“..the true way of Art is not by Instruments but by Demonstration: and that it is a preposterous course of vulgar teachers, to begin with the instruments, and not with the Sciences, and so instead of Artists to make their schollers only doers of tricks, and as it were Juglers: to the despite of Art, loss of precious time, and betraying of willing and industrious wits with ignorance and idleness.”
Wm. Oughtred 1630
A lot of searches for ‘traverse’ and ‘how to traverse’ find this blog. So far I have had a bit of a rant about what a pain traversing is and how TheoLt helps me get it right. The handful of good textbooks on this and an excellent online guide by the US DoD education dept. take you though the classic ‘measure 1st and compute after’ method and expect you to compute long hand.
Been there, done that (even got one published) but, to be honest, it’s a tiresome business and I now trust the whole maths game to software. I can now build a network of stations for building survey swiftly and produce all the required reports as I go along- this method has the huge advantage of letting me see the blunders as I make them and take appropriate action to keep things ‘on the level’.
So now I’ll have a go at explaining how I get my traverses to work and why I do them. I’ll try and answer some introductory questions and then move on to a step by step walk through of the TheoLt Pro procedures.
What is a traverse and who needs one anyway? A traverse is a method of getting precise control points (stations) measured by TST ( Total Station Theodolite) such that they have a precisely known and distributed error. The idea is to establish a set of points with the highest precision reasonably achievable to measure the detail from. An old survey adage is ‘always work from the whole to the part’. The traverse provides control of the whole so that measurement of the parts is possible.
What kit do I need? Traversing is a 3 tripod game, there is no point pretending otherwise: any saving you might make on using a single ‘pod and a hand held prism is lost in the time taken to re-do the work and the precision achievable will be compromised by the number of repeated set-ups you will need to get the required backsight/foresight shots- each set-up is a source of error!
You will need: TST, 3 Tripods, 3 tribrachs, 2 prisms, 2 prism holders, 3m tape, a selection of station marks, tablet PC with AutoCAD/ Bricscad and TheoLt Pro.
GDF 22: a prince amongst tribrachs, yes you need 3!
You will need to be able to set up the TST and targets over the station marks- if you can’t do this swiftly and easily you are in the wrong game.
How do I start? Walk the site and consider where your stations are going to be, I have found there are 2 things above all others to remember: how am I going to mark the station? and can I see back from this point to the last and forward to the next?
Setting up over ground that will take a mark can save a huge amount of time; a simple scratch mark, a Sharpie ‘x’, a road nail in a paving joint (never try and drive one into a pave unless you want to be charged for a broken one) all work, provided you have something to mark on. If I’m forced to place a station in a grassy lawn I’ll see if I can find a ‘hard detail’ feature to mark and work from- a peg is my least favorite mark: they MOVE!
You will need a circuit of points so a walk around the whole job is needed, spare yourself the surprise of discovering an un-crossable road or blind alley once you have begun, obvious places to work from might also be obvious places to get walked into: setting up at gates and doorways is not always the best thing to do. Trees are a pain: if you are going to return to a site in a different season you can expect your sight lines to be blocked by veg.
Once you are happy you can get around the site in the minimum number of stations and mark them safely, pick one (I like to start at the back of the property – it’s up to you) set up over your mark, set up your backsight and foresight targets. Note the heights of the 3 set ups. The disposition of instrument and targets should look like this:New survey, new project: Boot your PC, open a new drawing (metric), open TheoLt, go directly to the Project tab and start a new project:Select New....fill in the details and we are ready to build the network.
At the 1st Station. Select the control tab in TheoLt and hit the orientation button. Select the ‘Default Orientation’ tab, enter your instrument height and hit the ‘Align with Plane’ button:
Shoot 2 shots at a long wall of your building first on your left then on your right using the ‘All’ buttons. This will line up your job with CAD so that the front /back, left/right views will line up, this is a real time saver when you get into detail survey later on. With the 2 shots in hit ‘ok’, check over the panel and hit ‘orientate’…
…you will be prompted to set the Control type. As this is the 1st station on the loop and will be the point of misclosure it should be set as ‘Fixed’. Close the orientation panel and you will see TheoLt has plonked your first station into CAD with a default co-ordinate value (500,500,100 in this case).
So far so good, at this point I like to trace in some long lines off the building to make me feel better: the plan should plot in a logical alignment.
Progressing the loop Select the control tab, switch the TST to IR mode (to measure to prisms). Aim at the 1st target (to keep the numbering simple I take the foresight 1st) and take the shot, enter the target height check the Stn number and select the ‘Station’ option..
..you can change the Station name here if you need to. Set the station status to unfixed…
..and it’s plotted into CAD. Turn the TST to the backsight and repeat the procedure for the backsight (you may want to set the Station number away from the default of 3 here)
The first round. With the 3 stations in the drawing hit the orientation button and select the ‘Network Adjustment’ tab, it’s time for the first round of observations! Start on face 1 and the ‘flop the gun’ and take a set of face 2 obs: As each shot is recorded TheoLt will detect the station ID and prompt for the target height. The table records the observation data and the condition of the shots relative to their set-out values: green = good, brown = suspect, red=dud. The default tolerance for ‘Red data’ is set at 20mm (if you like the jangle of spurs you can set it to a wider value in Settings, Instrument tab: Max Error). Hit ‘view’ and the measured rays are plotted into CAD:Advance to foresight. With a round complete swap the TST for the foresight prism and move the backsight target to the next forward station:Orient to backsight. Select ‘Orientation to Reference’ tab:
The 3 stations will be selectable from the ‘Stn list’ buttons for the Instrument and Reference positions (AKA ‘station of occupation’ and ‘back sight’) When the station is selected you will be prompted for its height: check this carefully- TheoLt will carry forward the target height set in the Instrument data panel- it could be Zero!
With the stations selected take the orientation shot and check the result:
Hit ‘Orientate’ and switch to the Network Adjustment tab, shoot in the foresight, take the required rounds, when complete advance to foresight, switch to ‘Orientate to Reference’ shoot the back sight:Set out the foresight. Any shot taken whilst in Network Adjustment tab will be treated as a control observation. If TheoLt can’t find a station for a shot it will prompt you to create a new station. Once you have occupied a station and achieved an orientation to the backsight it’s time to set up the foresight and shoot it in.
Select the CAD symbol you want (circle or triangle do the same thing but it can be useful to differentiate between occupied and unoccupied points at times) enter the height of the target and station number…
..declare the station ‘Unfixed’ and you are ready to take another round of observations ..and round you go.
Trap dud data. Sooner or later something’s going to go wrong; usually in the height dept:
TheoLt will prompt you but inevitably crap happens and dud data gets into the Network table:
A double click inthe tick column will take the ob out of the calculation:
You can check the Network at any time by running the ‘Calc’ command. Note that for the ‘Calc ‘ to work 2 ‘fixed’ stations are required. These can be set from the table (by a double click next in the ‘I/E’ column on the station title row) or by setting the station status on creation. Entire station sets can be dropped from the computation, if desired, by editing the station list in the Control Stations tab:
TheoLt Pro lets you try out combinations of fixed stations, including observations or excluding them. The Network Adjustment table becomes a kind of ‘rear view’ mirror once the traverse is underway. As you build up the network data a quick check on the table keeps the job on the right lines. It’s reassuring to find the stations are auto detected as you shoot the rounds-and another source of error, the miss-matched station ID, is eliminated!
Building up the plan as you go around is quite possible and I have found this to be an effective method. It’s up to you; purists like to do the control 1st and then work off the adjusted points. I have been using TheoLtPro for a while and I trust it to apply the right adjustment to the data and keep a safe copy of the unadjusted work. I don’t expect to get the whole plan done from the network stations but I make the most of each occupation as I go around.
The closing angle is treated like any other observation in the table: this is observation which reveals the precision of the loop:
The misclosure. In this case the closing shot from STA 103 back to STA 101 is the last angle on the loop. The height misclosure is 14mm and the XY (on the backsight to STA 606) is 56mm. I’m still fixated on the misclosure value but TheoLt Pro uses a least squares distribution and although the closing angle is a good indicator of the precision achieved the error distribution will be based on a weighted station by station analysis not a simple proportionate distance distribution (Bowditch). The fixed stations can be changed and the calc can be run with or without selected stations ( i.e with or without spurs ). ‘Brown data’ can be dropped form the calc if you wish. With the loop complete run the ‘ Calc’ again to prepare the adjusted station data report and diagram.
The report. Once the Calc is run the adjusted station co-ordinates are available as a report. This this is created by the ‘Report’ button:you will need to type in the .xls bit and slect ‘All Files(*.*) in the file name fileds, hit ‘save’ and lo:
…all the numbers are there for the taking,personally I prefer the *.txt version as it’s easier to cut and paste into my report diagram.
The diagram: The unadjusted diagram is easy: just hit ‘View’ and the observed rays diagram is sent to CAD:
Once the ‘Calc’ is completed…
…the ‘Apply’ button is activated , it gives you 3 options:
1. Crate new Project: You may wish to take a conservative approach to the measured data and create a new project which will leave the current drawing uncorrected and create a copy of the current drawing file in a new project folder with all stations fixed in their adjusted positions.
Select the option and you will be returned to the Create new project panel:complete the panel and a new folder is created in the project base path path which contains the adjusted station list ready for import:
The stations will be ‘Fixed’ when the new project is opened in TheoLt.
That’s it. It’s the equivalent of a ‘save adjusted stations as’ operation.
2. Adjust Station positions: select the project folder and then the drawing for adjustment and 2 copies will be made one pre- and one post- adjustment version. Only the stations will be moved in the post adjust drawing.
If you open the Post Adjustment drawing you will see ( if you zoom hard enough) the difference between the measured rays and the adjusted station positions:and the unadjusted dwg is a safe back up if re-measurement is needed.
3. Adjust Station positions and detail : this is the clever bit. If you are running TheoLt Pro you will have TheoDisto.arx/brx running in CAD as a background service which monitors all measurement data by station as it’s recorded in CAD.When you select ‘Apply’ option 3 any radial observations you have taken are adjusted along with the station data. A duplicate drawing is saved with the adjusted data. You can run this at any time during or after the survey and build up a set of sequenced drawings if confidence in the network is low or if the control needs to be split up in a survey team.
Working with the adjutsed data. The TheoLt Building Survey (TBS) the command ‘theostationmgr’ gives you a list of all the stations used in the drawing and shows the data dependent in them. Open the ‘Plan builder’ and select the ‘stations’ tab: the table is there.
Select the station in the list and the data measured from it is high-lit in the drawing. Now you can troubleshoot by station and weed the data for poor shots. You can attach and detach lines from the station using the Add/Rem commands.
The dashed lines are the station dependencies for the selected station, useful when you are playing the ‘what the blazes is that?’ game in CAD. At least you can see where the lines came from!
Loop, line and schedule. A professional job needs proof of precision: TheoLt Pro gets you there!
TheoOffice: A post-process method. I have described how I use CAD on a Tablet PC for traversing a network with a reflectorless TST for a building plan. This is fine for building survey work but not so good for topo work where keeping a PC happy can be a bit awkward. If you log the data on the instrument memory as a GSI file the same adjustment process method can be used with the GSI data in TheoOffice.
TheoLt Pro Network adjustment works in AutoCAD, AutoCAD LT and Bricscad.