Checking Radius

[AuRules]

Got a tricky one fellows, how would you go about checking these 2 radius I have pictured in the attachment I need size and location. The bottom part of the picture is like looking at it from the top, then the top picture is looking at from the side if it was cut in half. I would like to create a diameter at a certain height on both radius and do it that way but I don't know how to get there.

[SWomack]

I would have the center of the part the Origin (naturally). Then use an auto circle to check it just like in the section view. To measure the R.400ish, take the diameter of the ID and divide by 2, then subtract the radius size, that's your "X" value. Then take your depth of the pocket minus the R.400ish radius and that's your "Z" value. Change your vectors to "0,1,0", adjust the start and end angles and viola, it should be finished. Rinse and repeat for other radius. You could do it on both sides to make sure it's concentric. Hope that helped.

[Rondog]

You could construct an intersection point between the two sides and then measure a single point on the center of the radius. This method is like using a rad gauge.

The only problem with this method is that you don't actually get a real radius. You would need to do some math to calculate this out at it treats it more like a profile.

[JEFMAN]

Don't you have to measure it as a part of torus ?
If you have to dimension radius only, I think Rondog is right, you should measure it as a profile...If you construct a mid line between each side of a radius, the center should be on this line. Then you offset a // line from the bottom, you construct intersect point and it should give the theo center.

[frazz]

Hoepfully those rads are actually important and not just called out because some engineer went overboard and doesn't understand the concepts behind what and why things get called for inspection and toleranced the way they do (not that it ever happens in any shop anywhere....). I know this isn't anything new to add to a solution for your problem, but do what Rondog said. +1 - AND REMEMBER TO MEASURE FOR SUCCESS!

[DaSalo]

Also, remember that there are other tools available than the CMM. If you really need a very accurate measurement of a radius (for example, you need to very both it's value and whether it is a "controlled" radius or not) some reproduction compound and an optical comparator are very valuable tools. I doubt you need to go that route here, but sometimes the CMM just isn't the way.

[AuRules]

What about this Idea, being that these 2 radius are a full 360 degree around the center of my part (origin), I had my engineer create me a diameter of both radius all the way around the part. This will give me size and location. Then check my radius with radius gage from the side to get the profile. So at a certain height I will get a diameter and we added probe comp angle to make sure we are good.

[AuRules]

Thanks for the input guys. Ran each scenario thru my head and maybe the profile is the best one. Thanks

[R2ah1ze1l]

Personally, I would measure those surfaces with points (5 or 7) for each rad. Then construct the points into a circle, now its really easy to report both TP and size (or profile if desired). If you have scanning capability, then scan those surfaces, and construct the circles.