No control data - TS1200/15

> Hi,
>
> I am about to start 6 weeks of track monitoring for UTX works across South east England railways. I am an assistant with a good understanding of setting out procedures using a Leica total station and general set up with known points. However, the sites I'll be assisting on have no control data whatsoever. I have been told we will set up using dummy co-ordinates i.e x=2000 y=1000 z=100 and I have researched the concepts of local / arbitrary grids but I just cant find exactly what is meant and how I achieve this. I want a better knowledge so I can assist in the best way possible.
>
> The gear to hand is a Leica TS1200 or TS15, 2 x tribrachs with legs and targets. The set up has to be as accurate as possible due to nature of work and we will be returning to each site numerous times for follow up surveys / monitoring.
>
> If anyone can talk me through the set up process It would be very much appreciated.
>
> Kind regards.

is the track monitoring small areas?

is each monitoring area far away from the next?

have you contacted the rail company's survey/engineering/land acquisition office for their control networks?

it appears you are the assistant. have you been tasked with learning which reference frame to use? regrettably, i have looked into similar cases in the past and have been told i am doing 'too much'. in other words, depends on management's attitude toward your effort (tread carefully)

six weeks of monitoring seems, to me, to be worth doing on a matching reference frame as previous surveys. if not, then some substantial portions of usable data is, well, unusable

> > Hi,
> >
> > I am about to start 6 weeks of track monitoring for UTX works across South east England railways. I am an assistant with a good understanding of setting out procedures using a Leica total station and general set up with known points. However, the sites I'll be assisting on have no control data whatsoever. I have been told we will set up using dummy co-ordinates i.e x=2000 y=1000 z=100 and I have researched the concepts of local / arbitrary grids but I just cant find exactly what is meant and how I achieve this. I want a better knowledge so I can assist in the best way possible.
> >
> > The gear to hand is a Leica TS1200 or TS15, 2 x tribrachs with legs and targets. The set up has to be as accurate as possible due to nature of work and we will be returning to each site numerous times for follow up surveys / monitoring.
> >
> > If anyone can talk me through the set up process It would be very much appreciated.
> >
> > Kind regards.
>
>
>
>
> is the track monitoring small areas?
>
> is each monitoring area far away from the next?
>
> have you contacted the rail company's survey/engineering/land acquisition office for their control networks?
>
> it appears you are the assistant. have you been tasked with learning which reference frame to use? regrettably, i have looked into similar cases in the past and have been told i am doing 'too much'. in other words, depends on management's attitude toward your effort (tread carefully)
>
> six weeks of monitoring seems, to me, to be worth doing on a matching reference frame as previous surveys. if not, then some substantial portions of usable data is, well, unusable

Thank you for your reply Moe Shetty,

So we're setting up track monitoring targets 30m either side of the UTX in 5m increments, shooting a baseline survey then monitoring them post UTX 3 times a day for 1 week and reducing that over a 3 week period for each site. The sites for UTX are varied and are in all sorts of places miles apart. I have simply been told there is no control data to go from and we'll create a local baseline grid for each UTX. The people I have spoken to say it is easy yet have not explained it and as I said I can't seem to find what I'm looking for.

As far as I know I create dummy points over pins and use them for each set up. I'm still confused though.

Info links or instruction would be great.

Regards

I think the term you want to spend some time searching is "resection" (the surveying kind). Here are a couple links to get you started:

http://www.leica-geosystems.com/flipbook/surveying_made_easy/en/index.html#/22/

http://www.raeng.org.uk/publications/other/10-resection

And here is one that is NO help at all, but it has a nice beat and shows some warm climate:

Enjoy.

> So we're setting up track monitoring targets 30m either side of the UTX in 5m increments, shooting a baseline survey then monitoring them post UTX 3 times a day for 1 week and reducing that over a 3 week period for each site.

I had to look up UTX in a British context to see that it refers to under track crossing.

This sounds like a pretty standard situation. Install a number of stable control points well outside the area of work -- I suggest a minimum of two beyond each end of the work area, including some on each side of the track and far enough away from the track so as not to be disturbed by railway operations. Assign an arbitrary coordinate to one of them, and an arbitrary direction from that point to one of the other points, and then tie them all together using that one fixed coordinate and the one fixed direction. Add in your monitoring points, and you're done. Remeasure all control and the monitoring points during each monitoring event, and compare the resulting values to those of the first event to determine the magnitude and direction of movement.

I believe I can only use a resection If I know the coordinates of the control points (which I unfortunately do not)

Thank you for posting though.

> > So we're setting up track monitoring targets 30m either side of the UTX in 5m increments, shooting a baseline survey then monitoring them post UTX 3 times a day for 1 week and reducing that over a 3 week period for each site.
>
> I had to look up UTX in a British context to see that it refers to under track crossing.
>
> This sounds like a pretty standard situation. Install a number of stable control points well outside the area of work -- I suggest a minimum of two beyond each end of the work area, including some on each side of the track and far enough away from the track so as not to be disturbed by railway operations. Assign an arbitrary coordinate to one of them, and an arbitrary direction from that point to one of the other points, and then tie them all together using that one fixed coordinate and the one fixed direction. Add in your monitoring points, and you're done. Remeasure all control and the monitoring points during each monitoring event, and compare the resulting values to those of the first event to determine the magnitude and direction of movement.

Thank you very much for the info, its making sense now. Is there any literature of web video you could re-direct me to?

Thanks again for sharing your knowledge to the British.

Matochen what I am getting from your post is all anyone is concerned about is the day to day changes in the track so the ability consistently tie into the initial reading system is all that is important. If this is the case I would suggest setting the total station in a spot where all data can be acquired in one setup (if possible) then set retro-reflective targets or fixed prism mounts at spots that give you strong resection geometry and forgo setting up on hard control. In general with good geometry you can resect much more quickly while maintaining the high level of accuracy needed for monitoring than you can set up over hard control.

http://www.berntsen.com/Surveying/Smart-Targets-Datums-Reflectors
I remember seeing an article where the rsak80 targets were affixed to the side of a track and used as the reference points for monitoring. The range values listed on the website are less than half of what most reflectorless total stations will shoot.

You are getting great input. This really is a great place.

It appears that there is no need to tie different crossings together so an assumed set of coordinates for each location is perfectly fine. No need to complicate things. Simply take the same set of readings over and over to discover movements that would indicate problems. It would be ideal to be able to set up in a single place and hit all targets but that is probably not going to happen. Try to go with the minimum number of set ups while maintaining proper angular relationships. That is, try to avoid shots where a minor angular difference would produce a significant distance difference.

> Matochen what I am getting from your post is all anyone is concerned about is the day to day changes in the track so the ability consistently tie into the initial reading system is all that is important. If this is the case I would suggest setting the total station in a spot where all data can be acquired in one setup (if possible) then set retro-reflective targets or fixed prism mounts at spots that give you strong resection geometry and forgo setting up on hard control. In general with good geometry you can resect much more quickly while maintaining the high level of accuracy needed for monitoring than you can set up over hard control.
>
>> http://www.berntsen.com/Surveying/Smart-Targets-Datums-Reflectors
> I remember seeing an article where the rsak80 targets were affixed to the side of a track and used as the reference points for monitoring. The range values listed on the website are less than half of what most reflectorless total stations will shoot.

Great stuff, thank you to all who have responded.

I hear what you are saying, a single set up where all data can be acquired would be optimum, that will be the aim. I have done a lot of monitoring before but always given designs with control data.

As for the set up, I'm still struggling with the concept of no control. Would I need to orientate the TS over a nail or point with dummy coordinates ie. x2000 y1000 z100 and then place retros or ground stakes with targets and shoot them as a backsight or resection or in another set up mode? Should I create the 3 control points with the same dummy coordinates?

I'm sorry but I'm still confused.

Set 2 points in the ground. Occupy one of them with the instrument. Key in your assumed coordinates and elevation for the occupy point. Measure the distance to the other point. Take that distance, assume a cardinal direction that best matches the orientation of the points in the field and apply the distance to the assumed coordinates of your occupy point and then key it in.

Example:

point 1
northing: 5000.00
easting: 5000.00

distance between points 250.00

assuming point 1 is south of point 2

point 2
northing: 5250.00
easting: 5000.00

Now everything will be relative to that baseline. If you need to use an assumed elevation on point 1, just measure your hi and hr and recollect point 2 using them. This will give you your relative elevation on point 2.

Hope that came out right.

One major consideration for deformation / monitoring-
Rather than assume a direction to your aiming point (second control point), orient the baseline to best present the probable distortions. In the case of a dam you would want the baseline to illustrate up-stream to downstream movement. With rails it should run parallel or perpendicular to the direction of travel. It makes the results easier to understand and increases the value of scatter plots.

To simplify the coordinates, I would probably set a PK nail in the center of the track on a tie at the beginning of each section. I would assign that 0,0. I would set another one at the other end of that section, measure the distance between the 2, say it's 1000 ft. I'd assign that one coordinates 1000,0. Then I would set control points either side as needed, negative coordinates would be on the left and positive on the right. Or you can use it as stationing and offset, with start station of 0 and end station of 10+00. Makes it easier to keep up with. If you are in a curve it's not as simple, but works the same way. Don't even have to set up on the nails initially if you have software to rotate your coordinates, just set up somewhere where you can see everything and shoot it.

Matochen,

What I typically do when starting a survey with no fixed control, such as what you are doing is create a point at 5000, 5000, 100 (pick whatever numbers you want it really doesn't matter) The next step is to turn the gun north and set my azimuth to 0. I then shoot all of my control and whatnot. After your initial set up you simply come back and use the same control points.

All that has been said is good advise. The only thing I would add is to keep your point numbers and coordinates unique to each monitoring location. That way, if you do need to plot things up in cad, the points won't all be on top of each other. For instance, location #1 would have control points in the 100's, monitoring shots in the 1000's and starting coordinates in N1000 & E1000, second location control in the 200's, shots in the 2000's, coords N2000 & E2000, etc. Sometimes a little organization good a long way.

The control points you set are all you have to work with since you will be on your own made-up coordinate system. I would set two control points close to the work and at least two more farther out and known only to me. If your control is lost, the monitoring is broken.

I live with an English teacher. She would pitch a hissy if she heard this: The control points you set are all you have to work with since you will be on your own made-up coordinate system. I would set two control points close to the work and at least two MORE FARTHER out and known only to me. If your control is lost, the monitoring is broken.

Deep down I know you meant that two more should be set farther out. She would just hear MORE and FARTHER back to back and cringe.

HMMMM!

What I meant "more" to modify was the number of control points not the distance "farther".

I guess I should/could have worded it more better.;-)