I need some tips for good model management. This is a challenge I know I can work around, but I would like to know how you take on this challenge.
Also please confirm if you actually use your technique in production or if this is a what you consider your best take on the subject.
I am working on a model and drawing for the ISIS I orb. It has a half dozen springs in the assembly and several levels of subassembly.
I have all the springs modeled as free-length (suppressed in the model) and as-used length (separate model).
The cross section on the drawing and in the model looks -perfect-...(!)
... and then I embark on the exploded view realizing I want to swap out the sprung springs for the unsprung springs.
My first instinct is to add the unsprung springs to the drawing so I can delete the suppressed components in the model (1) PDM BOM accuracy consideration) (2) file export consideration)
My second instinct is to manipulate the two versions of the springs on layers (...never like layers in Pro|E; better in Creo)
But there are inklings that says we can make this challenge work more fluently.
- What comes to mind is use of family tables for the springs but this would mean extra parts in the model (doesn't solve BOM/PDM issue)
- Combined states should be able to solve this dilemma but the master rep would have both lengths shown.
- Flexible components cannot change length from one view to another; not much help.
- I could take meds for my OCD and get on with life.
There are two situations also under consideration but are less obvious to a solution but worth considering.
- The assembly model exploded view -should- swap out the sprung springs with unsprung springs when viewing the exploded view. This is a candidate for view states.
- Also, detent springs actually have 3 states; unsprung (for part drawings), installed (for subassembly drawings), and actuated (for top level drawing).
Besides the coil springs, there are also 3 other components that have a sprung and unsprung state. The total is 10 flexy components in a ~33 component assembly.
How would you tackle this challenge if your employer was a stickler for accuracy?
This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.
Today was able to make a flexible spring tied to 2 assembly parts using the direct driving dimension in the spring tied to "distance" in the flexible dialog from the component placement dialog. The spring is tied between 2 pins and flexes in length and adjust its angle according to the mated components. Finally! ... got one thing to work as expected. Why I can't drive parameters this way is beyond me.
Expansion spring looped on grooved pin.
- Coincident axis of pin 1 to axis on spring (adjusted for tangency; not constrained tangent)
- Lengthwise plane of spring coincident to the axis of pin 2 (to accommodate tilt)
- Distance between the hooks of the spring set to "distance" between the two pin axes drives the spring length dimension.
Some visual aides...
No family table instances in the spring; no part duplication in the assembly; will follow assembly family tables for use in drawings.
OUCH OUCH OUCH ! ! !
SERIOUSLY ? REGEN ?
We need the little puke smiley so we get a green dot up there.
Q: "How many time to you have to regenerate to turn off the stupid little yellow light?"
A: "WAY TO MANY!"
So a flexible spring with 20 turns x 2 is some serious overhead in a 3-instance family table assembly.
Lesson: Move the flexible spring to the end of the tree and move the "insert here" above it while working on the assembly.
Regeneration was an absolute nightmare and time sink.
I was dealing with huge corruption issues on the assembly file. I replaced one of the springs and the squishable o-ring and the family table just went on a wild tangent. None of the states were accurate; zero positions for mechanism constraints became arbitrary; Regen warnings were off the charts.
I ended up deleting the family table and removed all flexible models and unconstrained everything (~20 parts). I started over! All new constraints, all new flexible components; all new family table - 2 hours!
Good news: No errors in regen and all the instances were now stable and reliable. Uuuugh!
...and I haven't even got to the drawing yet!
Next, my assembly drawing has 3 instances and the generic assembly. Although my assembly way perfectly happy now (green light), the drawing has all model instances displaying yellow lights. Regen, regen, regen... more regens! and finally... whoohoo! green light 
...though very short lived.
I had to set up an exploded view in model, of course, more yellow lights. I had to create some view states... more yellow lights. Yellow lights and "model has not changed..." huh? More regens!
And when it does regen, those springs are 100+ regeneration features. x3 ...x2. Easily an hour of regen time in 2 days work. Lesson: learn to manage your regen times.
So what did this teach me?
Allow for extra time to manage dynamic assemblies. Lots of extra time. in a 20 hour project, easily 25% of the time was managing the various dynamic aspects. Okay, fine. I can compensate for this as long as I know to consider this.
Again, I am not a novice at this. I have been using Creo nearly every day since it came out. I can make some pretty nice little mechanism widgets; flexible components are fairly straight forward; and even family tables are fairly basic. And I have 2 projects (1 personal that started this thread, and now a client model) and the work on the personal project really helped me manage the client project more effectively. But the whole system is highly sensitive.
So what about the fallback option?
Yes, the extra model concept. Could I have done this faster with multiple models? A spring for each; an assembly for each; 3 separate assemblies; no mechanism constraints?
Honestly, right now it is a trade-off.
The reason I say this is that during dynamic development requirements, the integrated system is extremely touchy. For a family table to "break" from replacing components in the assembly is not acceptable. Having mechanism constraints move their zero translation position is unacceptable. Having flexible components loose their "distance" values is also unacceptable. Does any of this sound familiar to any of you?
Okay, so I am venting... And I know the best teacher is experience, and I am not yet experienced in all aspects of using these very powerful features, but again, I am not a novice. It really shouldn't be this hard to create a sustainable dynamic model.
Back to regens... what is up with all these yellow lights? We need a button that says "regen all levels". This is most frustrating in drawings.
And... anyone else having PDF generation quit and then succeed the next time? This is getting to be prominent issue. Almost like it times out.
Wow! Sorry to hear that. I only had one or 2 springs in my assembly, and a fairly small one, so I noticed no issues with that, thanks for the heads up!
The parts for my design just came in, and I just got back from my trip, so I won't be on here much, but I'll try and help where I can.
I TOTALLY agree on the regen issue, it's been a pet peeve since I started on Pro/E.
We cancelled our trip to Snohomish since my friend was having major back issues, but I'm kind of glad we didn't make it, as I think if so I'd have worn out my tire somewhere in WY on the way back with those extra 800+ miles on it..... 
Best of luck!
