So you open an assembly that you know needs to be regenerated (in my case because I have a master part that I have modified which is referenced into about 20 child parts). How do you stop the assembly from regenerating? Ever since I have used ProE that "stop" button comes up and you are supposed to be able to stop regen by clicking on it. And ever since I have used ProE all it does is say, "Do you really want to abort regeneration?" You click on "yes" and it ignores you and continues regenerating.
I seem to recall three possible responses to the 'abort' query, two of which didn't prevent the regen and the least obvious (naturally) was the only one that did. But I can't remember what the three are or the proper response. I do recall, though, that I have successfully suspended the regen (or did I just quit the retrieve). Short of suspending the regen, I've taken to setting 'freeze_failed_assy_comp' to yes so that changing component geometry (or any other cause of damaged constraints) doesn't throw one into assembly resolve mode. Frozen components are a lot easier to deal with. This is also one of the best arguments I know of for using skeleton models, publish geoms and copy geoms for transmitting feature references between models: with a skeleton, it can be done without regen failures.
how do you tell frozen components from normal ones later on without deep digging into some info/component/etc. dialog?
Whenever I find some unexpectedly frozen component during a later redefinition it tends to be - just a little - out of place, which is totally inconvenient of course.
Funny that a "freeze" option is handy but there is no such thing as a "thaw" button! And like things frozen in a glacier, ProE seems to allow frozen components to slightly move...
I confess I still don´t fully understand all the concepts in ProE, though I have been using it for quite a couple of years now.
You can set your model tree to show the status of assembly components. One of the think it will show is 'frozen'. Personally, I think this should be a default 'out of the box' setting & have told PTC so, precisely because so many small companies don't have the time to hire someone to look after their system full time, & so end up using the decidedly ropy default settings. Think about it: Microsoft office works pretty well out of the box, so it can't be that hard.
If a component is frozen then it shows up in the tree. I think there's a couple of little bars next to it. My problem is not that the component's assembly needs to be redefined (the parts in question are usually assembled by default because they reference a master part), it's that the feature within the parts need to be regenerated. In the example I gave above they usually regenerate successfully, it's just that it takes forever and I'm only interested in the first 3 or 4 components. What I'd like to do is go into insert mode up to those first few components but I can't do that until it's all regenerated.
Anyway, now that it's regenerated I have suppressed all the stuff I don't want ATM. I guess I just need to think ahead.
Open / Open Rep / Create New ... Exclude top level entity ... go from there (i.e. Include 'master' and components you want to regen) might help the situation?
The Search tool (binoculars) is great for this. Select Component in the 'Look for' box and click on the Status tab. For the Criteria Value, pick Frozen, Packaged, Suppressed or any other condition. If you leave the 'Include submodels' box checked, it'll burrow into every level of the assembly. Pick components from the found items list and 'arrow' them into the selection list. These will expand and highlight in the model tree, making them easy to identify and deal with.
If a component is frozen then it shows up in the tree. I think there's a couple of little bars next to it. My problem is not that the component's assembly needs to be redefined (the parts in question are usually assembled by default because they reference a master part), it's that the feature within the parts need to be regenerated. In the example I gave above they usually regenerate successfully, it's just that it takes forever and I'm only interested in the first 3 or 4 components. What I'd like to do is go into insert mode up to those first few components but I can't do that until it's all regenerated.
Anyway, now that it's regenerated I have suppressed all the stuff I don't want ATM. I guess I just need to think ahead.
________________________________________
Well, I don't think that going into Insert Mode is a bad idea, providing you think to do that before you change the Master Model. And Simp Reps, even on the fly, will ease the business of opening the assembly. I was even wondering what would happen if you opened the components before you opened the assembly, would that ease the regen burden. Problem with these scenarios is that you're eventually going to have to regen these parts that are taking so long to regen. But what really bothers me is why modifying the Master Model would cause so much grief in the first place. And no error messages!?! That's strange. Did any of the individual parts have difficulty merging/solidifying? One component, taking an inordinate length of time to regen, can slow down the whole assembly regen. That's the main reason I thought to open and regen individually.
ou think to do that before you change the Master Model. And Simp Reps, even = on the fly, will ease the business of opening the assembly. I was even wonde= ring what would happen if you opened the components before you opened the as= sembly, would that ease the regen burden. Problem with these scenarios is th= at you're eventually going to have to regen these parts that are taking so l= ong to regen. But what really bothers me is why modifying the Master Model w= ould cause so much grief in the first place. And no error messages!?! That's= strange. Did any of the individual parts have difficulty merging/solidifyin= g? One component, taking an inordinate length of time to regen, can slow dow= n the whole assembly regen. That's the main reason I thought to open and reg= en individually.
In your assembly just above the tree, select 'settings' then 'tree columns' under the 'info' box select 'status' and move it to the right. this will show the status on all parts wheather they are suppressed, frozen, regened, children etc.. If you want then you can save the tree to load always. good luck, Gary
That's good thinking. Since the first post I have been careful to retrieve a Simplified Rep instead of the whole thing, but I forgot that I can make one on the fly.
ou think to do that before you change the Master Model. And Simp Reps, even = on the fly, will ease the business of opening the assembly. I was even wonde= ring what would happen if you opened the components before you opened the as= sembly, would that ease the regen burden. Problem with these scenarios is th= at you're eventually going to have to regen these parts that are taking so l= ong to regen. But what really bothers me is why modifying the Master Model w= ould cause so much grief in the first place. And no error messages!?! That's= strange. Did any of the individual parts have difficulty merging/solidifyin= g? One component, taking an inordinate length of time to regen, can slow dow= n the whole assembly regen. That's the main reason I thought to open and reg= en individually.
No, they didn't fail in this instance. That's because I'm goooood! ;-)
Actually it's because I added some geometry to the master, some curves or surfaces, which had no effect on existing geometry, but because it was the master file the child parts had to regenerate. And we know that when you do that the model can't possibly fail regeneration because that wouldn't make sense, now would it? :-P
The child parts won't regenerate unless the assembly is in session. I have done this before though. If the assy is in session I will open a particular child part and regenerate it, or I'll right-click on it (and maybe other parts) in the assy and regenerate those only. The OP situation only occurs when I have made changes to the master file and then I open the master assembly for the first time since the change.
"Horse's Ass" had the right answer - make a Simplified Rep on the fly. That's the way around it.
you´re right! The binoculars is one of the icons I tried once, not again because I didn´t get the concept in the first place. Pretty useful function, this one...
... do frozen components thaw by themselves eventually? Or am I supposed to look for them from time to time?
A couple points: 1.. To load each time, you must point the option mdl_tree_cfg_file to your tree.cfg file by browsing to it 2.. There are limitations to what it will show as frozen. For example, it will show a group as regenerated while every component in the group is frozen. You will have to show the tree in fully expanded form to see this.
3.. Same with subassemblies: the subassembly may regenerate while components within it are frozen, suppressed, etc, so, to see their status, expand all.
Pro/e assemblies are not experiencing climate change and spontaneous thawing. But, sometimes, redefining one will cause the others to suddenly wake out of their comatose state and assemble normally. But, as to what you do with frozen components in assemblies depends mostly on what you use the assemblies for. A drawing BOM is not affected by a component's frozen status. But it may be by other status anomolies (suppressed, suspended). And other functions like Mechanisms or analysis functions balk at frozen components. So, I've just gotten into the habit of maintaining clean assemblies, running Model Check and doing searches of anomolous components.
As Dave said, a frozen component will "thaw" "automatically" if its references are resumed in some other action. For example, if Component B is assembled to Component A and you suppress Component A, you can chose to freeze Component B (or suppress/delete it or redefine it). It will kind of hover in space without references. But if later on you resume A, then B will no longer be frozen, assuming the geometry is still there.
From theory, I´d agree. Practically, I run into some invalid refs of a component from time to time when I decide to manually redefine.
To my surprise ProE does not nag about it beforehand and I suspect this has to do with an earlier "freeze" - in order to get ahead to things I wanted to fix in a failed assembly primarily.
Forgotten´n´frozen components seem to stay in place until "thawed", i. e. selected to redefine... but, unfortunately (dangerously) not quite exact, though: there always seems to be a bit of drift.
Frozen components for me are always temporary. I always make sure everything is regenerated when it's going out, not because I'm a goodie two shoes, but because if it's not then I'll have egg on my face later on if something doesn't fit properly.
The small amount that they go out of alignment is quite mysterious. When the assembly first opens and regens, if you look closely, you'll see a message that says "Component ID X placement references missing. Using previous placement". If it is, as I suspect, using some offset from the default assembly csys, how could it be off at all, much less by the small amount that it's off? I don't get it either, but it would be just enough to do a major screw up of an interference check with fasteners in place. Wierd and just enough that I suspect Pro/GOOFY, the built in joke module of Pro/e. And you thought it was just random strangeness. Like converting unitless factors when changing systems of units. Stuff you'd hardly even notice but that cause a ripping good time when you do. Yup, lotsa laughs.
that´s what my boss would tell me ... if I had one :-) In reality, deadline is always close and I *have* to brush aside minor nags.
With insert mode activated in sub-assemblies the freeze nag appears far to often to not ignore it in order to get ahead to the real work.
Later on, I´d like a button "thaw all frozen components", or, better: automatically suggest refs to realign previously frozen components to. I frequently find that refs of dangling components are still there, but somehow ProE missed them and wants additional point´n´click work.
The hidden Pro/GOOFY module some frustrated outsourced el-cheapo coder freshman has hidden inside ProE´s Advanced Deadline Thwart Package (as David suggested) explains a lot... many thanks! I needed a laugh.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.