True Position calculation clarification

[ajxxx]

Hello folks

Firstly, thanks to everyone who takes the time out on this forums to help out. Really appreciate.

Ok bit of a newbie and requesting some help on a true position calculation

Attached a sample format, am very clear about the dimension Y (which would help me with delta Y)

For the X my belief is, since the datum C is specified on axis, it would be "theoritical centre of the width W" and the basic X dimensions would be D2/2 (for inner holes) and D1/2 for outer holes.

1. Would this be correct? and

2. Practically how would I be able to measure this? Say on a CMM...how would the datum be determined as 0 point for X?

3. Am sure we cannot but throwing it out there to check it (to get a "physical reference" some people have been recommending to subtract D2 from W and then divide it by 2 to get the centre of the outside holes (i think that is not correct but do not have a reference for the correct theory as well). Closest I can think of is Fig 4-33 of ASME Y14.5-2009. There is no note on the dwg which states holes are equidistant from the edges so a highly doubtful

If somebody could guide me here, would appreciate. As always thanks for your time and reading this - Appreciate

 

[drawoh]

ajxxx,

Your dimensioning scheme ought to be based on how your part must work. Your datums are not a good match with the way you apply dimensions. If the datum[ ]C feature is one end, the best dimensioning is ordinates originating from that end. For the centred holes as you show them, a better datum[ ]C feature would be your width[ ]W. Feature of Size datums are tricky. Inspection gets messy if the feature of size is not accurate. Note how these two datum and dimension schemes do not mean the same thing. An alternate datum feature would be to select one or two of the holes or to select the whole pattern. You could then apply a profile tolerance to the outline of your part.

If I were to inspect your part as shown, I would note that your holes are nominally centred on width[ ]W, and I would work out the nominal distance from the datum[ ]C end. I would measure accuracy from there. I would not worry about centreing. Is this what you want?

--
JHG

 

[jasoncwells]

Re #1) Yes, C is in the midplane of the part.

The way I read your PPT, you have not specified the position of the holes along the left-right axis. Just because datum C is specified as the mid plane of the feature of size does not imply that your holes have to be symmetrical about it. The holes look symmetric, but they are not specified symmetric. You need a basic dimension to one of the edges or you need a symmetry control. A former employer of mine required a center line with a "SYM" note at each end. (I never found that requirement anywhere but our local drafting standards.) The placement of basic dimensions does not affect their meaning. But you always have to get back to a datum with a basic dimension(s) somehow.

Re #2) (not a metrology guy) Generally a CMM will touch off on the part with a saphire stylus in multiple spots and mathematically produce your datum plane.

Re #3) see answer #1.

Regards,
Jason

 

[pmarc]

ajxxx,

1. Technically datum C on your sketch is not an axis but a center plane derived from the width of the plate (dimension W). You did not associate any position feature control frame with any of the holes, but I assume the datum order of precedence is A primary, B secondary, C tertiary. If that is the case, the theory in Y14.5 says that the datum center plane C must be perfectly perpendicular to datum plane A and perfectly perpendicular to datum plane B.

Additionally, you did not show a vertical centerline on your sketch that would graphically represent theoretical datum center plane C, but if you did it, the lack of basic dimension from that center line to any of the four holes would mean that the pattern of four position tolerance zones would be symmetrical about the center plane C (see implied basic zero dimension rule in para. 1.4(k) of Y14.5-2009). So it would be true that the basic dimensions from the "theoritical centre of the width W" would be D2/2 for inner holes and D1/2 for outer holes.

2. Like I mentioned, the datum center plane C needs to be perpendicular to both datums of higher order, that is A and B, so on a CMM you would have to probe datum feature A and construct primary datum plane from the collected hits, then probe datum feature B and construct datum plane B from the collected hits (that plane would have to be perpendicular to datum plane A), and finally probe both faces of datum width C and construct a datum center plane C from that (again, keeping in mind that the center plane needs to be perpendicular to datum planes A and B). This would be your 0 for X direction. All four position tolerance zones would lie at applicable basic distance in a direction normal to the datum center plane C.

3. Not sure I understand your description, so I will not comment on that.