mates + best practice

Should cylindrical features be fully constraint even if they don't have to be? An example would be a shaft and a hole. The shaft has a coincident mate
and a concentric mate. Should the shaft also have a mate to prevent it from rotating? Which is the better practice? What are the implications of each case?
Thanks
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
The best practice is to not add a mate to fully constrain the component. SolidWorks will take time to calculate the mate so adding the mate will slow SolidWorks down.
If your like me I hate seeing the (-) indicating the component is not fully constrained. I submitted an enhancement several years ago to have the indicator change if the component has one degree of freedom left. (r) would indicate the part rotates along a single axis, (l) indicates the part travels along a single axis, and (rl) indicates the part travels and rotates along a single axis. The addition of these indicators would let the user know how the component is mated by checking the FeatureManager.
Greg
Knotfreak wrote:

have to

coincident mate

it from

each
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
speaking of mates...
I would like to know if there is a way to use two mates to locate a part that slides on two pins. Think the top of a dieset sliding up and down on its pins (sorry I don't know the proper terminology). When I try to putting a concentric mate on the first pin to its opposite bushing thats fine, but trying to mate the other pin to its opposite bushing yields an "part is overconstrained error".
I know I could refernce off parallel faces of the dieset, but adding mates to the parts that do the guiding in real life is more "natural" to me.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Are the pins the same distance apart as the holes?
Are the pins and holes all parallel to the same direction?
If they're off even a little bit, it will easily overdefine.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Yes, and Yes.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Bill,
NEVER use multiple concentrics to locate pins, holes, etc.. They will go over defined even if they're perfect. Keep the mates as simple as possible. On a die set, or mold base, I pick a single pin or hole for concentric. I then add a parallel, either a plane or a face.
When Matt siad "a little bit" he wasn't kidding. Probably somwhere around .000000001". SW used to be more tolerant of this.
Regards
Mark

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Thanks, thats what I've found. I was just wondering if I was missing something, like an adjustable tolerance for concentricity, or something.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I agree on what you are saying, but disagree in the practice of it. My philosophy is, for example, that if you can't mate two holes in a plate concentrically, then you don't have them dimensioned the same, and I would like to know about the error right away. Now, I do understand that every once in a while it's unavoidable, but for the most part, if you want holes to line up, dimension them properly.
A common mistake is when mixing inches & mm in that someone will dimension one side to 25mm and the other person will dimension the other side to .984" because they don't like to work in metric. Maybe close enough to work, but then again, maybe not. So I tend to practice the more exact method of precision so my habits are there when it really counts.
WT

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

My experience is that, even when holes are dimensioned correctly and the center to center distance matches exactly to the eighth decimal point when measuring, it's still a crap shoot whether or not you can mate both concentrically.
Jerry Steiger Tripod Data Systems "take the garbage out, dear"
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Wayne,
Well, your welcome to fight with the results if that's what ya want.
There are lots of other less problamatic ways of insuring accuracy.
Regards
Mark

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Here's my take on concentric-concentric -
Put a 1-2-3 block down on a surface plate with the 3" height facing up (smalledst face supporting).
It's a lot like that - it will stay, but it's not the most sound state. Eventually the block will end up on its largest face.
Actually the block is really 6" long in this case.
:)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.