contoured surface as a Datum ( like a windmill blade )

[blinger]

We have a print that is calling the contoured surface of the blade ( like a windmill ) a datum.
Is this feasable ? How would you program a CMM to have this surface called a Datum ?
Does anyone know if the software ( PC-DMIS CAD ++ 4.3 )would accept it ? I don't quite
how a Datum on a contoured surface would work.
Any help would be appreciated.

 

[MikeHalloran]

I find it helpful to think of a datum as a feature of a fixture, not a feature of the part.

So, theoretically, you could machine a fixture to have a surface perfectly complementary to a perfectly contoured blade, rest the part on the fixture, and measure deviations over the entire surface.

For a physical fixture, that is a practically impossible thing to do, at least in part because it can only cover the case where the part surface lies entirely to one side of the theoretical surface.

For a CMM, I think it's less impossible, or slightly less impractical, but I think you need to call the theoretical surface a quaternary datum (if that's the right GD-speak), and use some features of the blade root interface as the primary and secondary and tertiary datums, so the inspector just has to bolt the blade to a simple fixture, and then map the blade's surfaces relative to those blade root datums and the theoretical surface.

... as opposed to first mapping the blade's surface and trying to then derive where the datum might best fit on or in it in such a way that the blade root features stay within tolerance, which is sort of what I'm reading into your problem statement.



Mike Halloran
Pembroke Pines, FL, USA

 

[geesamand]

I think you'll need to use datum targets.

We do inspections of propellers and in fact we need a combination of hard machined datums and datum targets, or simply use a number of datum targets and a best-fit solution from the CMM for alignment.

David

 

[blinger]

Here is a print of the part i was talking about as far as having a curved Datum ( B )
See the attachment below. As far as i can see there are no target points. The only thing i can think of,
as far as checking this with a CMM is to take points below each circle and constructing a Datum B plane on the
bottom. I'm looking for any ideas as far as to checking this part on a CMM.
thanks guys for your help so far.
Glen.

 

[ewh]

This may not help you with the CMM setup, but for what it's worth, contoured features may be used a datums.

AMSE Y14-5-2009 ¶4.13:
"It is sometimes necessary to identify a compound curve or a contoured surface as a datum feature. A mathematically defined feature shall be defined within a three-dimensional coordinate system. Where such a feature is specified as a datum feature, its datum feature simulator (derived from the math data) is used in establishing the datum reference frame. Aligning the high points of the datum feature with its datum feature simulator restricts movement of the part to the datum reference frame. Where the datum feature alone will not adequately restrict the required degrees of freedom of the part, additional datum features will be required."

I am more familiar with the setting up of the primary, secondary and tertiary datums and referencing those when establishing a contoured datum.

“Know the rules well, so you can break them effectively.”
-Dalai Lama XIV

 

[DaSalo]

Sorry for such a late response. You have likely already solved your problem but for what it's worth:

This kind of thing is a walk in the park for CMM software like Pc-Dmis. Just do the following:

- Use a Rayco plate or other type of flexible fixturing to build something to support this on the CMM table. The fixture should be fairly repeatable but doesn't need to be perfect. As long as it will hold the part steady, allow it to remain in a free state, and locate it in the same spot each time within +/- 2mm or so you should be all set to go.

- Next I would do an iterative alignment to 6 points: 3 on the contoured surface, 2 on datum C, and one on datum D. Iterate 5 times against a target radius of .015 or so and you'll be darn close.

- Now, you still haven't met the print requirement so you have one more alignment to do: You now need to take a point cloud covering the entire datum A surface. You have CAD++ so take advantage of the scanning options to make quick work of this.

- Now that you have that surface mapped you will use the best fit alignment tool to perform a 3D least squares or vector best fit to that surface. Now do another best fit, 2D this time, to a line measured along datum C. Constrain all translation so all it can do is rotate about Z (Rot about X and Y are constrained by virtue of it being a 2D alignment.) And finally, a best fit to a point on whichever end of datum D is sticking out furthest (corner that would hit first if sliding it along a rail contacting datum C until datum D hits a perpendicular plane.) This last best fit should allow translation along the datum C axis only.

- Done! Anyone with solid Pc-Dmis training (your CMM programmer? I hope...) should be very familiar with this process.

- To summarize:

- Iterative alignment to 6 points to get close.
- 3D best fit to datum A with all degrees of freedom permitted.
- Best fit to datums C and D with the appropriate degrees of freedom constrained.

Hope that helps.


CATIA V5 R20
PC-DMIS 2011 MR1