Sketches and circular array conflict

[BobvdBroek]

Hello,

My name in Bob, have been reading this forum for some time and its information has been helpfull on many occasions. But now I have a question my self:

It seems that circular array overrules sketches that are consumed in its creation proces. For example.

I have a model of an cillinder, diameter 50. With a fully constraint sketch I create a hole on 15mm of center of the cillinder. This sketch is then used to make an array of 4 holes, with the center of the array placed on the center of the cillinder. Now there are 2 place were i can change the pitch cirkel with 2 parameters, the constraints in the sketch and a parameter in the array feature. This is very confusing to me, but i can wrap my head around this. You can for example leave the starting hole position uncontraint and then use the array to control the pitch cirkel.

Bus it starts to get weird when you start changing the parameters. If you change the distance in the sketch (example to 10), the pitch chirkel changes (to 20) but the parameter in the array stays the same (15). Other example if you increase the pitch cirkel in the array (to 20), the sketch doesn't change (stays 15). Even if you increase the array parameter with (to 20) and decrease the sketch parameter with an equal amount (to 10), the diameter of the pitchcirkel in the model doesn't change (stays 15). So its like non off the parameter is leading or overruling the other. The center of the pitch cirkel doesn't change during the entire process.

Does anyone have noticed this behavour and offer een explenation to why en how this is happening?

(ps. I myself prefer using a sketch to draw an pattern, but i like this explaind anyway)

 

[Toost]

can you supply an example ?

Regards,
Tomas

 

[BobvdBroek]

This is the model descriped in the post.

Parameter P98=10 is the sketch dimension
Parameter P102=20 is the radius defined by the array feature

The pitch cirkel is 30 in the model

Regards

Bob

 

[Toost]

I have never noticed this before.
It seems as soon as one starts playing with the "Radius option" in the circular array, unexpected things happen. As long as one stays away from touching that value, all is ok.
To me, the original sketch carries the parameter that should be used to drive the radius, not the array feature.

Btw, use a datum axis for your circular arrays, it will then be associative to that axis.
If you use the point /direction method it isn't associative to the point/axis.

Regards,
Tomas

 

[JohnRBaker]

Yes, that is a bit odd, but I wouldn't lose a lot of sleep over it since the 'Instance Feature' functionality in NX 7.5 has been replaced by a totally new 'Pattern Feature' in NX 8.0. And if you were to create this same model with NX 8.0 using the Pattern Feature capabilities, then the dimension (expression) in the Sketch would be the only thing that controlled the so-called 'radius' of the Circular 'array', just what you would expect that it did.

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.

 

[BobvdBroek]

Thanks for the reply's. I wasn't on planning on this to spoil my night rest.

I'm a key-user with the company and i'm confronted with these kind of questions. I like to stay clear of the response 'it is because it is' or 'in the next version it wil be beter' because that doesn't help my collegues. We just switched to 7.5 and it looks like the next couple of years there wil be no upgrade. It seems that this one going to be filed under 'UG peculiars'.

 

[JohnRBaker]

Yes, I think that what we have here is a 'conflict' in what the model thinks is 'constraining' the location of the holes. The sketch says one thing, but the Instance has been told to do something else. Perhaps this might be an occasion for when it might be better for all concerned parties if you had left the Sketch un/under-constrained with the knowledge that the Instance Array feature will be the actual mechanism assuring me that I will get what I want and still be able to control it adequately.

And as I alluded to, since this has already been addressed in what are now considered to be the current releases, NX 8.0 and NX 8.5, there is little to no chance that anyone will consider fixing this particular NX 7.5 'peculiarity' ;-)

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.