On 11/22/2011 03:17 PM, Bob Camp wrote: > > On Nov 21, 2011, at 10:57 PM, Jim Lux wrote: > >> On 11/21/11 5:15 PM, Bob Camp wrote: >>> Hi >>> >>> You can have a surveyor come out and locate your gizmo to sub one inch accuracy for a lot less than a clock trip costs. A one meter ( or 3 ns) error would be pretty large these days. Both have been demonstrated / proven so often that they aren't really open to challenge. >>> >>> The total error is a sum of lots of things. Location and time of day are the easy stuff... >> >>>> OK, So assume an unlikely huge position uncertainly of one meter. On top of >>>> that let's assume the surveyor got it wrong too and missed by a full >>>> meter. Both of these added together can only account for about 10% >>>> of what they saw. Light moves across one meter in about 3nS You >>>> need to explain 60nS If the result is because of uncertainty in the >>>> location then the we are talking about 20 meters of position error. >> >> >> >> in the first paper, the distance uncertainty was given as 20cm > > > Of which the survey likely contributed next to nothing and stuff like earth tides contributed the majority of the error …. Earth tides account for 2-3 dm of rise, not really contributing much to the 18 m unaccounted for. Also consider that experiment ran for 3 years. Cheers, Magnus

On 11/22/2011 07:28 PM, Javier Serrano wrote: > On Tue, Nov 22, 2011 at 3:17 PM, Bob Camp<lists@rtty.us> wrote: > >>> in the first paper, the distance uncertainty was given as 20cm >> >> >> Of which the survey likely contributed next to nothing and stuff like earth tides contributed the majority of the error …. > > See http://operaweb.lngs.infn.it/Opera/publicnotes/note132.pdf > I think most of the uncertainty came from the non-GPS survey. They had > to stop traffic in the highway (OPERA is in a cavern at an exit of the > highway inside of a tunnel) but they could only close one lane. So > bureaucracy is a more plausible answer :) If the 20 cm uncertainty in position is insufficient, then redoing the survey is certainly an option. Using the 4 marks outside as reference and then using a couple of different surveying methods would certainly help. The LNGS laboratory is a bit of a challenge. It is the LNGS survey that contributes most of the uncertainty. Still, even if surveying is being redone on both labs (which could be motivated by the interest in this find), is this where we expect to find the 18 meters missing? Where did the 60 ns go? Uncompensated cables? Uncompensated equipment? Bias in detectors? Recall, the installations on both ends are quite large ones. If you want to find the measurement error, go and check them. If you want to prove the result, check them. Waving theories and deny the result? Go and check them. Cheers, Magnus