Significantly Out of Tolerance Reporting

[Rocket]

ANSI Z540.1-1994(R2002), para. 13.6 requires notifying customers, in writing, of:
A) Defective cal equipment or data in a report could invalidate the validity of results, and
B) When customer's M&TE found SOOT during cal process. (PARAPHRASED)

Since no specific guidelines are established for defining SOOT, what does it mean to your organization?

[RFCAL]

This also depends on whether the instrument was adjusted to within tolerance or repaired.If it was repaired,you do not owe any oot data as this is considered a hardware failure.

[Bryan]

The significance really is up to the customer, I pass it along to them if out of tolerances are observed period.

[OlDave]

I agree with Bryan. Notification is sent if the unit is out of tolerance by ANY amount. Significance is determined by the user.

[msrichmond]

Same here. Unless you are prepared to evaluate every single way the customer uses their equipment, you're best just reporting ANY out of tolerance conditions and letting them determine the effect on their processes.

[USMCPMEL]

This is why equipment is calibrated. You need to report any out of tolerance condition. I know guys that do not because of the extra paperwork. I believe that is doing a great disservice to our entire community.

[obp42]

I can understand the need for reporting lab standards.   As for our customer's equipment, they could care less.   I usually hear, what am I supposed to do with this or it get's round filed before they leave the building.   Just another tree killed.   I guess I'm old school.   We didn't use to do traceability reports, recalls or out of tolerance letters and we continued to have an acceptable takeoff to landing ratio.

[msrichmond]

I can understand the need for reporting lab standards.   As for our customer's equipment, they could care less.   I usually hear, what am I supposed to do with this or it get's round filed before they leave the building.   Just another tree killed.   I guess I'm old school.   We didn't use to do traceability reports, recalls or out of tolerance letters and we continued to have an acceptable takeoff to landing ratio.

You don't explain it to them? I have lots of customers that don't have a clue; military rotations put different people in the PMEL coordinator position who know nothing about PMEL. I consider it part of my job to educate them about how the OOT condition might have affected their equipment and to suggest they evaluate their end use to determine whether it did. Some people take it seriously, some don't but I feel I've done my part to ensure the best possible outcome.

If they say "What am I supposed to do with this?", likely they don't understand what you handed them and it is the work of a few minutes to explain. It might not affect their usage at all but...it might be the difference between a plane arriving safely or crashing into the ocean. My two cents.

[mdbuike]

Our policy is if it gets to 75% of tolerance, optimize it, and let the customer know it was optimized...still doing 7000 series scopes, and the first thing you do is check rise time (HF Comp)...

Yeah, we have a melange of equipment to support

Mike

[ck454ss]

I can understand the need for reporting lab standards.   As for our customer's equipment, they could care less.   I usually hear, what am I supposed to do with this or it get's round filed before they leave the building.   Just another tree killed.   I guess I'm old school.   We didn't use to do traceability reports, recalls or out of tolerance letters and we continued to have an acceptable takeoff to landing ratio.

In a manufacturing environment its a must to know how much my equipment is Out of Tolerance for quality purposes.  I work for a company who makes parts for your aircraft and we have had recalls on parts based on Out of Tolerance reports.  What defines a "Significant Out of Tolerance" is your quality manual/system.  For a 3rd party lab I ask them to give me data immediatly on equipment thats more than 5% OOT per my Purchase Order requirements so I can begin my process of determining if a recall is needed.

[Hawaii596]

I don't like the term "Significantly" as it is ambiguous and can be confusing to some.  I don't like the idea of having Significant in the language of a quality policy.  I think it is significant if it is out of tolerance, in that the instrument was designed to make measurements to a given tolerance for a given time.  If it does not perform to those specs while the user is using it, then I believe they need to know that their insturment did not meet its specs (the specs being how well they could expect it to measure for a given perioud of time).

The only way "Significant" should be reasonably part of a policy is if there is a well followed and documented policy to assure a given ratio between test instruments and what parameters are being tested.  If there is a 100% followed 10x policy between test parameter limits and tolerance (for example), that means with a well followed guard band, it is reasonable to presume that OOT's will not necessarily be significant until they reach a given percent of tolerance (too early in the morning to do that math).  If there is no such policy, and consistently followed, then there should not be a significant OOT policy, as it can only by definition be SOOT if it correlates to a given ratio between tolerance and test limit.

[obp42]

I'm sure some organizations take it seriously.   I'm dealing with knuckle dragging line apes.   We explain how it's supposed to work and you usually a get glazed over stare with some drool in the corner of the mouth.   Don't get me wrong.   I appreciate what those guys do and what they have to put up with but all they care about is sortie generation.   

[jimmyc]

This is why equipment is calibrated. You need to report any out of tolerance condition. I know guys that do not because of the extra paperwork. I believe that is doing a great disservice to our entire community.

very good point, when it becomes part of your normal process, it doesn't take anymore time.

[RFCAL]

If the UUT is inop due to a hardware failure, OOT or Soot data cannot be taken.Does a report still need to be sent?

[Hawaii596]

The philosophy I've always used with OOT versus INOP/MALFUNCTION/etc. is whether or not the user could possibly have made an OOT measurement (or generated output) and not known it.  If the unit is dead, then no OOT.  If there is something so grossly wrong with the UUT that the user could not make an OOT use, then I consider it malfunction.  If the malfunction can be corrected such that as found readings can be made without probability that the malfunction repair changed the accuracy of the UUT, I make as found readings and determine if there were any OOT's separate from the malfunction (such as changed fuse, broken power cord - oversimplified examples).  I spent a few years in an FDa regulated electronic mfg site where they were crazily strict about such things (hours long debates about where to draw the line between malfunction and oot).   That site, by the way, considered all OOT's to be significant.  I have a lot of respect for them anyway, as they ran a very tight quality system (one of the tightest I've ever worked in with exception of nuclear industry).