taking plane measurements

Technically it's an infinite number of points.
In reality you need to look at several things:
Method of manufacture
Tolerance involved
Time required for inspection
Basically, you need to evaluate what's best for your orgaization.

There is no document or standard that I am aware of and the only true answer is that you need to sample the entire surface in order to prove conformance. Obviously, that is not always practical so there is always a trade-off between trying to take enough data to accurately capture the surface and the time it takes to gather that data. There are SOOOOOOOOOOOO many variables that come into play...

• what equipment / sensor is being used
  
• what process was used to produce the surface being inspected
  
• what type of specification are you trying to verify
  
• what tolerance are you trying to achieve
  
• what level of uncertainty is acceptable
  
• how much time is available to perform the measurement
  

To name but a few. For things like a flatness check for example, it can often be much quicker to set up a part on a surface plate and sweep a dial test indicator over the whole surface than it would be to perform a comparable check on the CMM. However, it may be decided as acceptable to inspect the part on the CMM (and gather comparatively less data) because the flatness is not critical, the tolerance is well within the capabilities of the equipment and the CMM does not tie up an inspector for the entire duration of the measurement - he can simply load the part, start the routine and walk away.

Nobody will ever tell you how many.

Cause if they did, what if what they told you wasn't enough?

It's up to you.

Well stated!
I have people that tell me "thats all the hits you're taking!".... but they cannot tell me where its stated that I "HAVE TO" take a certain amount.
Then when that person has a part set up on a flat plate and three posts, they are only checking at three positions.... hilarious! I have been doing this for 30 plus years, but, fairly new on a CMM.... was curious on how this person was dictating on how many hits were acceptable. Thanks alot for responding!

Well said, also! (see my response to "paperback rocker")

Thanks for responding! This site rocks!!

Thanks for responding! Seems to be a common response : Whatever I want! (see my full comment to "paperback rocker")

Thanks again!

i had a customer i had to send all my CMM programs to in order for them to evaluate them. i submitted one program that checked a completely unimportant surface (plane) with a .050'' flatness. it was about 6''x6''. I put 10 points on the plane and submitted the program. it was rejected. the reviewer said i needed more points on the surface. so i put 100 points on it. it was rejected. he said it didn't need that many points on it.

i asked him how many points he wanted on the surface.

he said between 11-99. i almost lost my *** on this guy. we don't do work for them anymore. sunovabitch.

At my last job, whenever the engineer would come down to look at some failed part, he'd always ask how many points were taken.

If it was a lot, he'd say "try less points".

If it was a seemingly normal amount, he'd say "try more points".

holy guacamole

everyone loves the machine when it checks green, and everyone hates it when it rejects something. the struggle is real.