But with F2f I can run data through StarNet, import to CAD, and process a field file of topo in 10 minutes.
You're a better man than I. I do a lot of topo, and it's a rare day that I don't botch an HI or an HR or a description or a check point number (often a selection of each). I make a note in the raw data file of the goofs -- I mostly catch them as soon as I make them -- but that means editing the raw data file before I can adjust and move onto the F2F business. I don't think I've ever gotten that far in 10 minutes.
@norman-oklahoma I appreciate your (and others) time and input. I only get to work on this when I have the time so it's usually days before I can come back and look at the comments. I'll give your suggestions a try and let you know how it goes.
Why does at 5551 bs 5550 fs 5552 have an 84.66 second standard error at line 220 and line 2945?
Because the foresight length is only 10 feet. With a target centering error of 0.003', plus other contributors, it makes sense.
BTW - I usually (at least) start with 0.003' for instrument centering, 0.005' for target centering, and 0.01' for vertical centering. Do that and you can probably dial back on the angular and distance a priori errors. I generally go with the instruments specification on angular and distance errors, then mess with the target centering errors if necessary to tweak to a passing Chi-square. Not too much, though.
I don't recall the last time I took a measurement that had that much of an imbalance between BS and FS, so it is not something I just know through experience. I had to run the numbers and sure enough - when you have that 10 foot shot in the denominator it bumps the results of the target centering error formula and the instrument centering formula way up.
that much of an imbalance between BS and FS,
You are thinking in terms of a traditional chain traverse. This may be a case where he side tied a boundary mark from one control that was very close, then side tied it again from another control up the line. There would then be a weighted adjustment of the double tie. You just can't do that in a rigorous way with the compass rule.
Why does at 5551 bs 5550 fs 5552 have an 84.66 second standard error at line 220 and line 2945?
Because the foresight length is only 10 feet. With a target centering error of 0.003', plus other contributors, it makes sense.
BTW - I usually (at least) start with 0.003' for instrument centering, 0.005' for target centering, and 0.01' for vertical centering. Do that and you can probably dial back on the angular and distance a priori errors. I generally go with the instruments specification on angular and distance errors, then mess with the target centering errors if necessary to tweak to a passing Chi-square. Not too much, though.
I don't recall the last time I took a measurement that had that much of an imbalance between BS and FS, so it is not something I just know through experience. I had to run the numbers and sure enough - when you have that 10 foot shot in the denominator it bumps the results of the target centering error formula and the instrument centering formula way up.
I side shot an iron pipe from one control point. Then occupied that iron pipe so I could shoot another iron pipe. I didn't want to do anything with a 10' back sight but I did back sight my control point that was 10' away so I could turn to a further away point to back sight, which I did and then shot the second iron pipe. I wasn't really worried about the elevations of the irons and normally I would have imported those as side shots after processing my traverse in civil 3d. I don't think Starnet will let you shoot side shots from side shots. It kept giving me an error. So I made the first one an M record and just floated the vertical angle for those 2 irons.
Do you guys double your angles when doing a control traverse? If you have an active tracking prism (I use a trimble mt1000) do you use it actively or passively while traversing?
Do you guys double your angles when doing a control traverse? If you have an active tracking prism (I use a trimble mt1000) do you use it actively or passively while traversing?
yes. Usually 2 sets for control and one set for boundary ties. Although I posted the other day about how great single shots with an AP20 tilt sensor were working out.
I’m using Leica, so it’s passive. I have used Trimble and had no issue about using the active target on the precision front, but usually used passive targets for control and boundary
Do you guys double your angles when doing a control traverse? If you have an active tracking prism (I use a trimble mt1000) do you use it actively or passively while traversing?
I prefer using the traverse kits when traversing. However at times the pole and mt1000 is good enough I place it in semi active mode for control and property corners and any stake out for things that need to be tighter for elevation. That prism is not the most accurate the specs on it for sighting is up there limits than a single glass prism but good enough for some things. Are you running Trimble access or spectra old tds survey pro software what instrument are you running what’s the specs angular and edm I believe StarNet probably still has the routine you can run to see what you see to do based on equipment and procedure specs in order to achieve the results required. Aka how many rounds sets direct reverse based on travers length and average distance in set ups centering error usually above 2mm for a good traverse kit pole and mt1000 will be more bubble is only so tight sighting is higher etc when I locate rods property corners and or extra control I make sure one of the glasses on the 360 prism is facing directly at the instrument I use to keep a cheat dot On opposite side of nail polish don’t tell the wife I borrowed a bit please lol . If you are doing least squares a simple easy quick thing is after you observe the corner stay out re establish a BS now sighting you and then close back to you control that mean angle and the mean angle to you should add to 360 A poor boys way of closing the horizon and having a sanity check remember your traverse can close 1:10000000000 but one busted side shot to a monument is as good as the boundary can be
Do you guys double your angles when doing a control traverse? If you have an active tracking prism (I use a trimble mt1000) do you use it actively or passively while traversing?
I prefer using the traverse kits when traversing. However at times the pole and mt1000 is good enough I place it in semi active mode for control and property corners and any stake out for things that need to be tighter for elevation. That prism is not the most accurate the specs on it for sighting is up there limits than a single glass prism but good enough for some things. Are you running Trimble access or spectra old tds survey pro software what instrument are you running what’s the specs angular and edm I believe StarNet probably still has the routine you can run to see what you see to do based on equipment and procedure specs in order to achieve the results required. Aka how many rounds sets direct reverse based on travers length and average distance in set ups centering error usually above 2mm for a good traverse kit pole and mt1000 will be more bubble is only so tight sighting is higher etc when I locate rods property corners and or extra control I make sure one of the glasses on the 360 prism is facing directly at the instrument I use to keep a cheat dot On opposite side of nail polish don’t tell the wife I borrowed a bit please lol . If you are doing least squares a simple easy quick thing is after you observe the corner stay out re establish a BS now sighting you and then close back to you control that mean angle and the mean angle to you should add to 360 A poor boys way of closing the horizon and having a sanity check remember your traverse can close 1:10000000000 but one busted side shot to a monument is as good as the boundary can be
My setup for everything conventional is a trimble s5 3" instrument, a tsc 7 running survey pro and an mt1000 on a prism pole with a bipod.
"stay out re establish a BS now sighting you and then close back to you control" sorry, I didn't understand this part
"stay out re establish a BS now sighting you and then close back to you control" sorry, I didn't understand this part
I think he means that after you've shot your monument, use the monument as the backsight point and then tie into another control point (in many cases, the initial backsight point). That way you have a redundant tie to the monument. This is SOP for me.
@surveygrug sorry I was on some pain meds last night and it does not make sense good thing the wife came in right after and said get off role before you say something to crazy lol. I am not much on pain pills or any drug for that matter but I had to have some sleep and was hurting a bit. So example. Sit on point 1 BS point 2 this is on main traverse line. Turn to found property corner locate it. Now stay sitting on point 1. BS the corner you just located turn back to previous BS. You can change break the set up on your rod. Usually don’t walk back and break set up at gun location but that’s even better so it helps prove centering is correct. I usually turn direct and reverse just old habits die hard. To all property corners. If I get lucky and can see the same property corner from a different traverse station the. I just tie it twice from two different set ups and not worry about closing the horizon. The S5 is a workhorse I like that one. I don’t have a lot of time interacting with survey pro. Back when it was as TDS on a 48gx and a little time when it was on the old recon but nothing much in recent history. We have a TSC3 with it running on that hardly ever been used. But they have moved everything to Trimble access ow on TSC7 and all s7 robots. Trimble r12i’s. Sorry about the wording last night.
I changed the points from fixed to the 0.001 .001 .001 you suggested. not much changed, chi square passes but I'm still seeing the station coordinate error elipses approaching 0.30'. Do you think it's something I should be concerned with?
Under the "Adjustment Statistical Summary" you will now have an "error factor" for coordinates as well as angles, distances, zenith angles, etc. Play with the "0.001 0.001 0.001" standard errors until you find a number that yeild a number approaching 1.0 for the coordinate values. Play with your a priori errors for the other things until you find numbers that similarly yield ±1.0 error factors in the Adjustment Statistical Summary. (I usually leave the angle, distance, zenith errors at the instruments spec sheet rating and play with the centering errors.)
None of this is going to change your final coordinates significantly but it will be educational.
I'm not worried about those station error ellipses. Neither should you be. Those number are just telling you what our reality is. Possibly you can trim that down in the future by refining your field procedures, but not by all that much.
I'm processing my first project using starnet and since I don't fully understand it yet I'm also processing it in Civil 3d. I'm running my traverses from gps pair to pair around a lake. one of my traverses busts 1.20 feet.I'm fairly confident that it's not the gps coords and that I must have busted an angle.
my chi square passes and all of my numbers look good except for the station coordinate error ellipses which range from .01-.30 but none of the ones with higher error ellipses are in the section that busted. How would I ever recognize this error using only the information that starnet is giving me. both starnet and c3d are giving me basically the same results except for the leg that busted. I've attached my listing file just in case one of you has time to look at it.
I'm processing my first project using starnet and since I don't fully understand it yet I'm also processing it in Civil 3d. I'm running my traverses from gps pair to pair around a lake. one of my traverses busts 1.20 feet.I'm fairly confident that it's not the gps coords and that I must have busted an angle.
my chi square passes and all of my numbers look good except for the station coordinate error ellipses which range from .01-.30 but none of the ones with higher error ellipses are in the section that busted. How would I ever recognize this error using only the information that starnet is giving me. both starnet and c3d are giving me basically the same results except for the leg that busted. I've attached my listing file just in case one of you has time to look at it.
I've had only a quick look but it jumps out at me that you have the coordinate system set to "Local" while you say that you are holding GPS'd azimuth pairs. Set your coordinate system to the appropriate state plane zone and try again.
Nevertheless, I'm seeing residuals that I consider acceptable throughout. Out of all the points you have established I see only a single one with any component of the 95% error ellipses north of 0.10'. Give yourself a pat on the back and call it a good job. Remember that those numbers are the result of all the accumulated errors in all the measured legs it took to get to that point.
As far as your question about how you tell .... you have to look at all the numbers in the report. There is no single number that tells the whole story. BTW - You can set the listing report to sort residuals by standard residual ratio, which simplifies finding the big ones.,
im using gps points that we scaled to horizontal ground. our field data was collected in hg and when i tried setting starnet to grid it was reducing all of my distances to grid and failing the chi square test. i'm fairly confident in my control coordinates. i have other traverses running to and from these same points that close.
im using gps points that we scaled to horizontal ground.
The local setting is appropriate then. I'm not saying your approach is wrong but it isn't mine. I'd keep the control "unscaled", adjust in the projection system, and, if ground scale coordinate are desired, export scaled to ground coordinates using the tools provided by StarNet.
The accuracy standards for class A surveys say that neither axis of the 95% confidence level error ellipse for any control point or property corner shall exceed 0.10'. When I look at the station coordinate error ellipses ranging from 0.01 to almost 0.30 am I looking at the correct numbers? I have to record this particular survey and I also have to explain the numbers to my boss. I'm extremely skeptical of starnet at this point because if I wasn't also processing this data in civil 3d I wouldn't know that I have a 1.2' bust in one leg of my traverse. I could be wrong. I'll find out after I re-turn that portion of the traverse.
I suspect that your spec is going to apply to the position relative to adjacent monuments, not to the network accuracy.
This report is the one you should probably be looking at for compliance with that spec: