Ever year when a new release of SE comes out I try synchronous in the hope that, at last, it will work for me. This year is no exception and I chose a part already mostly designed in ordered but needing a couple of small mods which are quite laborious to do in the ordered environment. I converted the part to synchronous and tried to move the highlighted faces - they are two parallel faces containing three counterbored and tapped holes and they needed to be moved normal to the faces and away from the part by 2mm exactly. This short video is the result of a couple of hours experimenting.. (Please bear with the amateurish video - it's a first time for me)
You will see that no matter what I do Sync will not allow anything to move. By the time I recorded this video there were no locked dimensions anywhere in the model, every relationship in Live Rules has been switched off and still nothing at all moves in any direction at all. Worse, the Solution Manager makes no attempt to provide any useful information - just a notice telling me to unlock dimensions and relax live rules. Evidently it's not smart enough to know there are no locked dimensions nor does it know there are no relationship rules in force. More importantly, it's also not smart enough to know what the actual problems are either.
This seems to me to be the sort of thing sync should be able breeze through if it is to come anywhere near the claims made for it but pretty clearly still not much use when the going gets even a little bit tough.
Back to ordered and roll on ST8 I suppose...
Perhaps someone would care to comment?
PS - The 3D sketcher and the new measuring stuff in ST7 are very useful. Thanks.
Solved! Go to Solution.
Not a big user of it yet either for other reasons, but this probably has to do with all the rounding on the edges. From all the demos I've seen, it appears the first thing you have to do is remove the rounds, then perform the synch edit. This is why when hybrid modeling is discussed, rounds are recommended to be added in ordered.
I might try to recreate your part... unless you think you could send it to me...
Anyway, what's probably hanging you up are those tiny edge break rounds around the faces you're trying to move. Ideally, all the rounds would be ordered features while the main part faces are synch. You might be able to make it work by also selecting the faces of the rounds. There might also be something going on with the hole bottoms. You could select those more efficiently with a window select.
If you would like to donate that part to the cause, I'll make a little movie of those holes going where you want them.
As Matt stated, feel free to attach your model. I'm sure someone in this group would be happy to document a solution and provide feedback (including me).
Thank you for your replies everyone,
I'm home now (it's 10.30pm in UK) but I'll send the synchronous model tomorrow.
The point of my post was not really the specific problem in this model but why the "Live Rules" and the "Solution Manager" in ST7 are so irrelevant to this model - they gave no feedback at all, useful or otherwise.
I can certainly edit this model perfectly in ordered, but it'll probably take 10 or even 15 minutes to clean it up afterwards. If it's necessary to remove and re-apply all the rads, can synchronous edit it accurately at all (maintaining the design intent etc), and if it can, can it do it significantly more quickly? A challenge for you!
The other challenge for you is to demonstrate how you get the Solution Manager to tell you what is actually causing the problem with this model. (That's what it's for, right?)
Thanks for your interest, hoping to hear from you soon.
As promised, here is the synchronous version of the model - I'd appreciate any thoughts on why I struggled so much with it.
I edited the ordered version in about 15 minutes including cleaning up a number of errors.
Thanks for your help.
As others have said rounded corners can at times put a strain on LR's understanding what needs to be done.
What I did notice in your Pathfinder is you "Mirrored" one side. And with out removing this, your model isn't "perfect"
Watch here: http://screencast.com/t/bgKmzd2NqYL
So you may need to delete the Mirror Pattern on this, and the easiest way to do this is with a Boolean subtract... watch here:
As with History/Ordered model Synch still requires a few work arounds. And once you get to realize that neither approach is wrong, you finds ways to accomplish what you need.
Just don't dismiss Synch because of this type of challenge. This is a complex part and I suspect that it was constructed using outside parts that weren't perfect to begin with.
I also took a closer look at the part and while some of it can still be synchronously edited, I'm noticing some "issues" with a couple of the holes. The two holes on the left (upper and lower) are selectable as "holes". The upper right hole is also selectable as a hole but it contains an extra surface or cone (highlighted in blue). The bottom right "hole" is just cylinders/faces. These can be easily fixed by deleting them and adding/copying the respective holes from the left to the appropriate position on the right.
You mention the two parallel faces should be pushed out 2mm. Should the face in red be adjusted at all with this change or does it stay stationary?
I would agree with the speculation about the rounds. This particular test-case has the sort of interacting blend-patches which are best handled by doing a hybrid approach. Actually it was a model not dissimilar to this one that made us realize the *need* for a hybrid approach.
There are fundamental differences between Ordered and ST rounds. ST rounds work great at long as the interactions between rounds are relatively simple . . . when you get more than 2 different round chains interacting, I would strongly encourage using the ordered approach.
Actually if you pay *very* close attention . . . it works fine for almost 1mm of travel, until one of the patches fundamentally "changes" and subsequently errors out..
A hybrid approach allows for direct ST manipulation of the base features, while preserving the robust nature of ordered rounds.
I think of the hybrid approach like "a cake with icing" . . . the foundational shapes are ST "cake", while the "icing" is Ordered features. It works as smoothly as a normal ST move, and adds the robustness needed when blend patchs start bumping up against each other in complicated ways.
I would try *not* converting *all* of the model to ST . . . leave the rounds in ordered (creating a hybrid model), and attempt the edit again.
If you would like to post the ordered model, I can make an attempt myself.
Also . . . just a little extra tip . . . you don't need to specifically select the holes, they are "nailed" to the face, so they come along for the ride automatically.
Thanks for the feedback. Sorry for the confusion.
Good evening Bob,
Thanks for your interest.
The model was entirely created from scratch in Solid Edge Ordered,starting with ST6 MP8 and finishing in ST7. There are no legacy parts from any other CAD system although there are some part copies used for construction purposes from other stuff in the assembly. The synchronous version was made simply by converting the whole thing.
As you noticed, the model is symmetric and was modelled as one side only, mirrored and a few finishing touches done to features on the centreline.
If I interrogate the ordered version it is perfectly symmetric but there are clearly differences with the synchronous version. I had already noticed that the mirrored holes are no longer holes in the synchro version but merely concentric cylinders whereas the ordered version has maintained its hole information and you have shown other subtle differences which must have been introduced by the conversion process. I would have thought hole information should be maintained across a mirror whether it is synchronous or ordered shouldn't it?
I suppose I could have converted to synchronous before the mirror in the history tree and achieved a similar effect to that in your second video - I'll give that a try tomorrow.
PS - I've always wanted to make use of synchronous technology and have spent a great deal of time trying to use it - but the cryptic nature of the Solution Manager messages and the difficulty of trying to make sense of Live Rules in real world situations always drives me away in the end - life's just too short...
Thanks again for your efforts, I'm very much obliged to you.
Best Regards, David.