From: Timothy N. Stevenson

Sent: Friday, February 11, 2011 8:20 AM

To: Phil Heitzenroeder

Cc: Bob Simmons; Mike Williams

Subject: Re: CALC-13-03-01 Ready for Final Review

Phil

The text in 33 allows the checker to use their own judgement (graded approach). For instance, if the checker looks at the assumptions for a problem and looks at the inputs to ANSYS for a calc and the properties look right, the checker (who presumably knows ANSYS well enough) can conclude that the calc is OK or similar to the way they would do it. Methods could include back of the envelope, comparison to similar calc on something else, detailed review of inputs and outputs, etc... Performing the calc again would be a case like Titus and Willard just went through to resolve errors etc.

Does this help?

TIm

ENG-033 A

Cognizant Individual 1. Develops calculation in accordance with the format described in

attachment 1 or project equivalent. For software calculations

using code or software applications, Cog shall so document the

input and code used that a competent reviewer could determine

validity of the calculation.

Responsible Line

Manager

2. Appoints a qualified checker or reviewer for the calculation.

Checker 3. Reviews the calculation using the minimum requirements of

attachment 2. It is the responsibility of the checker to use

methods that will substantiate to his/her professional

satisfaction that the calculation is correct.

4. Resolves concerns with developer of calculation and signs

calculation sheet or project equivalent.

On Feb 11, 2011, at 6:26 AM, Michael Williams wrote:

Phil,
I believe the checker must conclude that, to the best of his professional judgement, the calculation is correct. Otherwise, the checking isn't meaningful.
Mike

On Feb 10, 2011, at 5:45 PM, Phil Heitzenroeder wrote:

Mike & Tim,
Just to get this straight in my mind as we enter an extended phase of
checking and signing off, for something like this particular case- a
force influence matrix - how can someone review this and sign off that
it's correct without essentially re-doing the calculation?  On the
practical level, one can check if the  coil configuration parameters are
correct, but to check the influence coefficients really takes re-doing
the calculation.  What concerns me is that the checker has to sign below
a line which states:

"I have reviewed this calculation and, to my professional satisfaction,
it is properly performed and correct."  

It's hard to say unequivocally "properly performed and correct" for
something like this or ANSYS calculations without re-doing it.   It may
be more reasonable for this sentence to be changed to read:  

I have reviewed this calculation and, to my professional satisfaction,
it is properly performed.  Also, perhaps we should leave a space for
comments, so the checker can add things like "I checked input values"
or "I discussed details with XXX and am satisfied with the methods
used".  This would add credibility to the process.

What do you think?
Phil

-----Original Message-----
From: Bob Simmons
Sent: Thursday, February 10, 2011 3:14 PM
To: Ronald E. Hatcher; Peter Titus; Phil Heitzenroeder
Subject: CALC-13-03-01 Ready for Final Review

Ron, Pete, and Phil:

Attached for your review and comment before we PDF and get signed.

Sorry I did not send around sooner.

BobS
<NSTXU-CALC-13-003-01.doc>

__________________________________________________
Michael D. Williams
Associate Laboratory Director for Engineering and Infrastructure
Princeton Plasma Physics Laboratory
P.O. Box 451
Princeton, New Jersey 08543
Phone: 609-243-2866
Fax: 609-243-2800