Saturday morning GD&T --- Surface profile of a diameter (cast cylinder)

It looks like the draft angle is included in the CAD/print. If that is the case, you could just build it as a cone, and report profile (form only) of the cone (that would also allow you to report the (included angle) of the cone to report the draft angle*2 (divisible with variables)). It seems to work fine with a cylinder. If not, take individual points, whatever is necessary to sufficiently represent the surface, create a set from those points, and then report profile of that.

If draft is not in CAD, I don't know how that would actually be possible to represent with PC-DMIS.

The attached drawing shows the 40.8 Ø (+ draft (3° per notes is draft)) on a cast OD with a profile of 3.6 and according to the control plans Engineering sent down on the inspection plans this is being called out to check as true position of 1.8 back to -D-.

I will discuss this with the engineer, if the print calls for profile why he is calling it out as TP in the control plan.

According to the attachment the profile is form only, you can take a series of vector points around the OD (assuming you're working with CAD)
Construct a feature set and report profile form only.

This might sound silly, but what I would do is create a complete dia 40.8, create my circle 6 to 8 hits create a center point, call that my "X" & "Y" (new alignment) then create 4 circles only taking 4 to 5 hits each

that would show me highs and lows of that circle

that's overkill but that's what I would do

The print says 40.8 +draft. So the diameter at the theoretical sharp corner at the top of the feature is 40.8 and it gets larger as you go down. I would measure a few hits on the top shoulder of the feature, use that as Z origin, and set X,Y origin to datum D axis. You could then measure the feature using the auto cone tool as Vinnie has mentioned. You will need to edit the hits on the lowest level because it appears that it extends further down on the right hand side of the print.

Now you can dimension the cone using a legacy position dimension as instructed by the engineer (make sure you check perp to centerline so it doesn't try to give you 3D position.) I would label that dimension as reference only since it is not called for on the print (the engineer might have a really good reason for asking for this information so best to try to provide it.)

Now I would also create a constructed set composed of the hits from the cone: Cone1.hits[1..numhits] or Cone1.Hit[1],Cone1.hit[2],Cone1.Hit[3],etc. (the second way is tedious but sometimes it helps avoid errors telling you that there aren't enough features for analysis)

Use a legacy profile of surface- formonly dimension with plus tolerance of 3.6 to dimension the constructed set. Just to be sure that all of your hits are being considered by the dimension create an analysis view to verify that you can see all of the hits in the image. Whatever you see there is what is being processed in the dimension. You can delete the analysis view once you have verified that things are working properly.

Now you will have satisfied both the print and the engineering request. I would then discuss with the engineer, as Roberto mentions, to understand the intent behind the request. Perhaps he doesn't understand GD&T and you can educate him on what surface profile is. Perhaps he knows something about the function of the part that is driving the request. In that case he can educate you on the parts function.

my bad didn't see the draft. This is why I don't program on the weekends. Just come in here and run parts then again do 2 circles. and create a cone. but I would do it Dasalo's way.

I will discuss this with the engineer, if the print calls for profile why he is calling it out as TP in the control plan.

According to the attachment the profile is form only, you can take a series of vector points around the OD (assuming you're working with CAD)
Construct a feature set and report profile form only.

+1

Also when reporting make sure to change to min. and max rather then deviation.

A simpler solution would be just to create a cone out of the lower drafted cylinder, using more points than required to get a more accurate cylindricity measurement, and then perform a coaxiality measurement back to Datum D cylinder, using standard tolerances posted on the print.