Was working on a project today that was dependent on two surveys we did about nine to ten years ago.?ÿ Those surveys were carried out using a total station to shoot all corners of a quarter section and split it into three metes and bounds tracts.?ÿ Our task today was to cut out a four-sided tract from one of the earlier tracts.?ÿ Today we were using the satellite-driven joystick.?ÿ First three monuments we shoot agree extremely well.?ÿ Number four seems to have a problem.?ÿ Tying that monument to one we had already shot should produce a line running North 89 something something West.?ÿ Instead it produced a line very close to South something something West.?ÿ OH MY GAWD!!!!!?ÿ This would produce an error in all three original tracts.?ÿ Cussing and discussing occurred.?ÿ Realized this line should be parallel with the south section line, so off we go to shoot that line that wasn't really needed for what we were out to do.?ÿ Find the monument needed, shoot it and BAM it agrees with the original North and West bearing, not today's South and West bearing.?ÿ So, back we go, but stop along the way to shoot a reference bar that had been set ten years ago along this troublesome line.?ÿ BAM it, too, agrees with the original description.?ÿ Now we are scratching our heads attempting to figure out how a bar that did not appear to have been moved could have moved nearly nine feet from the record position.?ÿ Guess what.?ÿ Reshoot that wandering bar and discover it is exactly where it is supposed to be this time.?ÿ The distance from monument to monument had been exactly the same as ten years earlier but somehow the bearing came up with different readings between the first time and the second time today.
Living proof that electronics can lie to you as easily as they can give you the truth.?ÿ Wham. Bam.?ÿ Sounds great from an efficiency standpoint but can really suck if you aren't the one who discovers the error.?ÿ We only knew it MIGHT be an error, which led us to take extra steps.?ÿ How many out there are not taking those extra precautions on every shot of every job?
Had an entirely different kind of problem on the same job today that some will find amusing.?ÿ We had two vehicles plus a four-wheeler present.?ÿ At one point I moved gear out of my vehicle and left it for about an hour while working from the other.?ÿ Eventually came back and started to open the passenger door to deposit the first layer of extra clothing that was no longer needed.?ÿ Looked through the window to see a cat in the passenger seat in a position and making motions that indicated it was eating something.?ÿ Two thoughts hit me simultaneously.?ÿ One, that the rear hatch had been left open.?ÿ Two, that there was no way I would be eating the remaining two Zingers in the pack I had opened yesterday.?ÿ Opened the door, watched the cat tear off via the rear hatch and dumped the tiny fraction of Zinger left out on the ground along with the 8000 crumbs that were all over the seat and its contents.
Reminds me of the time I went to an event and then camped at an acquaintance's farm.?ÿ While setting up my tent I left the car door open, and when I came back I found about 8 thin-as-a-rail kittens devouring my loaf of bread I'd brought for PB sandwiches.?ÿ Obviously the people weren't feeding them and momma wasn't able to keep up.
Any one single shot will be subject to error - both random and systematic.
The only way to avoid this is a second independent measure.
That means double working everything.
That's expensive.
?ÿ
Very, very expensive.
And not always justified (plus the Boss never likes the overhead).
On a five acre topo where we are shooting contours, I would say a its waste of time.
The cost of a single bad shot - which can be ignored later in the office - is small.
But what if the control shots were wrong? That's a whole new ballgame.
?ÿ
So...
If it matters, do it a second time.
Preferably as genuinely redundant check that can be verified on-the-spot in-the-field.
Double shoot the control; Close the traverse; loop the level run.
It wont be exactly the same.
The difference is a measure of the general error in your survey.
But you can go home satisfied and relaxed, happy in a job well done.
?ÿ
I am a fan of gps technology, but the 'fuzziness' concerns me.
A "95% confidence interval" means 5% of the shots are wrong.
I always do control and boundary marks twice with a gps
I have too much experience not to.
?ÿ
?ÿ
?ÿ
This is what makes javad users.
N
I am a fan of gps technology, but the 'fuzziness' concerns me.
A "95% confidence interval" means 5% of the shots are wrong.
Actually, a 95% confidence level means there's a 5% chance that any of the shots could be wrong.?ÿ?ÿ
Not the same thing.
All your shots are wrong, most by small amounts. In the long run, 5% of them will be wrong by more than the amount given by that confidence limit.
I was about to post the same thing.
And "wrong" only means "outside the computed error ellipse", which might be well under the required positional tolerance of that feature.
?ÿ
Dude, you are watching too much Emeril Lagasse. ?????ÿ
BAM?ÿ Whatta ya talkin' 'bout, Willis??ÿ BAM
Listen up Buckwheat, put in your hearing aid, put on your glasses, pull up your Depends (which most certainly require changing), and re-shoot everything instead of blaming the instrument. ?????ÿ
But, but, but, it was the instrument that provided the false data.?ÿ Gotta blame the instrument, once I was certain I wasn't responsible for the blunder.