A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Bruce J. Pennino
Date: 2013 Mar 28, 21:16 -0400
----- Original Message -----From: Brad MorrisSent: Thursday, March 28, 2013 7:20 PMSubject: [NavList] Re: Vertical Angle Measurement Errors
I had a good look thru your data. Here are some thoughts
I looked for the min and max readings in any given column, to find the span of the data. Once the span was understood, I can divide by two and define a plus minus range of data. Note that I did not find the mean or find sigma for a statistical distribution. I wanted a feel for your data, not a forensic analysis!
It appears that the data from inside your bedroom, thru the storm window runs around +/- 15 Arc seconds while the data when the tripod was firmly on the ground was more like +/- 9 seconds. Everything deflects under load. Its entirely possible that the floor was moving under the tripod and your feet. If it were me, I'd keep to a firm footing for the tripod and stay out of the building.
It also seemed like when you were aiming at a distinct, man made artifact, instead of the "mean tree line", the repeatability was better. Of course, the horizon doesn't have a nifty piece of plumbing to aim at!
When moving from 90 deg measurements to 270 deg measurements, you rotated in the horizontal plane. Further, my expectation is that this was a precision 180 deg swing done by index and not a "yup, that's about 180" kinda move. There's no notation about this, so I'm just confirming.
When aiming at a distinct target, up vs down was very repeatable. Your theodolite mechanics seem sound with little or no lost motion (slop).
I'm also guessing that the measurements are recorded in the order you took them. It looks like practice improved your result. That could be just the tripod footing.
The U/D front face U/D back face seems a very robust procedure. Average all four to get one true dip value. I like it!
All in all, very nice data. A statistical analysis will yield your repeatability and accuracy numbers. From this peanut seat, I'd say your data is plenty good enough.
BradOn Mar 28, 2013 11:15 AM, "Bruce J. Pennino" <bpennino.ce---net> wrote:
I've attached some angle data I've recorded with my theodolite. My goals are stated. I'm in the process of evaluating this information, and hope that others will look it over. I'll post my conclusions in a couple of days, and then we can discuss further. I've tried to give all pertinent information within the attachment.
A very little about the data..... No data points were excluded and there are no apparent blunders. There are some outliers. The data was recorded very carefully by reading the angle, rereading, writing it down, recheck reading, recheck writing, and then going on to next point. It only takes maybe 1 minute per value, so it goes quickly once the instrument was set up. The device is an earlier total station which requires "eyeball" reading of an index....no digital output. The instrument calibration is extremely stable, but it has not been to a shop in 10 years. I check it frequently in various ways, especially against other work I do with another instrument; or on occasion against work done by a professional land surveyor I know who is an absolute fanatic about precision, accuracy,errors,etc.
We are starting to learn a little. I hope to go next week and measure dip from a precisely known elevated position, Blue Hills Observatory outside Boston.
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
f1-Topcon-Error-Determination-D.doc (no preview available)
View and reply to this message: http://fer3.com/arc/m2.aspx?i=123192
View and reply to this message: http://fer3.com/arc/m2.aspx?i=123205