Feasibility of inspecting large radius with short arc length 4

[DoogieP]

Hello all,
Can anyone shed some light on the feasibility of inspecting the 4.375R on the attached drawing? We have an OGP smartscope and the operator is using video and\or laser scanning to measure the radius. What I'm trying to convey to him is first, I think he should be hard probing this feature. Second, because of the short arc length (about 17 deg)it's an insufficient amount to get an accurate reading. Our optical comparators you need atleast 30 degrees of arc to measure accurately. Thanks in advance for your help!

Doug

 

[MikeHalloran]

Physical templates and feeler gages would make sense to me.


Mike Halloran
Pembroke Pines, FL, USA

 

[TheTick]

Might be better to control as a profile than radius.

 

[DoogieP]

Thanks for the feedback.These parts are going to be checked by the end customer using a CMM so, I'm not sure if a template and feeler gage will cut it. This is a medical implant and they want to see as much hard data as possible. I agree that this should be controled as a profile because as it is currently dimensioned it is a form tolerance. The problem is if you lay it out, you only have .0001" total on form at the outer edges of the radius, see attached. It's hard to see the two arcs but, both are tangent at center line because the radius' location is dimension to the tangent. As you go towards the center of tangency you lose tolerance. Being this is for medical, we may not be able to have the customer change this at this point because of all the FDA red tape they would have to go through. Perhaps the wording of my question wasn't specific enough. I wasn't asking how to check it but, if we can realistically check it. Because the radius is faily large compaired to the length of arc and based on the tolerancing definition of a radius dimensioned with an R in ASME Y14.5M-1994 (drawing references this standard)it becomes a form tolerance. Our machine's advertised positional accuracy is .00016" so, that being said the tolerance has theoretically exceeded the capability of our machine. I'm looking into it right now but, I'm sure we are probably at the edge of our inspection equipments capacity as well.

 

[pmarc]

DoogieP,
Google for "Fundamental Good Practice in the Design and Interpretation of Engineering Drawings for Measurement Processes". When you find the document (it is 99 pages long and fully free), look at pages 64-66. I think this may interest you.

 

[SeasonLee]

Good search and excellent reference.

Season

 

[DeanD3W]

The feature control will not be successful, in my opinion, unless the red tape is dealt with in order to change the drawing. Profile of a surface will not consider a radius value or a center location, the only evaluation will be whether points on the considered feature are within the profile tolerance zone. If he feature must be oriented and located then datum features will be included, of course. Based upon what I am reading here I think the answer to whether the radius can be checked or not is "no". This makes me think of the other thread that asks why drawings and GD&T are done so poorly (the bad drafting thread). This is one of those cases where time must now be taken to do things a second time, since they were not done right the first time.

I don't find pages 64-66 of the "Fundamental Good Practice in the Design and Interpretation of Engineering Drawings for Measurement Processes" document helpful because it seems to focus on minimum zone vs least squares algorithms while neither one is useful in this case. It's good about pointing out the issues with partial cylinders, but where's the discussion of how a profile tolerance zone with the profile tolerance and tolerance zone constraint based upon functional need will provide well without any of the problems associated with finding a radius value and center location?

Dean

http://www.d3w-engineering.com