Dear all, I am handling a quality problem inditing 85000 parts involving 45 IP's, 45 part numbers.

The calipers failed the yearly calibration cycle. The calibration facility reported that the calipers were off by 0.003 on the I.d and 0.0055 on the O.D

Everything, ever since those calipers were last calibrated (a year back) is suspect. How do you do a risk analysis in this scenario ?

:(
senior members please advice !

Dear all, I am handling a quality problem inditing 85000 parts involving 45 IP's, 45 part numbers.

The calipers failed the yearly calibration cycle. The calibration facility reported that the calipers were off by 0.003 on the I.d and 0.0055 on the O.D

Everything, ever since those calipers were last calibrated (a year back) is suspect. How do you do a risk analysis in this scenario ?

:(
senior members please advice !

First and foremost a question, how do you decide on the calibration cycle? When high volume of parts are to be measured, then the calibration interval should commensurate with QUANTITY in stead of TIME. A caliper used by a Quality engineer has lesser usage vis-a-vis the caliper used by quality inspector, and the highest usage is by the patrolling inspection team. All the three calipers must have different calib. cycles. The popular mistake is in keeping all one-type-of-tools in a single calib.cycle.

:caution: The risk is great and IMO all components in question, must be quarantined and put to 100% inspection with a new caliper.

...................... Everything, ever since those calipers were last calibrated (a year back) is suspect. How do you do a risk analysis in this scenario ?............................

What does your procedure say?

First, I would go back through the measurement records. Are there any results within 0.003 of the tolerance. This will help to quantify the problem.
Second, assuming the calipers are worn, how would a part being oversize by 0.003 effect the fit and function?
Last, have the worst condition measured parts been used by the customer? Did they report a problem?

This will enable you to determine the level of risk and thereby the appropriate action to take.

Dear all, I am handling a quality problem inditing 85000 parts involving 45 IP's, 45 part numbers.

The calipers failed the yearly calibration cycle. The calibration facility reported that the calipers were off by 0.003 on the I.d and 0.0055 on the O.D

Everything, ever since those calipers were last calibrated (a year back) is suspect. How do you do a risk analysis in this scenario ?

:(
senior members please advice !

If the calipers were used day in and day out in one year duration, then hypothetically the wear on ID was @ 0.003/12 = 0.00025/per month. Similarly on the OD was @ 0.0055/12 = 0.000458/per month. w.r.t. Tolerance of the parts in question you may decide the safe months (starting from last calibration), as well as the border line cases. Beyond that, every thing is suspect. You may have to recall the supplies as well as quarantine the material in your facility.

Secondly, have another look at your calibration frequency, it appears out of sync.

Hope this helps

Umang :D

we verify ours every day against known standards. there is no way we could afford to put a year's worth of product at risk. Wow, sorry to hear aboutthis problem. Good luck to you.

This is also a good time to determine if standard calipers are the right tool for the job. Sure, Gage R&R will tell you if they function as an adequate measurement tool, but has no time function - calibration does. But, beyond that some knowledge of the gage is necessary. Calipers are handy because they measure many things, and do it fairly conveniently. But, unless you get special calipers, they generally do not have adequate measurement surfaces for high volume or high wear measurement (such as the carbide anvils on a micrometer). Not only that, when they do wear, there is little that can be done for them - whereas an indicator tip can be easily replaced a very low cost.

Although this is a very unfortunate problem (I have lived through them, too), it is a good learning opportunity for all of us. Do not rely on just gage R&R for gage choice!

What does your procedure say?

This was short and sweet, but makes an excellent point. If you don't have remedial action in your procedures, you need to put it in there.

The question is not if a failure will occur, but when. There are a lot of golden gems in all of these posts on this thread, and need to be considered when going through this entire process. Part of this should be determining how failures will be evaluated and handled. You really should implement Remedial Action for when you have calibration failures.

Risk analysis is important, here, and SteelMaiden's post is one to not be taken flippantly. If your risk is high, then you need to take steps to mitigate that risk.

How different was this wear from the last time they were calibrated? The instrument could have been near tolerance the last time they were checked. Hopefully, your instruments were calibrated by a competent lab, and you purchased data with the calibrations. All calibration documentation needs to be analyzed, and used for effective management decisions.

Let us know how this progresses.

Dear all, I am handling a quality problem inditing 85000 parts involving 45 IP's, 45 part numbers.

The calipers failed the yearly calibration cycle. The calibration facility reported that the calipers were off by 0.003 on the I.d and 0.0055 on the O.D

Everything, ever since those calipers were last calibrated (a year back) is suspect. How do you do a risk analysis in this scenario ?

:(
senior members please advice !

:topic:
In addition, you may want to look at the manufactures stated accuracy. 12" Mitutoyo dial calipers have a stated accuracy of .002".


As to your risk analysis - I think prototyper covered it pretty well. I really can't add to that advice.

:topic:
In addition, you may want to look at the manufactures stated accuracy. 12" Mitutoyo dial calipers have a stated accuracy of .002".


As to your risk analysis - I think prototyper covered it pretty well. I really can't add to that advice.

Good point, Carol. I didn't even think about going into that. You may need to purchase more accurate instruments, where small departures from the tolerance does not have an effect on the measurement process.

Also, buy a gauge block set. If you cannot afford a set, purchase two or three of them (across the range of use).

How about checking to see if the user is competent with their use.

I've seen calipers used as a wrench, a spanner and even a tack hammer.

Could be as basic as this.

Dear all, I am handling a quality problem inditing 85000 parts involving 45 IP's, 45 part numbers.

The calipers failed the yearly calibration cycle. The calibration facility reported that the calipers were off by 0.003 on the I.d and 0.0055 on the O.D

Everything, ever since those calipers were last calibrated (a year back) is suspect. How do you do a risk analysis in this scenario ?

:(
senior members please advice !

The calibration cycle issue has already been discussed, and a shorter cycle given the volume is certainly in order.

In theory, the ID/OD should be off by the SAME amount, not a difference of almost 2:1, I would question that right away.

Calipers will wear out with enough use. I presume several calipers were in use, but still.....

How about checking to see if the user is competent with their use.

I've seen calipers used as a wrench, a spanner and even a tack hammer.

Could be as basic as this.

I was thinking the same thing. You might find that the calipers are not worn, but actually sprung, bent, or have a burr. If this is the case, the response earlier for basing how far you go back for the part recall (or review) on wear over time no longer gives you a particular time frame (other than 1 year ago) to go back to.