I've been working this weekend on what didn't seem to be that unusual a project. It was just a resurvey of a 5-acre lot in a subdivision that had been surveyed and platted in 1995. For something surveyed in 1995, I expect good quality measurements, but it looked as if the whole subdivision had been staked with a total station with the wrong prism constant plugged in. Generally all lines that look as if they would have been run in the course of staking the subdivision are averaging about 36mm (0.12 ft) long. It doesn't matter whether it's a chord 35.36 ft. long or a lot line about 500 ft. long between corners. They are look as if they have approximately the same constant error between what appear to be the original markers.
Granted, 36mm isn't the end of the world, but it explains why, among other things, the right-of-way that was platted as 60.00 ft. is staked about 0.24 ft. wider than that.
As best I can recall, the surveyor who laid that subdivision out was using one of the early Zeiss Eltas. If they were like the Elta 46R that I was using at the time, you had to set the prism constant by dip switches on a board in the instrument and the convention was an odd one in that the instrument left the factory calibrated for a Zeiss prism which had a offset of about -34mm (if memory serves me). The subdivision looks as if it would be what you'd get by setting the offset correction to -0mm and using a -30mm prism. I'll probably have to push my luck and ask him.
I'm sorry to say that I did that one time; fortunately the error was caught by a county crew checking the monumentation before the plat was recorded.
The error was seemingly at random until I realized that since I had staked nearly all the corners from my control points, the erroneous monuments were on line from the control to where they should have been set but 0.10 feet in error. Fortunately, I had run the control traverse with the correct offset.
The problem was caused by a switch from a -30mm prism for control to a 0 offset prism for topo which never got set back to -30mm for setting lot corners later.
I had a Geodimeter 620 do similar. It stared measuring 0.11 feet off (I think short) on every shot. Didn't matter whether it was 5 feet or 1000 feet. I should have caught it when I tied some control. I got thinking about it on the way home (bad shots to control) for the day and set up on the driveway and did some checking. We had to send it in twice as the first time it came back and wasn't fixed. The second time they had to replace some board in the instrument and it works but I've never really trusted it since.
The job was way out of town also, 2-1/2 drive each way, so it caused me some grief.
Well, it goes without saying that a surveyor (and his crew, if applicable) should be intimately familiar with the problem of an incorrect prism constant. Unfortunately, I think there are those who do not pay attention enough to these sorts of things. That said, even for those who pay attention, it is an easy mistake to make.
I run my Sokkia SRX3 robot for both conventional 2 man surveying, as well as in robotic mode. I use -30 mm prisms for most of my work. However, when I am alone, doing robotic work, I find the 360 degree prism is quite helpful in maintaining lock. It has a prism constant of -7 mm.
Occasionally at the beginning of a day, following a robot session, I will forget to switch back to -30 mm on the data collector. As a result I will sometimes have to redo critical shots made prior to the discovery. It is an easy mistake to make, and is not helped by the fact that on my Survey Pro data collector, the icon for the 360 degree prism is EXACTLY the same as the icon for the 30 mm prism, so it takes quite a bit of care to be sure that the setting is correct.
I realize many of you probably set your instruments to send corrected data to the data collector, that is, the prism constant setting is made on the instrument. In my case, I chose to make the setting in Survey Pro, so the instrument sends uncorrected data. Two reasons: 1) When the setting is made in Survey Pro, the prism constant appears in my raw data file, and 2) when I had tried setting the prism constants in the instrument, I would have frequent trouble with the prism constant re-setting itself on the instrument without my having made a request.
> It doesn't matter whether it's a chord 35.36 ft. long or a lot line about 500 ft. long between corners. They are look as if they have approximately the same constant error between what appear to be the original markers.
I think you've found the "smoking gun" there Mr. Holmes, ..er, Kent. I can't think of any other explanation. Even a really bad bubble on a reflector rod would (in all probability) have some inconsistency.
> I think you've found the "smoking gun" there Mr. Holmes,
Yes, and it adds an interesting wrinkle to replacing missing corner markers since you have to make some sort of reasonable reconstruction of how things actually were laid out in order to put the extra 0.12 ft. where it originally was. Equitable proportionment doesn't make much sense where the distances are either essentially precisely right or precisely wrong by a predictable amount.
Layout mechanics
I'm guessing the original layout was "line work"; old school style of setting up on PIs or PCs. No 'radial' stake out apparently if the constant error is that predictable.
Guess you'll just have to figure out where his setups were at. Good luck.
Oh, one of punchlines to this story was that when a TexDOT right-of-way survey was made to take a sliver off one corner of the lot, the contract surveyors didn't apparently realize that the right-of-way of the subdivision street really was originally staked about 0.24 ft wider than the plat indicated. They somewhat arbitrarily chose one of the right-of-way lines as perfect and then set a new brass tablet in concrete in a location that's actually not on the other line of the street.
It was quite clear what had happened, but since the tablet wasn't actually on the street right-of-way line, the only feasible fix was to mark where the corner clip line intersected the street line 0.15 ft. distant from the station punch on the disc. 0.15 ft. fell on the edge of the brass disc, so now there is a second punchmark by the "T" in "TEXAS DEPARTMENT OF TRANSPORTATION".
Hey..
look on the bright side, it still hit brass didn't it?
When I worked for ODOT I use to joke about the size (3.5") of our bronze tablets. We bought them that way because we needed them that big!
Layout mechanics
> I'm guessing the original layout was "line work"; old school style of setting up on PIs or PCs. No 'radial' stake out apparently if the constant error is that predictable.
It looked generally like running lines with the exception of street intersections where it looked like a combo of running tangents and pay-for-spray on a couple of other points on the opposite side of the street.
Hey..
> look on the bright side, it still hit brass didn't it?
>
> When I worked for ODOT I use to joke about the size (3.5") of our bronze tablets. We bought them that way because we needed them that big!
Yeah, the Tex DOT tablets are 4 inches in diameter.
"T" in "TEXAS" Good Name For A Country Song
or is that "Tennessee"?
Paul in PA
Once there is two punch marks it becomes a free for all. I've seen them with maybe 20 punch marks, sort of like a territory marking zone for every surveyor that came by. A mini pincushion of sorts. Did you stamp your license number on there also?
> Once there is two punch marks it becomes a free for all. I've seen them with maybe 20 punch marks, sort of like a territory marking zone for every surveyor that came by. A mini pincushion of sorts. Did you stamp your license number on there also?
Well, the punchmark I added will be described in a metes and bounds description I'll prepare for what is now the remainder of a 5.00 acre lot less the corner clip. The same metes and bounds description also demonstrates why the punchmark at the edge actually marks the corner that the folks who set the tablet were trying to mark but failed to hit. The corner clip was a very small parcel and what are almost certainly the original stakes of the lot are still in place, if now out in a highway right-of-way where construction is taking place. So the metes and bounds description will document that evidence (that may not survive construction). The problem is so clean that there simply aren't multiple plausible solutions available. This isn't some indeterminate problem.
Did you really add a second punch mark to a DOT monument?
Well, it's certainly easier to punch the one monument, as opposed to punch all the other's on that side of the road to meet the platted road width.
> Did you really add a second punch mark to a DOT monument?
Sure, of course. The nature of the right-of-way clip was that the contract surveyors placed rod and cap markers at the corners of the clipped parcel and tied them to other original corners of the lot that remain in place. Now, the rod and cap markers are gone, the original corners remain, and there is a newly set tablet supposedly where one of the rod and cap markers was. All there is in the position of where the other rod and cap marker had been is a knot of concrete, the foundation for the other tablet.
So, basically, there is no original monumentation left defining the boundary of clip parcel aside from the lot corners less than 15 ft. away.
The 4" bronze tablet has about a 2" button where the station mark should fall and the corner fell in an indented ring right at the edge of the tablet, so I marked it with a punch. It isn't my client's fault that the DOT contract surveyors can't place markers very well and my client certainly isn't donating his land because of that inability.
What are the options? Call contract surveyors to come out an reset the tablets correctly? Who has time for that? Drill hole through tablet an into concrete to install another mark? in proper position? Possible, but time-consuming and therefore expensive. Punch the corner correctly at the edge of the tablet and give tie by bearing and distance to the station punch at the center of the tablet? Yes.
Why not get TDOT to let you move the marker? While they are at it have them fix all the rest as well.
Who has time for that?
Quickie-Dickie Surveys doesn't for sure!
Man, you're going over to the dark side.
".. is staked about 0.24 ft. wider than that."
That's weird. I imagine stake-out in the Hill Country is quite a bit different from this flat land. Wonder why the row doubles the error.
