Issues with repeatability

Alright, so I asked this once before, but with some rediriecting and checking (that and posting in the correct forum) I feel I may have a better shot at getting this figured out now. I am using a RA-7525 portable CMM. The issue is that I will take hits on the same part and not get the same reading twice, even when reading gauge blocks. The dimensions will very sometimes as much as .25" or more... My dads old stanely measuring tape is more precise than that. I've played around with the different heads (we have a 3mm, 6mm, and center finder), and still no improvments. I performed a cone test and got a min deviation around .124 and max of .245. Any possible ideas on what could be worng? Is this machine error at this point or is it still human error as my boss thinks?

1. talk to your trainer. send him/her the code.
2. post the code here.

Just two problems with that. One, I was never formally trained by hexagon, when I became a part of the qc department I was given an overview of the machine, the training binder, and told to practice at it when we have down time. Two, I have no idea how to pull the code.

Just two problems with that. One, I was never formally trained by hexagon, when I became a part of the qc department I was given an overview of the machine, the training binder, and told to practice at it when we have down time. Two, I have no idea how to pull the code.

When you are using a portable, you have to take two things into consideration. 1. Vectoring and 2. How steady your hits are.

Since the portable is not tied to an axis table like a bridge CMM would be, it has no way of knowing what vectors you intend for a feature to have unless you tell it and therefore, it could be compensating your probe incorrectly.

The best way to eliminate that possiblity is to always consider the centerline through the probe body as your axis and always try to come in at anything perpendicular to establish proper compensation. For example, for 2D features such as lines, circles, etc., your first hit should be as close to 90 degrees from your workplane as possible. The following hits don't have to be as the surface vector and probe comp have been established by the first hit.

The second and most common issue is just getting used to the "touch" of the machine and finding your sweet spot. If you bounce the tip, come in too hard, shake, or aren't careful that you aren't hitting more than one surface at a time, it's very easy for these things to occur and affect your end results.

I wanted to throw mine out the window before I finally got more accurate with my results. It just comes with time and practice and it can be very frustrating, but, from the sounds of things that is more than likely what I think you are experiencing. The best way I found to get some practice was to do simple measurments on a known artifact like your length check bar. The cone seats are easier to get repeatability on since they are short movements and you aren't moving the probe much. Try doing some length checks and see how tight you can get your results, as well as how repeatable you can be.

Just remember that a 7525 has an acceptable accuracy of somewhere around .0015" over the length of your bar, so don't be frustrated if you can't do better than that with the portable. It can be done, but, I just wanted to forewarn you of what you might see so you didn't want to toss it through the wall.

And BTW, if you've just been tossed into the Quality department without any training on PC-DMIS, then shame on your boss. PC-DMIS can have enough of it's own demons to battle, let alone trying to learn how to control the arm. If would definitely improve your growth potential if you were able to remove one of these obstacles.

Awesome, thanks for the tips! I have a few slow nights ahead of me this week so I will pull out the bar and see what I can accomplish. I appreciate the help!