Community Tip - Visit the PTCooler (the community lounge) to get to know your fellow community members and check out some of Dale's Friday Humor posts! X
How to add parameters with units using ModelCHECK? We want to avoid that users, after having run ModelCHECK to add missing parameters, still need to go into the Parameters table and select the proper unit via the dropdown lists. If parameters are not "unitized", then we get a checkin error in PDMLink. Thank you for your help! |
---|
I've worked quite a bit with ModelCHECK and I don't remember any way to choose the Unit for a parameter.
If nobody else has better news, perhaps look into whether or not Toolkit can do it.
Not an option in ModelCHECK to add units, you may try using mapkey.
Thanks for your answers.
That is correct when using straighforward ModelCHECK functionality.
However, I would be interested in knowing if someone has worked around it. We have tried, for example, making ModelCHECK add missing relations in this fashion: d1=5[mm], but we are unsuccessful so far.
Mapkeys would be a possibility, but in our case it is not so convenient because of technical reasons which are not worth to discuss here.
Within the parameter table is the ability to Load Configuration. I'd look into whether or not you could import parameters with units using this option. I'm not sure if ModelCHECK has the ability to load from file like this. You'll have to read the manual to see if it's even possible to do without a mapkey.
Thanks, I'll look into that possibility, although it's unlikely.
I know the manual well and still surprises me that adding a missing parameter WITH ITS UNIT is not possible, and it's also impossible to work around it, even using ModelCHECK's "Missing relations" capability and assigning that parameter a unit using a generic equation. To me it would be an essential thing to have: engineering parameters must not only exist but also have their proper units!
I copy the manual below, this is what one can do with parameters in ModelCHECK (and nothing more than this):
The following information can be verified in parameters:
• Are they the right types (string, integer, real, Boolean)?
• Are they PDM designated?
• Do they have assigned values?
• Do specific parameters have values from a standard list of valid values?
• Are numerical parameters less than, greater than, or equal to a specific value?
• Are string or Boolean parameters equal to a specific value?
• Do string parameters contain specific letters, or are they a specific length?
One thing to consider is the origin of ModelCHECK. It was developed a LONG time ago (by Rand Corporation I believe). PTC then acquired it and began packaging it with Pro/ENGINEER.
I'd guess that you need the Units in your parameters because of some functionality in Windchill PDMLink (or whatever system you are using for ERP/MRP).
Because ModelCHECK was developed before Windchill PDMLink was even a thing, and because the Units for a particular model are set in another location than the parameters, there wasn't a need to include things like Parameter Units in the ModelCHECK routine. (I don't remember if parameters even had units all this time or if it was something added after ModelCHECK was developed.)
ModelCHECK is also just that - a way to check the model. Because the model's Units are elsewhere than the parameter interface, and because the model's Units can be verified by ModelCHECK, there probably wasn't the foresight necessary to determine that parameter units may be required for anything outside the model.
In the intervening years, PTC has done little to no development of ModelCHECK. I, for one, am relieved they haven't touched it. Too many times I've seen an inadvertent 'improvement through destruction' happen. (Not just with PTC; with many other software as well.)
Thanks for the observation.
Yes we use PDMLink and naturally we want certain model parameters to be stored there.
I don't quite agree that ModelCHECK is just to check, since it guides you to correct the errors or even one can automatize many corrections using the ModelUpdate mode, but anyway, that would be a "philosophical" question 🙂
I also think that a small development would help, provided that they don't screw it up, naturally. But I guess it's almost too late, after so many years not minding it.
My point wasn't that it's just to check, rather that it's just the model. Should have emphasized that.
I'd be very fearful of any update they might attempt.