This won't be anywhere near as complex (or useful) as this one that Alin Vargatu posted, but maybe it will be of some interest. A few years ago I ran into a situation where an inserted Part was fully constrained with two Standard mates (Lock wasn't used, and this was before "lock rotation" on Concentric mates was available). Has anyone else run into this, or can figure it out if you haven't?
A coincident constraint (surface to surface) and a concentric constraint will work if the axis of the hole is NOT perpendicular or parallel to the constrained surface.
See example attached:
That was quick. I'm impressed, and that certainly qualifies, but it isn't the situation I referred to. I'll post it in a little while.
You should be able to use the 'Profile centers' mate and select the circular edges, this should allow for rotation only. After that you could mate two of the planes as parallel.
Stavros Pantelakis wrote:
You should be able to use the 'Profile centers' mate and select the circular edges, this should allow for rotation only. After that you could mate two of the planes as parallel.
Please notice I said only Standard mates, so I'm afraid that doesn't qualify, but thanks for posting it.
yea, I initially thought the concentric mate would function similar to the profile centers.
Glenn, since you aren't specifying the parts that leaves it open for us to define. I often insert countersunk fasteners. They come in with a coincident smart mate of a circular edge to a conical countersink. Applying a parallel mate to one of its planes or straight edge to some other suitable feature keeps it from rotating and makes it fully constrained. The same thing with a concentric mate paired with a parallel or coincident mate.
Two concentric mates of cylindrical feature pairs that are not parallel.
Now I don't know whether to give credit for the correct answer to Roland Schwarz, since what he posted is what I had in mind (at least I think this is what he meant), or to Dan Pihlaja for posting the first answer that fits the criteria.
1 Concentric,
1 parallel/perpendicular between planes
Ned Hutchinson, That would still leave one direction of freedom. The direction of freedom would be along the axis of your concentric mate.
Coincident with align axis between origins and/or coordinate systems = 1 mate.
Otherwise 2 coincident mates on 2 pairs of edges would do the trick.
"Otherwise 2 coincident mates on 2 pairs of edges would do the trick."
this was my first though. but..."nonparallel edges..."
As we all know excessive mating has it's own issues and so does having extra mates in SolidWorks Models.... Most importantly reducing the amount of mates in an assembly should be in your design intent criteria, as excessive mates can make SW crawl. With that there are choices we need to make when approaching a new design or re-doing an older design. Most of my designs have very little mates, everything is built in context and fixed to the point of origin and the three planes, however when I do need to use mates, (which are for multiple instances or interchangeable components such as hardware etc), I look for (2) straight edges or sketch lines to lock in the parts.. So if you need a shaft, how do you approach the design, a circle and extrude or do you use the revolve feature, if you use the revolve feature you have the sketch lines to use for mating.... You can also mate two circles with one coincident mate, however the parts will still be able to spin, but with a straight shaft without holes, it doesn't matter if the shaft spins..
John Stoltzfus said:
"You can also mate two circles with one coincident mate, however the parts will still be able to spin, but with a straight shaft without holes, it doesn't matter if the shaft spins.."
My OCD doesn't allow this......I am required by my inner feeling of insecurity to make sure that all things are fully defined and mated.
Dan Pihlaja wrote:
John Stoltzfus said:
"You can also mate two circles with one coincident mate, however the parts will still be able to spin, but with a straight shaft without holes, it doesn't matter if the shaft spins.."
My OCD doesn't allow this......I am required by my inner feeling of insecurity to make sure that all things are fully defined and mated.
Mine either. I can't have those minus signs in front of the components in the tree.
locking in the rotation of shaft or fasteners is almost the only time I use parallel mates.
Timothy Taby wrote:
locking in the rotation of shaft or fasteners is almost the only time I use parallel mates.
I don't bother with Parallel mates for that. It's so easy to right-click on Mates in the tree and choose "Lock Concentric Rotation". You can do it as each concentric mate is applied, but I often insert and mate several fasteners, then do this for all at one time. It's much faster than using Parallel mates, and works just as well (assuming you don't need them at a particular orientation). That's been available since SW2014 (see here).
I'm with you there on the Lock Rotation, Glenn. However, when I bring in the countersunk fastener it only puts a coincident mate between the edge of the fastener and the conical countersink so there is no concentric mate to lock. In these cases when I want to lock down the fastener I find the parallel mate is much easier to use with an edge or a face of the hex socket to some other straight face or edge of the assembly.
I'm thinking you got it Dennis Dohogne and if Glenn Schroeder steps up to the plate, he'll pass you 20K points
John, I'm more interested in sharing tips, techniques, best practices, clever ideas, etc., than in messing with the meaningless points. Although I must say you have have become quite the points philanthropist! Perhaps we should call you the Bill Gates of the forum points!
Dennis Dohogne understand it clearly but in the old days when there was a challenge - it either ended in a draw or victory - Glenn Schroeder is the one that took the liberty to call it a challenge - lol
We all know the point system on the forum don't mean nothing at all and agreed on the sharing/caring part
Mine neither, but in all reality there is nothing moving except the rotation and is another way to reduce mates, just using a standard mate..
that's why I requested the lock rotation feature back in the day. I couldn't stand seeing the unconstrained parts. not because fasteners rotated, but because with the ( - ) I was never sure if there were other missing mates.
John, you're making some good points for weldments whenever appropriate. NO mates!
Coincident mate of male/female conic features, & a parallel mate. 2 mates, fully constrained.
If the edge of the hole and shaft are the same in diameter use coincident and parallel
use parallel and coincident, if i get what you mean
I like this "challenge" as it offers the conversation of how we do the things we do and why.
I use varying techniques based upon criteria and project scope. Quoting projects usually consist of several small virtual assemblies contained by origin mates only and then all virtual parts in those subs with purchased parts put in as needed. So the only mates outside of origin are for purchased parts.
Doesn't work on all of our projects but it is one method that facilitates no red tree of horror down the road as changes are made.
2 mates to lock in a part is easy if the parts are made for it, I have a pin with a cross hole that goes into a receiver block with a hole for the pin and a clearance hole to align with the cross hole. 2 concentric mates will lock these 2 parts together.
Coincident to the origin point and a 2nd constraint to align to a plane?
One cool think about locking concentric mates is the fact that you can lock and unlock mates in bulk.
Do this test:
Filter by "Concentric", Shift-Select surviving mates and lock or unlock all of them.
Works great with selection sets also.
Behind the scene, SW will add a parallel mate between 2 planes. Just so you know.
Alin Vargatu wrote:
One cool think about locking concentric mates is the fact that you can lock and unlock mates in bulk.
Do this test:
Filter by "Concentric", Shift-Select surviving mates and lock or unlock all of them.
Alin,
There's an easier way. See my Reply #19 above.
Glenn Schroeder wrote:
Alin Vargatu wrote:
One cool think about locking concentric mates is the fact that you can lock and unlock mates in bulk.
Do this test:
Filter by "Concentric", Shift-Select surviving mates and lock or unlock all of them.
Alin,
There's an easier way. See my Reply #19 above.
Were do you see this numbers? #19?
Alin Vargatu wrote:
Glenn Schroeder wrote:
Alin Vargatu wrote:
One cool think about locking concentric mates is the fact that you can lock and unlock mates in bulk.
Do this test:
Filter by "Concentric", Shift-Select surviving mates and lock or unlock all of them.
Alin,
There's an easier way. See my Reply #19 above.
Were do you see this numbers? #19?
Dan,
You can only see those numbers in the Discussion's page. They don't show up if you follow a Discussion from your Inbox.
Glenn Schroeder wrote:
Alin Vargatu wrote:
One cool think about locking concentric mates is the fact that you can lock and unlock mates in bulk.
Do this test:
Filter by "Concentric", Shift-Select surviving mates and lock or unlock all of them.
Alin,
There's an easier way. See my Reply #19 above.
I agree. My technique makes more sense for selection sets containing mates.
The only issue with "Lock Rotation" is that it doesn't carry through with "Copy with Mates" while a parallel mate does.
I use "Copy with mates" all the time, so using the parallel mate once and then copying saves me time.
I share that frustration, Dan. I would like to see Copy With Mates work with however many (or few) mates a component has, even if it is not fully constrained.
Dan,
You can use the Parallel mate, but you can also add as many components with Copy with Mates as you want, and then lock rotation on all of their concentric mates with two clicks. And that's two clicks total, not two clicks per mate.
From Discussion page:
After selection:
Still no numbers:
Okay, I give up. Maybe it depends on which Browser you're using? I'm still on Internet Explorer.
In rigid body mechanics, a completely unrestrained body has 6 DOF. 3 translations and 3 rotations. Several years ago I made up a little table that listed which and how many DOF are removed by each mate. For example, plane-plane parallel constrains only 2 rotation DOFs (or DsOF?). Point-point coincident constrains 3 translations, no rotations. For a while I tried mating everything in my assembly with the absolute minimum required constraints to fully define each component. For example:
One hole-hole concentric (2 rotations, 2 translations), one point-plane coincident (1 translation) and one plane-line parallel (1 rotation).
One circular edge-circular edge concentric (2 rotations, 3 translations), one plane-line parallel (1 rotation), which coincidentally is another answer to the OP that I don't recall seeing (4 pages is a lot of posts, coulda mist it).
I thought that eliminating redundancies might improve performance. It's been a few years, but I seem to recall that as my assembly got larger, it actually increased the tendency for the assembly to be unstable. I think the redundancies as the mates solved actually ended up helping rather than adding overhead. One day maybe I will have time to do a rigorous comparison.
Redundancies are OK if they don't devolve into conflicts. Most times they don'e.
I prefer to constrain rotation by making a plane perpendicular to another. Much more robust than parallel.
For fully constrained with only 1 standard mate...
Coincident, coordinate system to coordinate system, with the Align axes box checked.
(Does not need to be origin to origin, you can use any coordinate system or origin.)
