Measurement Uncertainty (MU) Budget for a 0-1 inch Micrometer

[rich.dittmanjr]

I am a newbie to the uncertainty world. I started working in our lab a couple years ago and noticed that we should be working to ISO standards and not to the 1960’s mil specs. Long story short, I need a little guidance to see if I am on the right trail. We work in inch, so everything on my budget is in inches.
I also made an assumption that a 0-1 micrometer with a .001 resolution could be raised to a .0005 resolution due to the large space between the graduations on the barrel.
I have enclose the budget for people to look at, and to give me some feed back.

 

[Hershal]

The budget looks reasonable to me.....seems like you took into account at least the major contributors of MU.....

The space between the marks of the barrel might be best left alone.....if you let folks measure using the dead space, you now add parallax error, which will be equal to the resolution error (0.0005"), and so the MU will likely increase.

Hershal

 

[rich.dittmanjr]

Resolution Error

Thank you for your feedback Hershal. I feel a little more confident moving forward on other gages now.
The resolution of the micrometer still bugs me. I have read many different articles saying you can use the area between the vernier etch marks as a half increment. I also read you do not do that.
At this plant they normally round to the .001 and don’t estimate the distance between the marks. I feel safe because in this case, the calibration uncertainty based on .0005, would then be safe.
On the other hand, the Manufacturing team should not be using gages that do not meet the 10% resolution rule.

 

[rich.dittmanjr]

Is this right?

I sent a 0-1 inch vernier micrometer with .001 resolution to the calibration lab that we use for other gages. I did this to get a start on setting up our uncertainty and also do a little inter-lab comparison. When I saw what they had for stated uncertainty, I am really questioning the credibility of this lab. I cannot see how you can get +/- .000083 uncertainty from a vernier micrometer. Am I missing something?