor in SW, I don't know how many decimal places SW is accurate to, but if your holes are off by .0001, your concentric mates and other will cause errors.  In real life, there is wiggle room.


I will say that I think you are the first one that I recall using eDrawings instead of static pictures to help us help you.  

Flores
SW06 SP4.1

RE: Real world v SolidWorks

(OP)
Flores,

Are you saying I should be able to make those two holes concentric? I should and I can't, suggesting there is a problem. The problem must be in holes of the front back plates of the joint...... Checking it all now - v. embarassing if it is......

Charles

RE: Real world v SolidWorks

(OP)
Flores,

I think you may have hit the nail on the head so to speak. I think it is definately 'Apex Cover Back Plate' that is causing the problem and apex cover front plate is derived from it by hiding / showing an extruded cut.

I have modelled the back plate based on the profile of the 'Apex Cover Formed Plate'. I have used the formed plate  in terms of looking at the measurements / dimensions.

The arc was wrong and the profile just out when precision was increased. So..... can I make the back plate using the exact profile of the formed plate using some tool method that is outside of my knowledge. If there was such a way I would know the 'back plate' would always match the 'formed plate'.........the legs may then move and that would be a relief......!

Charles

RE: Real world v SolidWorks

(OP)
p.s. I know it must be the left and right hand set of holes of the front and back plates that are out of align but I should fix this problem first and irradicate all errors ....

RE: Real world v SolidWorks

(OP)
Flores,

Still no joy. The holes on the hinges as indicated in your picture can now be made concentric. There is still no circular movement around the bolts that attach the legs to the apex joint.

Is there a way to define that two parts should always move together.... This is what I see as the problem. If both legs moved together then all would be OK. In my tiny mind this just doesn't add up.....

The later stages of the assembly require that the legs can rotate with each other as the 'A frame' is braced to a central beam.

Without movement at the joints indicated as it stands SW can't model what I have standing next to me

Assembly reposted to

http://www.wilverley.com/assembly.zip

with the main file being 'Assembly - Side'

and eDrawings file posted to

http://www.wilverley.com/assembly.htm

Charles

RE: Real world v SolidWorks

IN SP4.1 I get a bunch of errors because you have over mated the assembly.

1) forget about the bolts and screws and just mate the assembly the way you want it to work.
2) make sure the assembly is working after each mate.
3) as you add the bolts continue to check the movement after each one. this will help you determine which mate is causing the problem.
4) The Parallel mate (legs) seems to be one of the mate issues, along with to many coincident mates.
5) I suppressed all the bolts, washers (except those that separate the legs) and deleted all the mates and started over.
6) You could make 2 configurations setting up each one to do one operation. You won't get this assembly to move in both directions with what you have on this assembly.
7) with a config you can suppress and unsuppress what mates you want to have activated. This gives yo uthe advantage to determine what's on and whats not.

but as to moving in both directions no I don't think is going to be possible, not only because of the Parallel mate, but also because ofhte hinge at the middle of the legs. They also control location of the legs. COnfigurations maybe the only thing you can do or maybe write some equations.

Regards,

Scott Baugh, CSWP
www.scottjbaugh.com
FAQ731-376

RE: Real world v SolidWorks

(OP)
Scott,

Thanks for having a look. I'm running a second hand copy of SP0 and I have no errors. Also the parallel mate is not really required as the other mates infer the hinges being parallel.

So I'm left with two seperate configs.... Am I just imagining it or should it really be able ro have both type of movement. If I delete the parallel mate nothing is any better.

I'll do as you say and essentially start again however its difficult as only once the legs are attached in all the places shown does it all go down hill.....I hope not to beaten by this !

Charles

RE: Real world v SolidWorks

I am using SP4.1 so it could be the SP you and I are on. I didn't delete them I just suppressed all the mates. It's very confusing that's why I suppressed the bolts and stuff to bring it back down to the basics. try getting to the basics and see what you can do there.

Regards,

Scott Baugh, CSWP
www.scottjbaugh.com
FAQ731-376

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close