7 Replies Latest reply on Aug 13, 2010 1:52 PM by Rick Davin

    Should US gal/day take the recipricol of its canonical value?

    Rick Davin

      The Volume class correctly has 1 barrel = 42 US gallons.

       

      The Volume Flow Rate class has a simple reference for 1 US gal/day equals 1440 US gal/min.  However, the canonical value of 0.090849882816 m3/s doesn't seem right.  If you divide that number by the barrel per day canonical value, you do not get 42 as you would expect.  It looks like the canonical value for US gal/min was mulitplied by 1440 instead of being divided by 1440.

       

      I'm thinking if the math is to be correct that the canonical value for US gal/day should be along the line of 4.38126E-08 m3/s.

       

      And then the bigger question becomes ... does such inverted numbers occur elsewhere when one UOM is based on another simple non-canonical reference?

       

       

        • Re: Should US gal/day take the recipricol of its canonical value?
          Ahmad Fattahi

          I got two points about the calculations you made:

           

          1) 1 US gal/day should be equal to 1/1440 US gal/min. When we make the time smaller (day to minute) we should have a smaller amount of mass in order to end up having the same flow. In my mind this one is the incorrect one.

           

          2) By the same token 1440 US gal/min is in fact equal to 0.090849882816 m3/s. Because 1440 US gal = 5.45 m3. Therefore to get the same flow per second, we need to divide this number by 60: 5.45/60=0.0908.

            • Re: Should US gal/day take the recipricol of its canonical value?
              Rick Davin

              Thanks for the reply, Ahmad.  Besides US gal/day, it seems US gal/hr is also wrong for the same reasoning.  A rate of 1 US gal/hr should be 1 US gal / (60 min), i.e. (1/60) US gal/min.

               

               

                • Re: Should US gal/day take the recipricol of its canonical value?
                  Ahmad Fattahi

                  Thanks Rick for the observation and the good point you brought up. Now that we know the answer to the main question I would encourage you to follow up this case with our tech support at techsupport@osisoft.com or +1-510-297-5828 to make corrections to the documentation.

                    • Re: Should US gal/day take the recipricol of its canonical value?
                      cmanhard

                      You sent me into a bit of a panic this morning, Rick. However, neither US Gallon per day(US Gal/day) or US Gallon per hour (US Gal/h) are supplied in the standard UOM converstion factors of AF (or Sigmafine).  Not sure where they originiated from on your system, but they are (were) defined incorrectly.  As noted, the conversion factor for US Gal/day should be 1/1440th of the US Gal/min rate.  Likewise, US Gal/h should be 1/60th of US Gal/min. 

                       

                      US Gal/min is an OSIsoft supplied UOM and is defined correctly.  Rest assured, we have gone to great lengths to ensure the UOM's provided by OSIsoft adhere to the International System of Units (SI) where applicable, and are correct and significantly accurate as possible.

                        • Re: Should US gal/day take the recipricol of its canonical value?
                          Rick Davin

                          Dang, Chris, you are right.  These were UOM's added internally by someone on our development team.  They also happened to be on every PI system I checked.  Only 1 of 5 people here could have done it, but no one remembers doing it, and it could have happened anytime in the past 12 months.  Seems to be a case of too many cooks spoiling the broth.  And we have no documentation regarding the changes.

                           

                          What would be helpful is some means to identify if a UOM or UOM class is a part of the standard AF shrinkwrapped product.  Those that aren't would get my first priority to go verify the formulas and document them somewhere.

                           

                          My apologies for the panic.

                  • Re: Should US gal/day take the recipricol of its canonical value?
                    Rick Davin

                    Digging further.  We have some 2.1 and 2.2 AF servers.  Wanted to make sure some other people had not tampered with the OSI-provided UOM 's.  No one has.

                     

                    I think the reference of US gal/day is wrong.  It cannot be possibly 1440 US gal/min.  If I am producing at a rate of 1 US gal/day, I would expect my per minute rate to be less than that.  Rather it would be (1/1440) US gal/min. 

                     

                    The good news is this looks to be one isolated instance of a problem with AF UOM, though I haven't checked it out with other rates.