Creation of Compound Radii Fillet

[seakins5100]

I am currently in the process of modeling an unshrouded axial wheel airfoil to be used for die creation. The customer requires a compound fillet between the hub of the wheel and the airfoil.

The larger of the radii must have its center located at a set distance from the 'platform' or inner flowpath of the part, and it is to be tangent to the airfoil. The smaller of the radii is then blended with the larger at the flowpath.

I've been able to recreate this profile in sketch mode, but I have been unable to use this profile to create the fillet I need.

Below is the profile I would like to use for my blend.

Any help would be greatly appreciated. Thank you.

http://files.engineering.com/getfil...file=UG_Help.JPG986184abaecb&file=UG_Help.JPG

 

[Xwheelguy]

Sounds like a job for Variational Sweep, possibly. Maybe use the large radius center as the guide/path and the 2 arcs (compound blend profile) as the section? Read up on this command in the documentation...it might be what you're looking for.

Tim Flater
Senior Designer
Enkei America, Inc.

http://www.enkei.com

Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.

 

[wmalan]

I would be interested to know how it worked for you. Our company machines bladed impellers and such. We find our customers supply us with models that are just patched surfaces to do this. These are from a MAJOR UG based company. Makes for extra work at the NC programming side...

--
Bill

 

[seakins5100]

I managed to perform the variable sweep with something very close the profile I wanted, but unfortunately it did not conform to the airfoil properly.

The three fillet types I have available to me are: edge, face and soft. I'll go back to trying to use one of these to get the job done.

I also forgot to mention, I am running NX4.0 with only basic licensing.

 

[Xwheelguy]

seakinds5100,

Another option would be to create a "helper" surface to which a Face Blend with the most surface area (not the largest value) is tangent, do not Attach this Face Blend to the airfoil yet. Then create a second Face Blend tangent to the first Face Blend and the second surface of the airfoil, again only let the Face Blend Trim, not attach. Finally, Sew all the surfaces except the helper surface then use Patch and see if you can Patch the Face Blends into the solid airfoil.

Tim Flater
Senior Designer
Enkei America, Inc.

http://www.enkei.com

Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.

 

[EDCSRPM]

Create an offset inward of the hub sheet per dwg reqt's
Thicken the hub (any reasonable value)
Unite the AF to the thicken
Blend with large fillet
Trim resuling body with original hub sheet
Unite body to wheel model
Instance the body with blend
Apply small blend using instance option

This also works with variable blends for either or both blends that comprise the compound.

Hope this helps

 

[seakins5100]

Thanks for all of the help so far everyone.

I am currently at the end of the process which you described TurbBldDsgnr, but unfortunately the small blend is now giving me issues due to the a small part of the geometry of the larger blend.

I'll let you know how this turns out.

 

[hudson888]

You could always post the model and see if we can't sort it for you. Can't see your jpeg, must be the chinese internet I'm using great fire wall thwy call it.

Cheers

Hudson

 

[Xwheelguy]

I thought it was the Great Wall of Fire?

Tim Flater
Senior Designer
Enkei America, Inc.

http://www.enkei.com

Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.

 

[EDCSRPM]

Seakin5100,

You might try a conic blend. The problem is that it is very near impossible to replicate the same results as the compound, but you can get close. I have used sheet blends and the patch method suggested by nkwheelguy. Main problem is discontinuities that can cause issues for analysis/Mfg CAM. An AF typically has a radial edge somewhere. If that occurs at the extreme LE or TE it can attribute to the blending problems due to a very complex set of vertex generated where the two face edges meet. You might try using the stop short blend option to end blend before problem area. You could then try various freeform methods to close the gap.

One more possibility is extract some V parameter curves on the AF face and sub-divide the AF face using them. At minimum, this may help you trouble shoot the blend issue. Isolate the LE and TE first.

 

[hudson888]

On rare occasions when blending with radii won't fully describe the geometry that I require the technique is to use blends of curves offset from the edges to build the blend tangent curves that I require and surface in between those to create the blend that I need. In doing so you often need to stratgize to avoid three sided surfaces at the corners becuase they're really hard to build using manual surfacing techniques. The object of the exercise would be to create a set of nice tangential surfaces that desribe the blends you want, then sew them together and patch them on to your solid model.

Cheers

Hudson,

 

[seakins5100]

Thanks again for the suggestions guys. I would give more details or even the model, but this is a military part. Here are a couple photos to give you an idea of what I have on my hands - looks like you're spot on TurbBldDsgnr. The pictures shown include the larger radius offset from the flowpath I mentioned earlier.

It's been a little hectic here lately, so I may not get back to this immediately.

 

[hudson888]

I don't know if it will help but have you looked into face blends using tangent curve controls. It appears to be the root radius of a turbine blade am I correct? Earlier I had envisioned a tip radius, but from memory relatively few blades have any radius on the tip end.

To be honest your images aren't screaming at me what your problem is.

Cheers

Hudson

 

[Xwheelguy]

You're not alone Hudson...and I'm really interested in what's going on around the bottom of the blade with the odd looking edges coming to points. I initially thought this was a blend from the blade to the shaft, but I don't see one there, unless my eyes are playing tricks on me again or the images are prior to the blending.

Tim Flater
Senior Designer
Enkei America, Inc.

http://www.enkei.com

Some people are like slinkies....they don't really have a purpose, but they still bring a smile to your face when you push them down the stairs.