It's a classic SB bolt composed of a screw and a nut from our library, I didn't redesign my screws and nuts. They are driven by a family of assembly parts, which drives screw and nut. The bolt is composed; A screw, blocked on the origin From a nut, coaxial stress to the screw but, and not stressed in the axial direction
Application: I install my washers independently of my bolt, and I install my bolt directly from my bookcase. I make it flexible. I then constrain the nut on the washer I apply a repetition to the " flexible " bolt I rebuild and close my assembly I open my assembly again and the pim pam poum on forced error the scandal !!
It seems that flexibility and repetition do not mix. So far no problems but now I'm having them; a colleague had this problem of editing the function, in short it's quite... Or a clever little guy removes parts from me and reinserts them
I confirm that flexibility is to be avoided for this kind of assembly (one or 2 mxi assemblies per assembly when possible. Otherwise every time you move anything SW recalculate the whole thing! And it regularly happens that a flexible assembly has its constraints that goes red. You make it rigid and then flexible again and the constraints come back well. But for the nut, and the screw, forget about the flexible assembly if you don't want to bother! The easiest way is to leave them independent and add to each of them a reference of constraints on the edge:
For flexible constraints (not tested the case of demand) I just realized that it is related to the rotation lock function of coaxiality constraints. Another thing coded with the feet because personally I don't see how a rotating lock of a screw prevents it from moving in space when you move X/Y. Rather, it feels like they used a blocking feature. Basically, by releasing these locks, the flexible assemblies behave properly.
To answer you, because I just tested this error again. And here the observation is clear, it's completely random Repeat state without sync of functions in the pattern You edit the repeat function several times and it gets wrong. We edit the function and this corrects the constrain errors. The constraints are just in yellow. Or I make the room rigid and flexible again. In another case, I open an assembly, the constraints are red and yellow and in a state of deletion due to the loss of a surface. In short, the only solution after editing the repetitions and removing and restoring flexibility, removing constraints and restressing the bolts. As if one part is replaced by another that has nothing to do
Another info, my colleague doesn't block the rotation
Yes @Cyril_f agree on the coaxiality locked that bugs (I often add a constraint // not to lock it with the function that bugs or leave it free. On the other hand, flexible constraints don't like to name more the constraints with 2 components different from the flexible assembly. (not made for that in my opinion). Basically, you should not constrain the assembly with the screw that is fixed. And in the assembly where this assembly is inserted, reconstraining the nut very often generates (see each time a stress bug.) The recalculation of constraints goes awry almost every time. For me, the flexible assembly must be inserted and fixed with constraints on the fixed part (not on the moving part under penalty of big problems!)
If we use bolt assemblies: We end up breaking them down to avoid stress problems, but it is still more practical than importing bolt components one by one... As for the rotation constraint lock: Remember to right-click on the constraint paperclip to enable/disable it: But I still prefer this possibility of rotating blocking rather than the use of parallelism constraints...
@sbadenis , unless I am mistaken in my various tests, if there is no constraint that limits the displacement in the subassembly, the positioning constraints in the upper assembly do not pose a problem. On the other hand, with repetitions I have a big doubt because the 3D models are linked so they all behave in the same way. To come back to the rotation lock, if I take a linear guide rail in assembly whose carriage is free on an axis (sliding along the rail), if I fix the rail itself with screws whose rotation is locked, it puts errors everywhere and prevents the assembly from moving in space (even though the rail is perfectly fixed, it's the cart that moves). So it's a " bug ", I spent almost a day understanding why my assembly was in error and that I couldn't move it to a higher level assembly despite all the flexibility it takes.
@Maclane Yes parallelisms to be banned, SolidWorks makes constraint control loops, and just because of a redundant parallelism to another constraint and it's the drama
Is it possible to share an example? (bolt assembly, destination assembly). I use flexible assemblies regularly (for cylinders, linear guide, etc.) and I don't have this kind of problems (yet)...
The big difference is that you use them on one or two components where @FRED78 use them on your bolts (so dozens or even hundreds of flexible assemblies). It is important to understand that flexible means that SW raises the constraints of the flexible assembly to the upper assembly → it makes it heavier. To hope to have something that works, you must not have hyperstatic constraints on these flexible assemblies: for example, to fix the face of the nut, use only one point (like the origin or a point created especially for this) and especially not the support face (otherwise we will stack the parallel surfaces together and after a while → all red).
No, unfortunately I can't provide you with a context But I can give you my method, Drilling in the room with sketch-driven repetition In the implanted screw assembly with a feature-driven pattern that is in the part (Sketch-driven Previous Pattern).
The bolt: Our library is homemade Screws and nuts are made by Ø, and controlled by a family of parts The bolt is a pilot assembly per family of parts, which points to the nut screws of the written library above. As with all fasteners
In the case of your cylinder you don't do repetitions?
When you plan to make flexible assemblies you do not constrain the part (Nut) that will be reconstrained in the assembly. So my nut is free, the screw is fixed in the assembly originally maybe that's the problem? Should the screw also not be fixed?
On the other hand I don't understand the story of parallel faces!! For me this is not a problem??
@FRED78 there may be an intermediate solution if you want to continue using this type of assembly: through Configuration publisher.
The first advantage would be that Solidworks will systematically ask for the size of the bolt to be inserted each time it is inserted. -The second advantage is that you will be able to drive a distance (free or under constraints of the mini/maximum type) between the nut and the screw head.
on the other hand, the disadvantage is that you will have to leave your screw-nut assembly accessible to write, which will allow Solidworks to create the new configurations on its own (screw size + lg screw/nut).
Subsidiary question: why do you use an external image host (Goopics)? The forum is however able to directly accept copy-paste for many formats (images / video / links...).
… And I don't have a problem with constraints when reopening.
For your part, what exactly are the constraints that appear to be in error when reopening? Stress of the nut on the screw? Nut stress on the first-level assembly component? Screw stress on the first level assembly component?
I noticed one thing: The flexible state of the bolt subassembly is maintained for repeated instances. On the other hand, the constraint that positions the nut is not duplicated. And we end up with all the nuts of the repeated instances free in translation.
It forces us to redefine constraints. Isn't that a bit restrictive?
I would take some time to look at publisher. The only problem is the standards on the screws, one of our guys could create a length that does not exist in the standard. If you saw the table I had to make for the screws (What to use and in what case) But the idea is good I see how it works
Try to make a hole driven by a sketch And the screws driven by the previous repeat function.
But I have a doubt that a clever little guy would have deleted my folder and who would have blown up the links, I have some clever guys here, formatted like the game of throne of geniuses
having them; a colleague had this problem of editing the function, in short it's quite...
also not be fixed?
But the idea is good I see how it works
it's weird