Hello,
I've recently worked with a group who had never processed through opus and in general is making me wonder. While I am not an expert I would like your opinion on some of these methods:
Only working within the area between your set up station and your backsight station?
Setting control via GPS for a 2d project collected in GPS, as opposed to bringing your grid to ground?
Rotating the the GPS pole 180 or through the cardinal directions than averaging for more accuracy?
Having the R10 readout face in the same plane as the TSC3?
In any event your input would be appreciated.
There's room for a book, about practical application of gps, along with excellent, good, and poor procedure.
Not too big of a book, but one that covers basic common pitfalls.
"Practical GPS surveying for dummies" comes to mind.
N
1. This is an old-time rule that isn't a bad one, it forces the errors in backsighting and leveling to be minimized. It is also valid when setting up a GPS base, the closer the better, although I would argue that it can be taken to extreme and hurt productivity which is not good for business or taking care of your clients, you need to know your error budget and what is necessary for the job. If I'm setting very tight points, I don't want to be on a hill 1500' away laying out building corners to a 1/8". ?ÿ
2. I don't understand the argument. Why would setting up control with GPS not include ground? GPS has nothing to do with SPC, it is only a tool that can get you there if you tell it to, if you tell it to be on ground it will spit out control on ground. GPS is an excellent tool for getting control on SPC, UTM, LDP, Surface Coordinates, ground distances, it's just a dumb machine controlled by the operator.?ÿ
3. Nothing wrong with doing that, it takes into consideration your plumb errors, gives you more than one location and allows your point to be more accurate.?ÿ
4. Lost me there, no idea why that would mean anything, unless you want to watch the blinking lights on the R10. If there is a reason, I would like to know what it is; always interested in learning.
4. Lost me there, no idea why that would mean anything, unless you want to watch the blinking lights on the R10. If there is a reason, I would like to know what it is; always interested in learning.
?ÿ
With the R8-3 the factory modeled phase center was something like 0.01' towards the MUI (iirc). I don't know about the R10. This wasn't an accurate certified phase center.
Rotating the the GPS pole 180 or through the cardinal directions than averaging for more accuracy?
I would argue that consistently having the antenna oriented the same way would produce more accurate coordinates, especially if the software knows how to correct for that particular antenna mode, and assuming your pole was accurately plumbed.?ÿ
If the major source of error was in the pole bubble, then rotating would average that out for world coordinates, but would add in the now incorrectly modeled antenna offset. I'd hope plumbing wasn't the biggest error anyway. Consistent orientation would cancel out errors for legs between project points.
By habit, I have always had my antenna connection pointed north even tough it does not affect the results.
Only do L1 static so never have sent anything thru opus.
Once or twice have been required to have specific location for a wellsite and a topo for some state water department.
All my boundary survey is WGS84 and it gives me true north and has been a great investment.
Around here there is very little difference between between grid and ground results so I do not ever worry about a scale factor as my results from GPS to TS distances have always been 0.04?ñ foot.
It is really going to upset me if L1 is ever turned off for out use.
Rotating the the GPS pole 180 or through the cardinal directions than averaging for more accuracy?
I would argue that consistently having the antenna oriented the same way would produce more accurate coordinates, especially if the software knows how to correct for that particular antenna mode, and assuming your pole was accurately plumbed.?ÿ
If the major source of error was in the pole bubble, then rotating would average that out for world coordinates, but would add in the now incorrectly modeled antenna offset. I'd hope plumbing wasn't the biggest error anyway. Consistent orientation would cancel out errors for legs between project points.
Bill, in the old days it made a difference, with the R10 I see nothing in pointing it any direction, I assume the javad is the same
Thanks to all, here is a better explanation of#2
The project is along a 1.5 mile major arterial with PLSS monuments at centerline intersections relatively. Object is to collect a 2d look of light poles, traffic signals,bushes,trees, stumps, boulders,property wall, back of curb(only at the returns given a tangent) back of sidewalk visible utilitys for a landscape architect. The project leader was proposing to collect PLSS record corners 1/2 mile on either side, average for a CGF then bring everything to ground.?ÿ Given the no 3d model required and the difference at the 4 th decimal place would yield a difference of .01 to .05 depending on how far away from the scaling point from grid to ground and National cad Standards I suggested setting a control point about halfway, using a here point to get started, collecting static as we collect RTK, download to opus, find the difference between record and measured on PLSS and shift, compare that to Opus vs here, translate measured to record, bring to ground or leave in grid. State Plane is your control....that is my argument, why would you need to set control?
Thanks to all, here is a better explanation of#2
The project is along a 1.5 mile major arterial with PLSS monuments at centerline intersections relatively. Object is to collect a 2d look of light poles, traffic signals,bushes,trees, stumps, boulders,property wall, back of curb(only at the returns given a tangent) back of sidewalk visible utilitys for a landscape architect. The project leader was proposing to collect PLSS record corners 1/2 mile on either side, average for a CGF then bring everything to ground.?ÿ Given the no 3d model required and the difference at the 4 th decimal place would yield a difference of .01 to .05 depending on how far away from the scaling point from grid to ground and National cad Standards I suggested setting a control point about halfway, using a here point to get started, collecting static as we collect RTK, download to opus, find the difference between record and measured on PLSS and shift, compare that to Opus vs here, translate measured to record, bring to ground or leave in grid. State Plane is your control....that is my argument, why would you need to set control?
That just sounds like an incredible amount of gymnastics to avoid using something like Starnet (or any of several other adjustment packages) that can combine GPS and terrestrial measurements seamlessly and output either grid or ground scaled results, as may be your preference.?ÿ
I'm with Mark on that, you can figure out a scale factor for the project in a couple of minutes on the computer, set it up and fix it for the project before you collect any data in the field. Go to Google and get a few elevations matched up with Lat, Longs, put it in the computer and mean a SF. You can also do what I prefer and look at real worldish state plane points around the project (if you want ground distances in state plane), inverse between, compare the ground vs grid distances and there you have it, your scale factor, ground/grid; or if your program prefers grid/ground. Cut it off at 6 decimal places which gives 1ppm or 20 foot in elevation change across the area.?ÿ
As an aside, you are going to locate 1.5 miles of detail without any control??ÿ
I strongly suggest at least a few intervisible control points be set in areas that can be easily accessed.?ÿ
The general rule is 1500' or less in rural areas, one point per block in urban areas.
This landscape project needs no elevations?
Yeah, I've gotten plans from landscape architects, you basically go to the field and design it yourself, fixing all the issues that are left over from a "concept design".
I'd at least pay attention to elevations, it doesn't take much longer.
This isn't an add-on to a much larger project is it? Usually this would be in done during some reconstruction and DOT or City/County would have some data-bases already done, I would use it if it's available, heck there may be some control already. ?ÿ
?ÿ
The Topcon Hiper V told you to always face the panel North. Some receivers are weird like that. But I think it is all nonsense due to their accuracy and ppm.?ÿ
The Topcon Hiper V told you to always face the panel North. Some receivers are weird like that. But I think it is all nonsense due to their accuracy and ppm.?ÿ
The Topcon Hiper V told you to always face the panel North. Some receivers are weird like that. But I think it is all nonsense due to their accuracy and ppm.?ÿ
Maybe, BUT ALL NGS antenna models are done with antennas oriented to North using the manufacturer specified "reference point", older antennas had a north mark, most now specify something else, the antenna connector is common for external antennas. Most CORS operators also follow the same convention so that a particular antenna has the same reference to north as the calibration. Orienting all antennas in the same direction as they were during calibration actually can improve GNSS results, although these days the phase eccentricities are generally so small that ignoring orientation of the antenna doesn't appreciably degrade the results. Old habits die hard, I still orient the coax connector north on all my occupations other than topo shots.
SHG
It was my understanding back in the day, that the antenna pairs, base/rover, both orientated north cancelled out any error between them. However if you mix a compact L1/L2 with a micro-centered 5800 then there was no reason to do it.?ÿ
?ÿ
If your software uses the antenna calibration data, then whether you mix or match you are better off orienting the antennas.?ÿ
Granted, the horizontal calibration values are small for most modern antennas, and even some older ones: TRM33429.00+GP NONE L1/L2 microcentered is within a mm horizontally.?ÿ My TRM14532 has 3 mm or so horizontally, which is worth correcting.?ÿ Although that isn't the biggest error in most occupations, pointing it south instead of north would give 6 mm of offset.
https://www.ngs.noaa.gov/ANTCAL
If your software uses the antenna calibration data, then whether you mix or match you are better off orienting the antennas.?ÿ
Granted, the horizontal calibration values are small for most modern antennas, and even some older ones: TRM33429.00+GP NONE L1/L2 microcentered is within a mm horizontally.?ÿ My TRM14532 has 3 mm or so horizontally, which is worth correcting.?ÿ Although that isn't the biggest error in most occupations, pointing it south instead of north would give 6 mm of offset.
You didn't get the link quite right, or they changed it on you.?ÿ Correct direct link to the page is here:
https://www.ngs.noaa.gov/ANTCAL/#
?ÿ
I guess the slash on the end is required.?ÿ Most web sites don't care about that.
Hello Nate,
I've got a reasonable library, thanks for the recommendation.
Paul
?ÿ
Having the R10 readout face in the same plane as the TSC3?
In any event your input would be appreciated.
That will be to get the correct orientation of the e-bubble on the R10.?ÿ?ÿ
One of the amazing differences between conventional survey gear, (total sta. theodolite etc) and static gps, is the fact that you don't need a backsite with gps. And, that directions, or bearings are so accurate, and consistent.
That's one of the first things...
N