AnsweredAssumed Answered

mates flip in different configurations

Question asked by Andrew Michell on Mar 28, 2020
Latest reply on Mar 29, 2020 by Elmar Klammer

Like • Show 0 Likes0
Comment • 4

I have a mechanism in which a control wheel can be placed in any position from 0 to about 225°. I have various configurations which fix it at different angles. Initially I defined those positions using an angle dimension between two planes, but discovered that in changing from one configuration to another Solidworks cannot tell the difference between 10° and 190° - or presumably, that it does not always measure the angle to the same side of the planes. On the advice of my VAR I then created a sketch like a slice of pie, defining the angle subtended at the apex, and then mating my wheel to one edge of that slice of pie. Now when I change from one configuration to another the slice of pie in the sketch does always have the correct angle, but the mate between the wheel and the line flips..... so I am back to square one.

I have had this problem before, and I suspect so have countless other users. There must be an easy way to ensure that the angular position of a an object can be reliably controlled through 360°. Does anyone know how?

No one else has this question

Outcomes

Solid Air Mar 28, 2020 8:31 PM

Yes I have ran into this before. Unfortunately about 5-6 years ago. I will have to search. Please post a picture of your slice of pie sketch and what mates you are using. Sketch is the right way to go though.

Actions

Neville Williams Mar 28, 2020 8:53 PM

Once you get over 180 deg all bets are off as to what is going to happen.

Have just done a job that has a simple arm that can be flipped up. I had to show it in a few positions and this problem reared it's ugly head. Also using this assembly as flexible in another assembly and I have a feeling the program gets itself in a twist with this enabled

I used a Limit mate.

To me, Mate alignment somehow gets confused or program is having an issue with plane normal and flipping it. I thought I would be smart and use the primary top plane but same result.

I know this does not help you, but thought you should know I'm one of those "countless other users" who wish this would be fixed - it has been an issue for a very long time.

Actions

Vladimir Urazhdin Mar 28, 2020 10:21 PM

What if suppress the flipped mate and create the dedicated mate with a right alignment for each configuration?

Actions

Elmar Klammer Mar 29, 2020 5:59 AM

Hi Andrew,

Let's put aside that this has been a ongoing quirk of SW. I found a combination of the two approaches works most often.

Use a sketch to drive the angle. But don't constrain to the sketch line itself. Sketch lines have a direction too. The pick point sequence matters. And somehow the solver gets confused if the orientation of the line switches from top to bottom. That's why I add another angle helper plane that is perpendicular to a main plane (preferable top, bottom or right) and is colinear to the sketch line. Since planes have a direction as well. Then mate to the newly created helper plane instead. The idea is that the normal vector for the helper plane comes from the main plane and not the sketch. So when the sketch line flips direction it doesn't affect the normal plane vector itself. Has worked for me in the past. But there has been cases when this didn't work as well. In those case I try to define my helper plane around the axis or rotation and angle dimension to main plane.

Additional. Don't use flexible mates whenever you can avoid them. Only use them if you have a good and pressing reason. They kill you in larger assemblies and are likely to wreak havoc for the mates solver. Most sub assemblies that need to move can be split up. I.e. a hydraulic cylinder. I make a configuration for the rod and on for the barrel. Then I set the BOM quantity for the Rod to zero. That way my counts are correct. Now I mate them as needed. You can also use envelope feature if you need to show the complete Zylinder in your SUB. In my experience, this beats flex assemblies by a long shot for most common applications.

Using the reference vector option in your angle mate dialog window is advised.

To be aware of. The amount the angle mate changes between configurations makes a difference. If you go from 10-75-150-210 is often more stable then going from 10-210 for instance.

The order of your mates is important. It can happen that higher order mates have an adverse effect on lower ones.

Clean & simple mating is a must. Keep the number of mates low. And one more thing. A fixed component at the top of the feature tree is less likely to cause mate errors. I know there are some that think all components should be mated and none fixed. I found the approach with a single fixed anchor component works better.

Actions

4 Replies

Solid Air Mar 28, 2020 8:31 PM
Mark Correct
Correct Answer
Yes I have ran into this before. Unfortunately about 5-6 years ago. I will have to search. Please post a picture of your slice of pie sketch and what mates you are using. Sketch is the right way to go though.
Like • Show 0 Likes0
Actions
Neville Williams Mar 28, 2020 8:53 PM
Mark Correct
Correct Answer
Once you get over 180 deg all bets are off as to what is going to happen.
Have just done a job that has a simple arm that can be flipped up. I had to show it in a few positions and this problem reared it's ugly head. Also using this assembly as flexible in another assembly and I have a feeling the program gets itself in a twist with this enabled
I used a Limit mate.
To me, Mate alignment somehow gets confused or program is having an issue with plane normal and flipping it. I thought I would be smart and use the primary top plane but same result.
I know this does not help you, but thought you should know I'm one of those "countless other users" who wish this would be fixed - it has been an issue for a very long time.
Like • Show 0 Likes0
Actions
Vladimir Urazhdin Mar 28, 2020 10:21 PM
Mark Correct
Correct Answer
What if suppress the flipped mate and create the dedicated mate with a right alignment for each configuration?
Like • Show 0 Likes0
Actions
Elmar Klammer Mar 29, 2020 5:59 AM
Mark Correct
Correct Answer
Hi Andrew,

Let's put aside that this has been a ongoing quirk of SW. I found a combination of the two approaches works most often.
Use a sketch to drive the angle. But don't constrain to the sketch line itself. Sketch lines have a direction too. The pick point sequence matters. And somehow the solver gets confused if the orientation of the line switches from top to bottom. That's why I add another angle helper plane that is perpendicular to a main plane (preferable top, bottom or right) and is colinear to the sketch line. Since planes have a direction as well. Then mate to the newly created helper plane instead. The idea is that the normal vector for the helper plane comes from the main plane and not the sketch. So when the sketch line flips direction it doesn't affect the normal plane vector itself. Has worked for me in the past. But there has been cases when this didn't work as well. In those case I try to define my helper plane around the axis or rotation and angle dimension to main plane.

Additional. Don't use flexible mates whenever you can avoid them. Only use them if you have a good and pressing reason. They kill you in larger assemblies and are likely to wreak havoc for the mates solver. Most sub assemblies that need to move can be split up. I.e. a hydraulic cylinder. I make a configuration for the rod and on for the barrel. Then I set the BOM quantity for the Rod to zero. That way my counts are correct. Now I mate them as needed. You can also use envelope feature if you need to show the complete Zylinder in your SUB. In my experience, this beats flex assemblies by a long shot for most common applications.

Using the reference vector option in your angle mate dialog window is advised.

To be aware of. The amount the angle mate changes between configurations makes a difference. If you go from 10-75-150-210 is often more stable then going from 10-210 for instance.
The order of your mates is important. It can happen that higher order mates have an adverse effect on lower ones.
Clean & simple mating is a must. Keep the number of mates low. And one more thing. A fixed component at the top of the feature tree is less likely to cause mate errors. I know there are some that think all components should be mated and none fixed. I found the approach with a single fixed anchor component works better.
Like • Show 0 Likes0
Actions

mates flip in different configurations

Outcomes

Related Content

Recommended Content