how to find the center of a circle resulted from a sweep cut 1

[godpaul]

NX 8.5

after the sweep , i want to find the center of the circle in the last sketch (picture), and that center will be used as my reference for other feature, how to proceed?

thank you

pic

http://files.engineering.com/getfil...-2bcc-49a9-9715-e3b55006bf7e&file=Capture.PNG

prt
tp://files.engineering.com/getfile.aspx?folder=0a6bf0f2-3835-484d-bea9-293dcf502b57&file=how_to_find_center_of_a_circle.prt

 

[JohnRBaker]

Since the resulting edge of the opening is NOT a circle you can't simply create a point at the Arc Center. However you could use an analysis tool to find the 'centroid' of the opening. To do that, go to...

Analysis -> Advanced Mass Properties -> Area Using Curves...

...and select the 'Boundary (Temporary)' item, then select the edge curve of the opening, hit OK, accept the default tolerance, select the 'Create Centroid/Axes' item, toggle ON the 'Model Centroid' option and hit OK and then 'Cancel'. After refreshing your display you should find a point object at the 'center' of your opening which you can query to find the coordinates of or which can be used to dimension the location. Note that the point created is NOT associative to the opening so if the model changes you will need to repeat this procedure to get the correct point.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[godpaul]

thank you Mr. John,

after posting the question, i tried again and again, and find a way but dont know if it is good.

since associativity is VREY IMPORTANT in my model, i am very careful in term of building sketch and assign dimension. here is what i did:

1. use project curve to get the "circle"(it's already in spline i know)
2. radonly add 3 three point on the "circle"
3.draw a sketch using 3 point method.
4. all done!

if i change the dimension of the cirle in the very insitl sketch, the geometry will be updated as well

good or bad?

prt

http://files.engineering.com/getfil...ile=how_to_find_center_of_a_circle_solved.prt

 

[JohnRBaker]

Yes, that will work as well but you don't need to create another sketch. Simple go to...

Insert -> Curve -> Lines and Arcs -> Circle Point-Point-Point...

...and with 'Assocative' ON (you can toggle it ON/OFF at the top of the list of 'Lines and Arcs' that you'll see during the last step above) and your 'Snap Point' set to 'Point on Curve', select your three roughly evenly-spaced 'points' around the opening and now you'll have an associative arc that you can create an associative point at the center of.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[cowski]

Another alternative is to calculate how much the original section (circle) will rotate based on the helix pitch and distance between the surfaces, use instance geometry to create a copy, then project the circle up to the top surface. It should work as long as your surfaces are planar and parallel.

www.nxjournaling.com

 

[JohnRBaker]

Yes, but either the OP's sketch idea or my three-point arc will account for not only any change in the path and/or block thickness but also changes in the size and location of the original arc.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[godpaul]

you always have nice tricks!!!

 

[cowski]

Yes, but either the OP's sketch idea or my three-point arc will account for not only any change in the path and/or block thickness but also changes in the size and location of the original arc.

I believe my alternative method would/could account for those variables as well, could it not?

www.nxjournaling.com

 

[JohnRBaker]

Perhaps, but it just feels a bit too complex when compared to the more direct methods proposed, don't you agree?

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[cowski]

Perhaps, but it just feels a bit too complex when compared to the more direct methods proposed, don't you agree?

The "area using curves" method is unassociative. The "three point arc" method is an approximation (albeit a good one) that should be associative; but in my experience, I wouldn't be surprised if it errored on large input changes. If this is going to be used as a seed file (as the other threads seem to indicate), I'd go with the most robust solution.

www.nxjournaling.com

 

[godpaul]

i tried both of your method

the helix calculation is actually a strategy i used to determine how much angle does the circle rotate after a certain distance, but i never realized your method can be used to locate the center...

anyway, thanks for making my model robust!