Laboratory Policies

[dallanta]

ROFL, I read this for the  first time, hoping you are joking, but can see you are not.  How has it been going since they started this crap?
  Yep a few nrts and calls to the item managers and the poop would be in the fan I hope.

[flew-da-coup]

If AFMETCAL is so much better than civilian than I guess taking as found and as left data for tracking trends in error is a waste of time.  AFMETCAL does not require this, But NIST does. I guess NIST needs to be sending there stuff to AFMETCAL for cal. Give me a break. You like Airforce labs because they are easier. I worked under NRC for a standards lab for nuclear power plants and the requirments there make 00-20-14 look like a Joke. I also have worked in a Air Force lab so I know what I am talking about. You may have worked for a joke lab in the civilian sector which has given you the wrong idea about civilian metrology.

[flew-da-coup]

I also guess the K procedures  are better because they don't check every 10% of each range on something like multimeters. That one point check is much more reliable. God forbid no one has ever heard of linearity errors on a meter.  Or just requiring only 6 frequency points on an attenuator like lets say a DC-18GHz and you check 2G, 5G , 8G, 10G, 14G and 18GHz God forbid there will be a hole at 6.5GHz. I can go on all day long. AFMETCAL is actually behind the curve. Hell they are just now getting into automation.

[MIRCS]

I agree.  Hmmm....the dreaded as found/as left data.....all filled out by hand due to Part 11 of the CFR's

I miss the days of just looking at the display and moving on..........

and to continue with what was being said....................why is the 470 20" Sine Bar not calibrated for flatness?

[flew-da-coup]

Here is another AFMETCAL super metrology moment: For the longest time AFMETCAL had on their website that you should clean your Ruska 2465 high pressure piston with acetone. Their statment " this should give you about 7-8 could readings before you need to clean the piston again". Well, I say just clean the piston like Ruska suggest with cashmire bouquet soap and you can get about 200-300 good readings. AFMETCAL does it again. Will someone please back up AFMETCAL before I really start tearing apart some airmans heart who thinks AFMETCAL is GOD. I guess not, because you can't. I just think it is funny when someone thinks a government entity is better than the private sector.

[MIRCS]

I just think it is funny when someone thinks a government entity is better than the private sector.

Well that's a double edged sword there. On one hand they are better.......the military in general. On the other some parts of the private sector are light years ahead......while other parts are just figuring out a round wheel is better.

Nuclear Power..........is probably light years ahead of most. Pharma.......well some are and some aren't, but they are being forced into it as we speak.

I currently work in Pharmacueticals and I can without a doubt say................the people here that matter....ie QA, Validation and manufacturing.............really ain't gotta a clue at this site. I am the first person to work at this site with any metrology training.......eye opener to say the least to myself and those that work here. we are slowly working through the problems.......this place will be rock solid soon and we will be in line with the majority of the private sector that is lightyears ahead of the military.

[flew-da-coup]

Hey MIRCS,
  Are you an old IM. I was an IM before I went to ET school. IM "A" School, Pensacola FL. April 1990. 

[MIRCS]

Hey MIRCS,
  Are you an old IM. I was an IM before I went to ET school. IM "A" School, Pensacola FL. April 1990. 

How else do you think I know what MIRCS is. IM "A" school Pensacola 91

IM's RULE!!!!!!!!!!!!!!!

USS Prairie
USS McKee
USS Cape Cod
SIMA San Diego
TRF Bangor
UNCC Metrology Center
Moody AFB PMEL
Hospira

[MIRCS]

And actually I remember one of the recent worldwides had concern for the Phys-D world since there were no more IM's, and that skill would be lost to all.

[Hoopty]

Alright, here's my 2¢ on the subject...

On the one hand, we are all metrologists here.  Pretty cut and dried.  Ultimately, we make quantitative measurements by comparing a known standard to an unknown.  We're not engineers or rocket scientists, nor do I think a monkey should be doing our job.  We're not saving the world, but the world definitely needs us around.

On the other hand, each sector has a different mission with different priorities.  The private sector is primarily about the bottom line.  While the AF is bombs on target.  Sure, money is a concern, but not nearly to the extent that it is in a commercial lab.  You just can't compare the two.  You're comparing apples to oranges.

If AFMETCAL is such a joke, why aren't more planes falling out of the sky?  Is it perfect?  No way.  But it works for us.  Is the private sector perfect?  I doubt it (can't atest to that as I have no first hand knowledge).  But again, it seems to work for them.

Sure the K-pros have problems, but you mean to tell me the commercial procedures get it right every time?  Again, I doubt it.

Now, I'm not trying to get involved in a pissing match here.  And I'm not your "airman that thinks AFMETCAL is God" either.  I just don't think that you can accurately state that one program is better than the other.  I don't think that anybody's program is perfect, and you really can't run the comparison that you're trying to.

[docbyers]

Hoopty and Flynn make good points.  Generally speaking, I think it might all come down to the mission.  In the military, it's "bombs on target."  In the civilian world, it's profit.  Is there a way to find a balance between what needs to be calibrated to accomplish the mission versus how effectively can we do it so that we can make a buck?

Take the 4x principle, for instance.  In PMEL School we were taught that this rule worked pretty well- your standard needs to be 4x more accurate than what you're testing.  Fluke, Tek, H/P- they all made calibrators and signal generators that performed nicely for the meters and scopes we worked on, and DOD was happy to buy what we needed.  Now, here is what guys like MIRCS and I go through routinely: you have a nice weight scale to calibrate, and you want weights that are 4x better than the scale.  First thing you have to do is explain the 4x principle to management, who's usual response is "The weights cost HOW MUCH?!?"  Then you do a thorough evaluation of just how accurate the scale needs to be for what you're weighing, and find that, well, yeah, I DO need weights that good.  You get your weights eventually, after much wailing and gnashing of teeth...  MIRCS' experience is like mine- nobody but me in the plant is PMEL-trained, and no one, from the president to the janitor, know what metrology is all about.

I think another difficulty comes from the fact that all good metrologists are military or ex-military (we all went to the same school, either in Colorado or Mississippi).  It's easy to work for a contract lab at a military base.  It gets a little harder to work in a purely civilian environment, because we're trying to apply our military training and experience to a civilian job.  Again, the mission is different, and we have to adjust our thinking accordingly.  When I wore a uniform to work every day, all I worried about was doing high-quality calibration work in a timely manner- that was it.  These days, I worry about the cost of weights, and do I really need the uber-spiffy ones, or can I get by with a cheaper set?  I WANT the uber-spiffy ones, because then there's never any question about my scales, BUT I have a budget that doesn't leave me enough money for paper for the tractor-fed printer...

At the end it all comes down to doing what is right from a metrology standpoint; it costs what it costs, and I can usually justify the expense of the nice weights or the Fluke DMM vs a Radio Shack model...  Getting my troops paid what they're worth- now there's the next battle to fight.

[OlDave]

OK, let me toss my grenade in here as well....

I have to agree that the military (at least the AF) provides a good starting point. But the AF does NOT make metrologists. They provide you with the basics to understand and repair electronics and a little of the theory to understand the importance of calibration. I thought I was pretty damn good with 20+ years of experience in both type 2B and 2C labs, numerous advanced courses at Lowry, and experience prototyping instruments for AGMC, but I was absolutely HUMBLED at how little I knew about metrology (the actual SCIENCE of measurement) when I took my first course at NIST.

Control charts? F tests? T tests? Statistical analysis? Propagation of uncertainty? Uncertainty analysis? Guard Banding? All unknown quantities with the AF. I had the mechanics, but I had to mentally move to another level of thinking. That's when I realized that the AF only has calibration technicians. AFMETCAL does not allow you to be metrologists, and honestly, most people in the lab couldn't cut it.

How many of you have the tools in place to detect a 0.5 microgram shift on a 1 milligram weight? How do you know for an absolute certainty that you didn't have a speck of dust on your standard 2 gram weight that you just used to calibrate another weight with? Has your SPRT's Ro drifted a couple of milli K? What confidence level do you report you measurements to? Can you pull the historical data on an item and see how the previous calibrations differ from this time? What are you going to do about it?

But does the AF need that level of metrology? No, in most cases not. But on the same token they need to encourage the people that can and do master the statistical witchcraft of metrology instead of blindly requiring conformance to a calibration procedure that may or may not be correct or adequate. But now you're expected to detect and correct the difference in the 18 character part number that signifies the pressure gage you just stamped should have a phenolic case with a male pipe thread connector instead of the cast aluminum female thread it has. Is that really that important to the accuracy or reliability of the item?

How many of you participate in proficiency tests or inter-laboratory comparisons? ISO 17025 practically requires that you do that. How else can you measure how your capabilities compare to both what you report and the capabilities of other labs? Back in the REALLY old days AGMC used to send artifacts out to the labs to be measured. But that usually required people to think outside of the box, something that is discouraged or prohibited now days.

Unfortunately I have found that in most cases the military calibration program and the associated regulations and procedures are written for the lowest common denominator, i.e. the newest or dumbest person in the lab. So make the most of the training and experience the military offers but always keep an open mind because there is a LOT more out there in this field than you are probably aware of.

[Hoopty]

OK, so now it's about the fact that we're not really metrologists in the AF...   

If you want to get technical, no we're not.  And really, I can buy that.  We (the majority of us AF types anyway) are just calibration techs pretending to be metrologists.  OlDave, hit the nail on the head in his post.  With the key point being, that it's not necessary for AFMETCAL to take it to that level for the most part.  Good stuff to be aware of, but not a necessity for the average joe.

BUT, the point I was trying to make in my post, was that even though we all work in the same metrology (or calibration) world, the roads lead to different places.  And you just can't say that one is better than the other... because what works for one, may not work for the other.


Now, I'm all for discussion, but it seems to me that this thread has gotten way out of control.  I've thought and thought and thought about this and even attempted to draft a decent reply, but in the end it made my head hurt and it was mostly just rambling.  I don't know, maybe this just shows my ignorance about the overall big picture...

[OlDave]

Well it's nice to see I wasn't locked out and banned from the forum....

I apologize to Hoopty and anyone else that feels I cut just a little deep on my last post. But I believe the point I was trying to make was made. The AF PMEL world has a basket load (not an SI unit) of outstanding calibration technicians. But in my opinion there are few, if any, true metrologists doing true metrology work even at the AFMETCAL level.

There is a world of difference in the requirements at every place you will ever work. Some places are in it for the money, some places are in it to keep the FDA off their butt, others are concerned with keeping the missiles locked on target. But we all have very little other than our own personal integrity that keeps us honest and the data correct. And that's the thing that sets PMEL troops apart from the rest. When WE say it's right, you can bet your a$$ on it!

[Freezer]

I got nothin..

   Cept maybe a statement that I'm proud to be whatever I am, and that most of the "work related" part of what I am is because of AFMETCAL, in a trickle down sort of way.  If I'm a calibrator, well then, I'm proud to do that for a living.

  If someone wants to call me a Metrologist, that's OK too.  In the same sort of way my wife believes the measurements in the bedroom are correct, if you get my drift .

I'm not pretending to be more than I am, or less, I'm just happy to be in the career field, with a big ol' bunch of very inteligent and equally proud people, call them what you want (they've probably been called worse).

  Discourse and debate is good, but let's not let it get too out of hand.  Remember, this is FUN!  Go read some of Docs jokes for heavens sake!  Make fun of Minot!  Tell InLikeFlynn he has a big nose!  The possibilities are endless...

(Is that a Lee Greenwood song in the background?)