KevinFoshee, post: 350998, member: 8314 wrote: I prefer straight-line offset to 90 degree offsets.
Either one works well if executed with care. And I often use 3 offset points in order to add redundancy.
Jim Frame, post: 351000, member: 10 wrote: Either one works well if executed with care. And I often use 3 offset points in order to add redundancy.
Yes, I agree that a 3 point resection is best. This is why this CoGo function was recently added to J-Field.
I know a surveyor who does dd int (about 90å¡) BUT he always makes his crew do a 3rd shot. With a 3rd dist to the corner that needs tied. So that he can catch blunders, in the measurements, or in the GPS. I think it's a good idea.
N
Hey, Matt, that's the idea.... but to have it incorporated into Jfield is a super idea!!
N
Straight line works well IF you use a string or straight edge (folding rule?) between the offset points. Not so well if you have to eyeball the line or are going for the center of a post that is in the way of the string.
Straight line works well IF you use a string or straight edge (folding rule?) between the offset points. Not so well if you have to eyeball the line or are going for the center of a post that is in the way of the string.
Straight line works well IF you use a string or straight edge (folding rule?) between the offset points. Not so well if you have to eyeball the line or are going for the center of a post that is in the way of the string.
Nate The Surveyor, post: 351021, member: 291 wrote: I know a surveyor who ... always makes his crew do a 3rd shot. ...
Actually, you know 2 such surveyors, if you include on-line acquantances.
What I used to do with my Topcon GPS, to set a corner in the woods is:
Use float to get within 10' or so. Avoid trees. Set it up, wait 15 mins or so, get a fix. Store it. Use compass and box tape, set cor. Move up to corner, (Unless it is too close to a tree) In that case, move to another location, (change the multipath environment) and regain a fix, and check the corner. IF it is within a tenth or so, then average it, and pound it.
What I do with the Javad, is often use ONE fixed engine to get within 2 ft. (One fixed engine is pretty weak. It has very liberal tolerances, for going fixed) and then, let it go. It will eventually come in fixed with several engines. Now you can set the corner. Move up to the rebar. Shoot it with verify. it has always been within 0.05' so far. It's pretty cool. Without verify, you are running a little risk, of having it wrong. Of course, you can reset the GPS engines, manually. etc.
N
Old times:
Set the gps up, in the woods. Get it FIXED, and store a shot. RESET the GPS, and repeat. Raise rod, repeat. Lower rod, repeat.
After a while, and you have about 5 coords, for the same point, average them, and go to the next.
Modern times.
Set the GPS up, in the woods. Tell it you what confidence factor you want. Tell it how many shots to take. It does it's thing. While you get witness trees.
It does "Old times" for you. And, you can EVEN tell it to auto store, when it meets criteria. And re-shoot. You can even leave it out all night. It will store, and retest, and repeat all night. In the morning, you will have a pile of several hundred shots. Then, you can average THESE, and you now have near POST processed, results, in an RTK system. I think I'd do that and leave it while I went and ate lunch, but overnight would be overkill. But is sure will let you see what it does.
N
@nate have you tried reoccupying a "dodgy" place and comparing the results over a period of time?
Leaving it as explained above to say collect 5 minutes of useful data then repeat later?
Yes, I have. Worst difference so far, is about 0.12'. The poorer shot, showed a LARGE box, and poor RMS value. It was easy to see which one was poor. (This was on a job I previously had Total station and Post Processed static on.) I was playing with it... It was same job (Post Number 2) as listed up above, but a different point.
N