I have a job where we have to monitor a railroad bridge and rail where we are replacing half the bridge and half the abutments.?ÿ A 15" rain a couple years ago undermined the south side of both abutments on an old girder bridge with 2 tracks.?ÿ They want to remove the south half of the bridge and the concrete abutments on each end while maintaining train traffic on the north track.?ÿ We have driven sheet piling between the tracks at each end of the bridge and there will be anchors thru that piling under the remaining track drilled up to 85' long at a 60 degree angle to hold the existing track in place.?ÿ Here's a pic.
Need to monitor all 4 corners of the bridge, the rail over the abutments, and the sheet piling for movement.?ÿ They say the allowable movement is 1/8".?ÿ However unrealistic that is, we have to monitor it.?ÿ Due to the equipment that will be in the way and the inaccessibility on each side of the track, I believe I will have to set control points on an adjacent road bridge about 300' upstream, a crossing about 600' west of the bridge, and near an overpass about 300' east of the bridge.?ÿ Will have to set up near the west end of the bridge and resection to those 3 points.?ÿ
?ÿ
?ÿ
Will be using a 1 second Trimble SPS 930 robot.?ÿ Questions are:
1. At those distances, will a nodal point prism permanently located at each control point increase the accuracy of the setup, assuming setup locations may have to vary some depending on what's in the way??ÿ Is the accuracy of a robot increased by locking onto a nodal point prism vs a standard prism?
2. Has anybody used these type prisms;
?ÿ?ÿ?ÿ https://www.berntsen.com/Surveying/Survey-Prisms/Mini-Prisms/ctl/ViewProduct/mid/1600/itemID/1550
?ÿ?ÿ?ÿ with a robot for monitoring rail, and if so, were you satisfied with the results?
?ÿ
Trying to come up with the most efficient a quickest way to get all the shots needed. Having to walk to each one with a rod and bipod would be pretty time consuming.?ÿ If I can set up the robot once, and shoot everything robotically, it won't be too bad.
?ÿ
?ÿ
?ÿ
RE #2
I've used those with good results except...halfway through the monitoring campaign the railroad came through with their ballast grooming equipment and knocked every one of the targets off the rails?ÿ
For campaign monitoring you really don't need to worry about nodal prims, or offsets for that matter.?ÿ You are looking for change.?ÿ As long as each prims is shot from the same control.?ÿ When setting up a long term real-time monitoring system you can shot through single pain glass in an instrument enclosure.?ÿ Even though the true position is skewed,?ÿ the vector you are measuring remains constant unless you have movement.?ÿ As for the prisms, I have never had much success getting them to bond to the rail web.?ÿ They tend to fall off which really messes with your results.?ÿ You can get brackets that clamp onto the base of the rail that work great.?ÿ That is until the tamper comes through.
Good luck
My concern is I won't be able to set up in the same location every time, so I could be shooting the control from slightly different angles.?ÿ Forgot to mention, the 1/8" allowable movement is both H and V.
We just got done monitoring a bridge abutment while the slope protection was removed and a soil nail wall was put in place.?ÿ We use the same instrument but used adhesive backed reflective targets.?ÿ We had 18 points to monitor 3 times a day, 7 days a week, for almost 4 months.?ÿ Worked out extremely well.
We just got done monitoring a bridge abutment while the slope protection was removed and a soil nail wall was put in place.?ÿ We use the same instrument but used adhesive backed reflective targets.?ÿ We had 18 points to monitor 3 times a day, 7 days a week, for almost 4 months.?ÿ Worked out extremely well.
Were you shooting the reflective targets robotically?
You need to get the kind of equipment these guys had...
They said the bridge has moved less than 1/1000th of an inch........
"The device uses mirrors, a laser and a series of sensors to test every half-hour 24 hours a day for any movement in the bridge.
The good news is the bridge has moved less than 1/??1000th of an inch since the fire."
?ÿ
More like smoke and mirrors to get that accuracy.
No, manually sighting direct and reverse.
I can't tell you how many monitoring specs state 0.001ft.?ÿ Good luck with that in at lab setting let alone on a bridge.?ÿ I think it is a case of national cut and paste from one set of specs to another.
I will give them the benefit of the doubt and assume they were misquoted in the article. But even if they meant 1/1000 of a foot, still not achievable.?ÿ
Interestingly, I found out today that I will be working with the company that did the bridge monitoring on a state wide photo control project. My client was awarded the contract, and to meet their WBE requirement they have three minority or woman owned firms and my company (not wbe) on the team. We will each do 1/4 of the state, and then I will review and adjust all of the data and prepare the deliverables.?ÿ?ÿ
Resections for high accuracy monitoring. Doesn't sound like a good idea. You will need very stable points, yes definitely nodal prism on each. This may be difficult in an unstable construction site. Better off setting permanent position for the instrument location.?ÿ But sounds like you don't have that luxury.?ÿ
How many points total?
I would possibly try static GPS. Building a bracket or weld 5/8" bolt for each point. If you have say 8 points to monitor, then use 9 static receivors, for a single simultaneous occupation or session, I'd say an hour at 1 second epoch. Along with one GPS on a very stable position.
Just my $0.02
?ÿ
One point to bear in mind. You talk about setting up your instrument in locations such as an adjacent road bridge, etc. Make very sure that your setting position isn't within any movement joints on the structure, or that could throw all your observations. I once saw an engineer use a "reference" target he had fixed to the end of a 50ft. high steel frame building - at the expansion end!
?ÿ
1. At those distances, will a nodal point prism permanently located at each control point increase the accuracy of the setup, assuming setup locations may have to vary some depending on what's in the way??ÿ Is the accuracy of a robot increased by locking onto a nodal point prism vs a standard prism?
"nodal" is a term some use to describe the relationship of a prism's corner, as seen by an observer, relative to the intended target. If you don't have an intended target other than the prism corner, then a prism by itself is neither "nodal", nor non-"nodal".
if you rigidly fix a prism to something (I'm not sure if that's what you intend to do) then you need to apply some correction to your measured distance to these fixed prisms in order that the corrected distance coincides with the viewed reflector corner. otherwise your computed reflector coordinates will change according to viewing angle.
I will give them the benefit of the doubt and assume they were misquoted in the article. But even if they meant 1/1000 of a foot, still not achievable.?ÿ
As it was described, from "a few feet away", let's say 6 feet, 0.001 inch perpendicular movement is 2" of arc if my calculations are correct. With care, this movement relative to another close reference point is in the realm of being detectable.
you should set up?ÿ your robot one day and devise a short range experiment to see just how sensitive it is to detecting minute movements at close range. you may be surprised, or not.
Resections for high accuracy monitoring. Doesn't sound like a good idea. You will need very stable points, yes definitely nodal prism on each. This may be difficult in an unstable construction site. Better off setting permanent position for the instrument location. But sounds like you don't have that luxury.
How many points total?
?ÿ
There are 16 points to monitor. The only place I can see them from is on the railroad bed adjacent to the live track on the 25' fill. No place to set a permanent setup position. ?ÿThe surrounding area is succeptible to flooding, so I can't put anything down there ?ÿ
My thinking is that the prisms permanently mounted on the control points should be nodal so that even if the instrument is set at various locations, ( by that I mean I think they would all be within a 50' radius depending on what is in the way at the time) the accuracy of the resections would be as repeatable as I can get them. ?ÿMy question is will a nodal prism increase the accuracy at 200'-600' (distance from setup to control points) shooting rounds to them robotically? ?ÿMy intent is to have the instrument automatically shoot rounds to all the points once the resection is done.?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
I would used fixed prisms for control as well so you do not have to set up tripods every visit. If properly executed resection from fixed points outside the area of subsidence is almost always the best approach when working in a dynamic environment.
A variety of targets for this kind of work are used underground. Ordering in the states can be a bit of a chore.
?ÿ
?ÿ
It is great you are looking for advice on how to do this task.?ÿ It seems to me if it is important enough to call 1/8" spec on movement it is important enough to ask for the surveyors input on how to best achieve that objective.?ÿ You should not have to work around others they should have to work around your control if that spec is truly needed.?ÿ I say that because we as surveyors always take the back seat and don't stand up for what we need to do a job correctly.?ÿ I use to have the same problem with high rise building control.?ÿ If the contractor is going to require my expertise in providing building control for upper floors or bridge monitoring then he is going to have to work with us with project planning and you let them know what it is going to take to achieve the goal.?ÿ My 2 cents, Jp
A combination of a closed control network and reflective targets?ÿ worked for me on a building monitoring project.
I've about given up on trying to predict all the things that can destroy my control points. Reflective targets are so cheap that I just set them everywhere I can and hit them whenever possible.?ÿ
?ÿYou can mount them on metal plates and fasten them just about anywhere. If your project calls for 16 to be placed on the structure, I'd set at least 32 and possibly more if I figured out an efficient way of mounting them.
?ÿ
?ÿ
?ÿ
?ÿA combination of a closed control network and reflective targets?ÿ worked for me on a building monitoring project.
I've about given up on trying to predict all the things that can destroy my control points. Reflective targets are so cheap that I just set them everywhere I can and hit them whenever possible.?ÿ
?ÿYou can mount them on metal plates and fasten them just about anywhere. If your project calls for 16 to be placed on the structure, I'd set at least 32 and possibly more if I figured out an efficient way of mounting them.
?ÿ