Significant numbers

https://sparrowmissions.com/wp-content/uploads/2016/05/The-A-Team.jp g" target="_blank">https://sparrowmissions.com/wp-content/uploads/2016/05/The-A-Team.jp g"/> &imgrefurl= https://sparrowmissions.com/2016/05/09/i-love-it-when-a-plan-comes-together/&h=367&w=500&tbnid=-v1_ZV5H0ZULSM&q=when+a+plan+comes+together+meme&tbnh=95&tbnw=130&usg=AI4_-kRhPTPTdE2fc9TE_DPhJvvTGHh_oA&vet=1&docid=6H5XRvAmu0r1VM&sa=X&ved=2ahUKEwikzJiaq5OUAxWam2oFHSSUNtMQ9QF6BAgFEAQ

What you did beats a calibration by miles.

Stories like these just make me ....... so much happier that my state has a defined set of low distortion projections. 

Nevertheless, it is always nice when a system works like it should. 

We are building a 135 million dollar senior living complex in San Antonio, TX.  We were given site control by the civil design company.  We fire up a rover and dial into a network VRS to start tracking to the points.  Everything was 700'+ offsite to the NE.  I got to reading through the plans and saw where they converted grid to ground by a factor and scaled about 0.0 instead of a centroid.  Those coords are huge down there so the distortion is wild.  Evidently this is standard practice by TXDOT so everyone does it.   I had to do the math on our GCPs and get things back on site so our drone, which flies connected to the network VRS, would align with the site in the drone processing software.  I had to copy all of the CAD files that I overlay on the orthos and move them as well.  A bit of extra work but a fun exercise.  The dirt contractors just calibrate their GPS to the site so they don't realize what's going on. 

This a snip of the ortho image.  The blob of checkerboards to the north is where the provided site control lands in the real world after being scaled about 0,0.  Our site is at the south end.  Network VRS puts our imagery close to where it's supposed to be and the adjusted GCP's tightens it up.  

Posted by: @jflamm

↑

The blob of checkerboards to the north is where the provided site control lands in the real world...

...when you treat it as State Plane coordinates. It is not State Plane Coordinates!!!!

Yes they scaled around 0,0 and now had a coordinate system they could use on the ground for construction. Pretty standard practice from my experiences. Also pretty standard (unfortunately) that they don't truncate this new system so these coordinates end up in the hands of those who don't know what they are holding. Sometimes it only causes frustration as in your case. Sometimes there are actual costly consequences. Don't make your State Plane coords scaled to ground look like actual State Plane coords!

Also don't give someone control for construction that is realized on a significantly different plane than where the construction is to occur.

😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 

I sleep well every night because I chose to never worry about such things.  I gladly never pursued such jobs in the first place.

On the other hand, I am happy to know that others live and breathe on such projects.

The ground projection is actually scaled around 600,000.0000, 0.

Posted by: @mathteacher

↑

The ground projection is actually scaled around 600,000.0000, 0.

The coordinates aren't normally scaled from the origin point of the zone. That would defeat the purpose of the project scale factor. In fact, it isn't scaled "from" any point. The coordinates are multiplied by the project factor, which also means that it mimics scaling around 0,0. However, change the scaling point to anything near 0,0 and it makes very little difference to the final numbers of the surface coordinates at the project location. A change of 0,0 to 100,100 will only make a change in the coordinates of .01 feet for project coordinates using a 100PPM scale factor. Even a northing or easting of 2,000,000' will only change .01'. Same as a coordinate of 500,000'. 

Significant numbers. 

That's why if you have surface coordinates clearly scaled from some nearby point, don't sweat too much getting the precise location for that point. You only need to get close. The PPM of the scale will allow you to set an error budget of how far off is too far. A PPM of 300 will mean 100' from the real scale point introduces .03' in the locations. However, even that is dependent on the directions from that "real" point and the site location. If the miss is north 100' and the site is east, the resulting error is nil in both northing and easting. That's why I had my PC run the numbers using two scale points. Made almost no difference. 

@lurker 

Truth.  Friends don't let friends scale grid coordinates without truncating.  The lack of meta data in these examples shows exactly why.

Exactly and well stated. What the (0, 0) is really saying is that we just multiplied state plane coordinates by a factor to create ground coordinates. I used the False Easting for the Texas South Central Zone, but any spot would do.

Each Texas county has its own factor, the Surface Adjustment Factor (SAF). San Antonio is in Bexar County which has SAF = 1.000170. Some counties use more significant digits (this one has only 3) but dividing the ground coordinates by the SAF should produce State Plane coordinates.

In the few Texas DOT projects I've looked at, both state plane and surface adjusted coordinates are published and labeled and the arithmetic works. 

It works and it's not too difficult to understand.

Posted by: @john-putnam

↑

@lurker 

Truth.  Friends don't let friends scale grid coordinates without truncating.  The lack of meta data in these examples shows exactly why.

Truncating was SOP back in the day. Normally working with state plane we would remove the million number. A northing of 1,500,000 would become 500,000. The same would happen for lessor numbers, an easting of 850,000 would become 50,000 (you've got 10 miles of play to get to the next integer). This wasn't done for scale factor issues, it was done cause no one wanted to write down the same number over and over. No point in the 1 million feet since it was on every northing. When you're typing or using pen, pencil and paper it's a useful short cut. Everyone understood the reasoning behind those truncations. This happened for real state plane numbers or modified numbers. Usually, we would begin a control traverse starting on a known NGS monument, develop a scale factor for the project, then survey using surface distances to the next closing NGS monument. For really large projects each number would be reduced to the plane way down below the surface and the state plane coordinate would be calculated. Those were normally large mapping projects. 

Then in the late 70's early 80's DOT started to multiply the coordinates. I hated it cause I couldn't plot points on quad sheets, they were always way NE of the real position. But over time computer programs wiped away hand plotting and it's a non-issue today. Put a Quad in Autocad that inserts at SPC, hit SC command, pick the quad, pick 0,0 as the base point, type in the factor and the quad sheet expands to the site properly, same with photos, drawings ect. 

The only metadata needed is the scale factor. However, once the coordinates are old-timed truncated all that becomes more complex, so DOT won't do it. I suppose where things are low, flat, the numbers are near one for a scale factor it gets difficult to notice when you're on a multiplied coordinate system. Where we are it's clear instantly, normally the shift is +600' NE, you can't think it's state plane. Truncating becomes irrelevant and counterproductive. Since we're at elevation it's somewhat rare that anything is done on state plane surfaces, I never expect any coordinate I see to be SPC. 

My only purpose for truncating is to eliminate errors and costly mistakes and what has occurred in this thread. Surface coords truncated never get mistaken for State Plane coords. It is simple and it is safe.

Most importst-metadata. Write down what you did so the next poor guy can follow and not have to guess.

Truncating coordinates disconnect those coordinates geographically. Without metadata they are the same as a set of 10,000, 10,000 with a compass bearing as a basis. 

If the metadata is available why bother truncating? Simply to be sure someone doesn't "think" that they are SP? Not much thinking going on there. Either way both sets have the same issue, no metadata, then no geographic basis. 

With multiplied coordinates all that's needed to know is the SF. And if it's not given it's trivial to reproduce. Much more difficult with a scaled from nearby point set, although not that difficult either. 

If the metadata is available why bother truncating

Exactly and we have a whole thread because it is not available. It tends to be not available and or even wrong over half the time.

We, and most in this area never truncate and always scale to 0,0.  Being doing it that way forever without any issues at all.

@bill93 I can testify that even if, and that is a big if, metadata was originally provided, chances are it will not be present on a set of engineering plans.

Posted by: @mightymoe

↑

Truncating coordinates disconnect those coordinates geographically. Without metadata they are the same as a set of 10,000, 10,000 with a compass bearing as a basis. 

They lost that connection as soon as you scaled them.

Posted by: @lurker

↑

If the metadata is available why bother truncating

Exactly and we have a whole thread because it is not available. It tends to be not available and or even wrong over half the time.

We always did one thing once we were presented with truncated coordinates. That happened often in the days of NAD27 since people didn't want to type or write down the repeating numbers. Once computers became available the given coordinates would be un-truncated. I would do the same with a set today if I ever see one. 

Truncating doesn't really solve anything more than shifting the coordinates to a far away location within SP. If coordinates look like SP but aren't they need to be dealt with somehow, truncated or not. Coordinates for a set of monumented control without metadata should never be assumed to be "on State Plane".

However, there are easy steps to get any set of carefully measured control points attached to our global geographic system with defined projection parameters that can be shared as metadata. 

First locate the control points, do they produce repeatable distances and were the interior angles of the figure created by the points accurately measured. If so then that figure can be placed on the lat, long system without any calibration (even 10,000, 10,000 points). To do it quickly simply set on a given coordinate, use it as an origin point, project through it creating an LDP, add an angular rotation and you have definable projection through the coordinates. To be more advanced figure out the central meridian, scale and origin coordinates. That takes more advanced knowledge but it's very doable. 

If it's modified state plane there are two usual ways they get modified to surface:

The simple way is the multiplied coordinate. Occupy the points and divide. You will get a scale factor easily that way and apply it. If the points were poorly measured then there isn't anything that will fix it, not a scale factor, not a calibration. 

For scaled coordinates around a point there is an easy way to deal with it, and a more difficult way. The more difficult way is what I related in the OP. There I had the given scale factor, if I didn't have that I could have figured it out by inversing ground vs. grid on the points. The results would have been very similar. 

The easy way is when you get the real SP value and find it's off a foot or two from the given coordinate, simply use the given "off" coordinate for the point setting up the base over it, locate other points, get a ground vs grid scale and apply it to the base point holding the foot or two foot off given coordinate. I don't like that way cause it's basically doing a here point with the resulting error, but it is quick and it works. And it's still much better than a calibration. Plus it doesn't take more time than a calibration, it's probably faster. 

I get it, it's frustrating to see something close but not on SP. It's easy to push the button and get it to crank out repeatable numbers that mesh with given on the control positions. SP can kinda do that. But there was a thread I posted from a DOT supervisor who was upset a newby contractors surveyor starting a project telling him all the control was "off" OPUS by 2 to 3 tenths and everything for the job needed to be shifted to match his GPS. LOL. What was shocking was to see so many posters agreeing with the newby. Soon 2022 will come out. Already I'm getting some info from DOT about the transition. They explained that they won't use the new coordinates without modification. Same as always, the max 75PPM is just too much "error" to allow. Since 2022 will be LLHT and no longer tied to the stability of NAD83 things might get really interesting.