after reading the posts in this forum, i have to admit i not as analytical as some of you. but, i have a conundrum that has me baffled.
i laid a a small house with the gps on Friday. i had previously tied all the pcp's with the gps (hiper v) and they they checked within .02 +/-.
i set up Friday and localized on the pcp's. checked in perfect. as a check, i also re-staked the pcp's and boundary corners and again checked perfectly.
so i went about laying out a 40 x 65 envelope. after layout, we always tape the cross-ties and ties to the bndy corners. the boundary corners ties were fine. the cross-ties were fine. the long side was fine. but the short side (40 feet wide) was out .2 (long)
i checked the coordinates in the data collector. all coordinates were all calculated correctly, the staked out matched the office coordinates i put into the data collector from drawing the bearings and distances from the plat.
i went back out today and set up on 2 different pcp's. the hubs matched exactly the coordinates input in the data collector from the office. the inverses in the data collector matched exactly 40 x 65.
i remeasured with the tape and the cross-ties checked and the long side checked. the short side again measured 0.2 too long. (note i used a different tape to remeasure)
i can't seem to figure out went wrong or where my probelm lies.
on my gps, i don't need state plane scale factors so it is set to 1.
Unfortunately, that is the nature of RTK at times. Each shot has an error ellipse, and you can get some anomaly that will cause one shot to be out.
Personally, I only use rtk for rough grade staking.
Are you sure you have parallel sides? Not input in a bearing wrong that would then mean all your staked points would obviously fit the design as input.
That's (0.2) about 10' (10 minutes) over 65 feet.
What do you get from your DC if you get true line offsets holding one side and calculating to points on opposite side?
00 and 45, and 65 and 45?
I also agree with Jimmy and wouldn't use GPS for accurate house setout.
That said I'd check your input.
How can the cross ties match if one side is off
Sounds to me like it got pushed out evenly between each end.
billvhill, post: 426149, member: 8398 wrote: How can the cross ties match if one side is off
If he's set out a trapezium the cross ties will match. Measuring the corner angles of what's been set out might be a useful check.
The best you can expect from layout RTK is +/- 0.04' under ideal conditions. Add any type of multpath and you got more error.
With RTK you can set a 2"x 2" hub, but you can't set a tack.
I find it hard to believe any Surveyor would set building corners with rtk. This is what gives them a bad reputation. This is why I hear so many Mason's hating layout done by boundary Surveyors. If they all checked the cross ties as the OP did, this practice would stop.
billvhill, post: 426149, member: 8398 wrote: How can the cross ties match if one side is off
Don't the cross ties of a trapezoid match?
leegreen, post: 426159, member: 2332 wrote: With RTK you can set a 2"x 2" hub, but you can't set a tack.
Sent from my iPhone using Tapatalk
I agree with others that you shouldn't do layout with GPS. But set one side with GPS and either turn angles with a totalstation or last resort use 3-4-5 and a tape measure.
Sent from my SM-G925V using Tapatalk
I have had luck setting offsets, locating those, then swing ties to the desired point. 3 tie points minimum.
I would stake a building with hubs, two with tacks. Let the form crews work from there. That would be for cheap contractors not paying for a TS layout of every corner with checks.
Sent from my SM-G930V using Tapatalk
so then with this logic, the GPS is useless. you won't layout anything with with GPS but you will use to tie in boundary corners or set main control points. then these points will be out .04 x .04 plus (your logic). so basically i spent $25000 for a rough layout tool.
as far as concrete work, i have never seen a foundation installer use any type of instrument other than a string line. have you ever looked down the form board line (even with brand new lumber) it fluctuates at least a 1/2 inch (both ways).
i'm sorry i give surveyor's a bad name by using the current technology, checking every which way i can and give the client 110%. Again, i don't have a phd in surveying or know the 100 % workings of the GPS system, but i believe there is error in any instrument whether it be GPS, t2 and tape, total station, etc. and humans. all i can do is the best i can, with all my ability and check, check check!
Allowable, or expected, error is the critical element. Just how much "slop" is permitted for a given application. Scott Zelenak's work at the World Trade Center site is vastly different from a six-mile long route survey for layout of a new highway. Traversing a standard section in PLSSia is not the same as locating setback lines for a $10,000,000 building on a 48-foot wide lot.
cls5095, post: 426171, member: 6344 wrote: ....you won't layout anything with with GPS but you will use to tie in boundary corners or set main control points. then these points will be out .04 x .04 plus (your logic). so basically i spent $25000 for a rough layout tool....
If you tie points points 1000 feet apart to an accuracy of +/-.02' that's an error ratio of 1 in 50,000. That's good. If you tie points points 45 feet apart to an accuracy of +/-.02' that's 1 in 2,250. That doesn't cut it. The appropriate tool for building layout is the total station. In fact, a transit and steel tape would be still better for short distances like this.
Nevertheless, I doubt that GPS measurement is the source of your .2' bust. There is something in your calcs that isn't right. It happens.
cls5095~
Please take the information you get here as that,,, information. Don't be offended. Everyone will make a comment. Everyone will do it different. If you want to layout with GPS. So be it It can be done. I think you are on the right track in being very cautious and checking distances, etc, etc. Best thing you did. You are aware of the errors that is unavoidable.
Did you draw the 40x65 square or did you use the engineer/architects drawing? I've seen where the engineer/architect wanted the building to be offset from the property lines. Only issue was the property lines (most of the time) are not at 90å¡. They offset the property lines then trimmed/extended the lines to connect. Copied and connected the other building lines.
There is something off (apparently or you wouldn't be asking.). It happens. I'm still looking for .5 foot of elevation error.
Good luck and let us know what you find.
I got handed a parallelogram once for stake-out, opposite corner inverse aren't equal in that case. Someone in the office had simply offset the boundary lines in an almost square lot.
The quickest check on the math is inversing the opposite sides for the same distances, then confirming the cross-tie inverses are equal.
Property ties on a long skinny lot work best to insure the front setback but, don't work well for side ties. A quick stacking of two laths and eyballing the side ties should show you 0.2'
Per Pythagorus, the cross ties of a 40.00' x 65.00' rectangle should measure 76.32'. If your inverses say otherwise we have narrowed the search for the source of your error.
cls5095, post: 426171, member: 6344 wrote: these points will be out .04 x .04 plus (your logic). so basically i spent $25000 for a rough layout tool
Sorry to have enlighten you, I'm just the messenger. Yes, RTK is a rough layout tool. It will get you inside a golf ball size error ellipse with 95% confidence under ideal conditions. With multiple epoch averaging, taken 20 minutes apart (change of SV's), then yes you can get RTK accuracy of +/-.02'. But can you afford to take that much time for layout.
Just a few days ago I had to inform a licensed surveyor that he was measuring the ARP incorrectly when using a tripod for static observations with OPUS. He was not happy either, as he'd been doing this for five years, without a single check.
Scotland, i don't take offense. but, its seems everyone has a different take on there surveying style. i don't have anybody in my area to talk to so this is my resource.
i did come in today and recalculated my entire layout. it matched what i did Friday. i even went out with my total station, rechecked the hubs, and found them to entirety reasonable. an .04 x .04 error would be 0.06. i honestly did not have a 0.06 on any of my 4 hubs. still baffled.
as far as OPUS, that aspect of surveying i have never tried. or no one has explained.
leegreen, post: 426182, member: 2332 wrote: ....yes you can get RTK accuracy of +/-.02'. But can you afford to take that much time for layout.(?)
Good point. In order to achieve 0.02' accuracy with RTK you have to occupy your points for between 15 seconds and 5 minutes. Twice over, at least. That makes economic sense when you are tying up to a few dozen control & boundary points in a day. It doesn't work, economically, for stakeout work.