linear tolerances of imported solids

[Lorenzo77]

Hello everyone, it's kind of bothering me about managing linear tolerances in nx and I wanted to ask you a couple of information.

first of all I wanted to know if there is a command to know the tolerances of a solid in a model.

the second question concerns the importation of maths which do not know tolerance, in fact normally I work with nx to 1e-4 and always amount to 1e-4 but not knowing the tolerance of the imported file I was wondering whether the tolerance was converted both positive and negative and the imported solid used the import tolerance value or maintained that of origin.
(e.g. iges model with unknown tolerance but actually is: case a) 1e-3, b)1e-5; 1e-4 work model and import parameters set at 1e-4.
how will you take the imported solid into my model in case a, and in case b? )

Finally, I ask you about the operations on solids/surfaces with an adjustable tolerance parameter, the example case may be: they are in a set and I have two models, the model with solid 1e-4 tolerance and a model with solid b to 1e-2, I must follow a boolean operation of a-b subtraction with tolerance set to 1e-4, the solid to now that tolerance has? and if the tolerance of the operation had been set to 1e-3 because with tolerance 1e-4 did not go well now that tolerance has my solid to?

Thank you.

 

[Utente NX]

Since you talk about imported solids, once imported and recognized as solids, tolerance and standard nx, then from them you don't have to worry anymore.

 

[IPdesign]

I do not want you to consider modeling tolerance as a tolerance that is usually indicated on the tables.

 

[ceschi1959]

I do not want you to consider modeling tolerance as a tolerance that is usually indicated on the tables.

Hi.

the problem reported by lorenzo (we talked about it myself and him) is much more serious and for nothing trivial.
is the most serious case when he has interlocutors that when you talk about modeling tolerances answer you wath is it???? (naturally in narrow Bergamo). so the file that should possibly change and fresare does not know how it comes out. we have made very significant import and export tests.
a possible solution would be to understand in advance with what tolerance the various formats are exported: iges, step, parasolid.
It seems to me to remember that there was already a half discussion about this where the fields were indicated by opening iges or step with a text editor.
Do you remember anything?

Hi.

 

[avalon3d]

from my point of view in solid modeling does not affect so much tolerance (by talking about mechanical values): If we find mistakes in the cad for tolerance modeling with solids means that something is wrong.
with surface modeling instead, close to 1 micron better guarantees the "goodness" of the work to which I know... (Czech, correct me, please)
However, here we are talking about tolerances that are not even perceived by machines ut to here, mathematicians are fed. . .

 

[avalon3d]

Oh, sorry!
Hi, hi, hi.

 

[IPdesign]

Hi.

the problem reported by lorenzo (we talked about it myself and him) is much more serious and for nothing trivial.
is the most serious case when he has interlocutors that when you talk about modeling tolerances answer you wath is it???? (naturally in narrow Bergamo). so the file that should possibly change and fresare does not know how it comes out. we have made very significant import and export tests.
a possible solution would be to understand in advance with what tolerance the various formats are exported: iges, step, parasolid.
It seems to me to remember that there was already a half discussion about this where the fields were indicated by opening iges or step with a text editor.
Do you remember anything?

Hi.

I remember something:
the following is the first part of an iges file

1
1h,,1h;,28hc:\temp\xxx,26h and:\xxx, g 1
7hnx v8.0,27hsiemens plm software nx 8.0,32,38,16,38,16,28hc:\temp\paol0g 2
23c2ba4ig09.prt,1.0,2,2hmm,3,0.0,15h20111201.065238,0-005,10000.0, g 3
11,0,15h20111201.065200;

what I avoid in red is the value of export tolerance .
usually I to avoid problems amount with a higher decimal, but not always everything goes smoothly.
or sometimes I get mathematician with too high tolerances (e.g. 0.10 or 0.254 ). in that case I notice the customer that the geometry has been modeled with inadequate modeling parameters and there may be discrepancies between the imported geometry and the quoted 2d design.

It should also be said that as practice I always carry out a table of control of the models I receive based on the 2d provided.
It's a waste of time or useless work at first, but don't imagine how many times I intercepted modeling differences that had not been updated on 2d.

 

[IPdesign]

Oh, sorry!
Hi, hi, hi.

It was a pleasure to meet you in person yesterday at the nx8 presentation

 

[IPdesign]

...
However, here we are talking about tolerances that are not even perceived by machines ut to here, mathematicians are fed. . .

Let me not agree.
I happen to design molds that have dimensional tolerances even of millimeter cents.

when I receive maths with tolerance of 0.256 I cannot afford to feed them to a cam
I have customers working with 0.2 cutters on the tempered and such a hole for them means making a difference.

Of course if I cut truck bumpers I don't give a damn even if I have 4 tenths of opening on a surface.

 

[ceschi1959]

It was a pleasure to meet you in person yesterday at the nx8 presentation

too!

:smile:

 

[avalon3d]

too!

:smile:

I quote everything that says baskets!

the speech can be valid in the case of creations of molds, masks or measuring instruments, but if you manage the production of the pieces the tolerance does not serve so "closed". I don't talk about exporting to 0.254, but besides the penny normally you don't use.
a micron is, as I said before, the tolerance of std modeling, which should ensure a certain safety margin in all conditions.

 

[Lorenzo77]

I would like to clarify my/our processing needs; my mechanical company produces molds and 45% of the printed products require centesimal tolerances (high speed windows rather than optical fiber attacks to cite some products), of course for the processing machines and for the processing process at the practical act it is ready to maintain these precisions for which at the design level at least the thousandth of precision I think it is indispensable, to this precision they are added customers who work with complex surfaces and then require the restitution.
At this point I need to clarify the management of tolerances with nx and its import and export processes so as to be able to manage everything at the best by disconnecting me between models of suppliers, external designers, old mathematicians, models of customers and those who put it more.

 

[IPdesign]

I understand, I have your own problem.
I almost solved by asking them mathematicians in step, you get a solid model and almost always you can work
Some people don't want to understand it, but when you ask to recognize you for hours to fix geometry, see that iges don't dare to give them any more:-d

 

[Lorenzo77]

ip, even when I can send the models in step to overcome problems normally related to iges, but if I have to be honest while with iges I can understand the tolerance of solids by controlling them with an editor, with the step I do not know where to look to know the tolerance of models. Also not even with the steps I have the guarantee of having a correct model, sometimes it happens to find surfaces that lose the limits, or are missing or corrupt probably for a problem always linked to tolerances.
for the speech of giving us mathematicians with acceptable tolerances realize that the external design studies to which we support do not provide us models in prt (and also they have nx) but they provide us with models in step, the inability and the company menefreghism allows it ingorating the hours of work that go in addition to working with maths in native formats, and if on one side I can say that it is missing
heredi prepared the system to find tolerance in iges how do I know the tolerance of a step model? and how are nx tolerances handled with operations between different tolerance solids?

 

[avalon3d]

bhe, if imported files do not respect modeling tolerance, and you try a boolean, the default error that generates ug is "non manifold solid", even in case a piece of surface is not "closed".
I recommend that you use the examine geometry command, which you find under the analysis menu, setting the tolerance you want to analyze the imported solid if you have any doubts!

I hope I was useful.