I've finally completed my control network that is sufficiently accurate for my purposes; My next step to to develop a highly detailed topo map of the 25 mainly sloped, wooded acres. I've done a first pass TIN, using just about every point I've gathered during the process, but am finding that, using the software I'm using (TPC), I could probably draw a topo map more accurately and more quickly by hand. There're just not enough points to achieve the kind of detail I'm looking for. I don't have a robot, so I've tried to make good use of techniques to capture features (trails, streams, ditches, a pond etc., but it's terribly slow going.
There are only a few pages in Ghilani and Wolf on the subject, and while there are several threads here on fairly specific problem solving, I'm looking for more general information on workflow or process. So far, it's just been: 1. Capture data points, 2. Create a TIN, 3. Add Breaklines, 4. Plot contours, 5. Repeat
I've thought of manually just adding hundreds of interpolated points, with which to make a more detailed TIN, but there's no easy way to do that either with TPC. Editing the surface (by grabbing contours and draging them around) does not seem doable either (If I could do that, I'd be set, because I know what the terrain looks like and know how to replicate it in a topo map). Finally I've thought of turning the entire surface into poly lines, dividing them, and dragging them around, generally trying to do it all in CAD, but that seems kludgy as well.
So, the question is: Are there any good resources out there that describe the proper way to achieve this? Or is it too highly software dependent, such that even if one understood the right process to follow, would not have good results without the right software tools? And finally, Is it possible that no matter what software or technique I end up using, will the results be inadequate if I just haven't captured enough data? Thoughts welcome.
The KEY to any topo is detailed shots from the field crew. If you do not have the shots needed to define the site, your tin is worthless.
". Capture data points, 2. Create a TIN, 3. Add Breaklines, 4. Plot contours, 5. Repeat" WAY too many steps. This is where a good software package comes in. One that is easy to use and gets the job done "automatically" The "automatic" part being driven by your field codes viaa field to finish program.+
"I've thought of manually just adding hundreds of interpolated points, with which to make a more detailed TIN, but there's no easy way to do that either with TPC. Editing the surface (by grabbing contours and draging them around) does not seem doable either (If I could do that, I'd be set, because I know what the terrain looks like and know how to replicate it in a topo map). Finally I've thought of turning the entire surface into poly lines, dividing them, and dragging them around, generally trying to do it all in CAD, but that seems kludgy as well." NEVER, NEVER, NEVER DO THIS!! You are opening yourself to a lot of liability. Interpolated points ail leads to disaster. Can you justify that addition in a court of law if necessary? Of course not.
My training is as a Cartographic Drafter". I LOVE doing topos. The idea is to get software that will do all the initial grunt work for you so that you can focus on tweaking the tin to get a good surface.
It appears to me that your issues arise from the li8mitations of your software. I understand budget issues, but "you get what you pay for". I would strongly suggest that you take a look at the Carlson Survey software with embedded AutoCAD. The Intellicad version is good, but it still has a few qwerks that I am waiting for them to fix. Those may not affect what you do, so the budget may fit it better.
Carlson's phone support center is awesome. Ladd Nelson often visits this site to answer questions. Ladd and Jennifer DeBonna are the King and Queen of Carlson software. They are the best. Then there are local folks, like me, that offer on site training or group training if desired.
You can get a 30 day evaluation copy of the software from your Carlson dealer, or Ladd, or Jennifer. Get one and take a run at it. You won't be a pro on the first try. That's where Jennifer's manuals come in real handy (thatgcadgirl.com). Get the software, a little training and you will be off to the races.
Remember, the secret is enough shots in the field and a standard set of descriptors (field codes). Once you have that all done, you are ready to go. If that kicks your butt, send me your field code list and coordinate text file and I then can build you a Field to Finish file and run up the base drawing for you and then you can take it from there. This would be a one time process, because at the end you will gave the field to finish. FLD file that you need for all future jobs. The cost for this process would depend on how large the files are, ie how many descriptors you use etc.
Carlson is you answer. What you do after that is all up to you.
cptdent:
You give great advice, except you seem to be missing that RFC has been teaching himself surveying as a hobby, with the help of this Board.
Ken
So, that means he does not need survey software that works for him" Wuhodathunk !? Haven't been on this site for awhile, so I guess I did miss those details. My bad.
RFC I maybe missing a point but software and processing doesn't make a good DTM, or surface.
It can but as you've said it doesn't seem right I'd suggest first and foremost is good field practice.
You could blast your site with points (and waste heaps of time) but equally achieve the same results with far fewer points strategically taken to include changes in grade.
On even slopes you only need start and finish of that grade.
Are you surveying in ditches? You need top, toe (2 if it has width at bottom), top. I prefer taking such at same section of ditch. Makes better looking, more correct model.
Often changes of grades are imperceptible and the landscape is a progression of changes or like a rolling slope with no distinct change point. That's when discretion takes place and depends on contour interval you're aiming at.
That's also when you take a few extra points.
If you've surveyed it yourself you'll have a feel for break lines and with care insert your own.
Coding helps there.
Basically its looking at the ground you're standing on, thinking how you can best transpose that to a computer in terms of points and commands to break the surface 'here'.
Then start surveying.
Throw a few extras in for further checks later in areas of doubt, process, then remove those points from surface and see what changes take place in your output.
I love topo surveys and tell clients who wonder at contour interval, it's often little extra work to achieve 250 mm intervals over 1 metre interval (confidence of interpolated points is what counts here, for design purposes)
Have fun and plow on.
What is the end product going to be and who is going to use it?
1. If it is lines on paper that approach is good, then maybe use some 3d polylines to make it look logical and correct the poor TIN.
2. If the deliverable also includes a TIN for design then you need to know the users software and how to force that software to draft your best shot at proper contours and infill spot elevations. (faked spots, breaklines, etc.) For that you need a dark room and lots of coffee... and time.
I am pretty sure I don't miss doing TINs
The last time IO played with TINs was 25+ years ago with a $100 third party topo program in AutoCAD 10, generally it was swapping TIN faces, that is picking the best two triangles between 4 points. Once I started using a CAD package, AutoCOGO, I learned to create a 3D node if I needed to force a contour. Otherwise it was easier to create 3D breaklines, also using horizontal and vertical 3D offsets.
Paul in PA
rfc, post: 438279, member: 8882 wrote: So far, it's just been: 1. Capture data points, 2. Create a TIN, 3. Add Breaklines, 4. Plot contours, 5. Repeat
I don't know how TPC works, but for every software package I've used the sequence would be:
1. Capture data points
2. Add Breaklines,
3. Create TIN/contours
5. RefineWith a good set of breaklines you should be able to generate a decent TIN/contours.
cptdent, post: 438296, member: 527 wrote: So, that means he does not need survey software that works for him" Wuhodathunk !? Haven't been on this site for awhile, so I guess I did miss those details. My bad.
Not at all. Great advice is great advice, no matter the capabilities of the recipient. Thank you for all of yours.
I've not yet tackled FTF, and know I need to. That would speed the process significantly. But I fear that the elephant in the room getting in the way of "the secret of getting enough shots in the field", is the lack of a robot. I can really see how a solo operator, armed with a robot and a good FLD file, could cover a lot of ground in short order.
I'll "recalibrate" my sights on a smaller area to start with, perfecting the process first.
Finally, on the subject of Carlson, the last time I did the 30 day demo, it was *way* to pre-mature, and ended up being a waste of time. I might have to take another look.
Thank you again, Sir.
RFC I surveyed a site today and as it was such a lovely day and not a big job I did an overkill to demonstrate the way we can saturate points to little if any benefit.
Site was even slope and client asked for 1 metre contours.
I modelled my complete survey then deleted some points from the DTM and remodelled.
The yellow highlight shows the deleted points.
You can see in the upper parts where slope is very even it has little to no effect.
Lower down it does play with them which is understandable as its flattening out and I (deliberately) took out some where the grade starts to change.
As said this one is so even a slope to make it somewhat simple, but don't be fooled by 'even looking slopes'. They can be in reality a large rolling slope rather than an even grade slope.
Well as stated you need lots of points the more the better, but also realize that it only takes one bad point to really mess up your TIN. You have to have a feel for your project and look for bad shapes and slopes, then edit out any points that may be incorrect, and re process.
rfc, post: 438347, member: 8882 wrote: Not at all. Great advice is great advice, no matter the capabilities of the recipient. Thank you for all of yours.
.....
Finally, on the subject of Carlson, the last time I did the 30 day demo, it was *way* to pre-mature, and ended up being a waste of time. I might have to take another look.
Carlson is very intuitive to a knowledgeable and qualified surveyor.
Maybe you should reconsider what is a waste of your time.
Paul in PA
Paul in PA, post: 438377, member: 236 wrote: Carlson is very intuitive to a knowledgeable and qualified surveyor.
Maybe you should reconsider what is a waste of your time.
Paul in PA
What I meant, Paul, was that *at that time* it was not as productive as it might be now, or even sometime in the future. As intuitive as it is, two years ago, I was jumping in the deep end on may aspects of the art, and attempting to become productive with Carlson on top of everything else was less than useful. I chose, instead, to focus on SurvCE and Starnet.
Richard, post: 438352, member: 833 wrote: RFC I surveyed a site today and as it was such a lovely day and not a big job I did an overkill to demonstrate the way we can saturate points to little if any benefit.
Site was even slope and client asked for 1 metre contours.I modelled my complete survey then deleted some points from the DTM and remodelled.
The yellow highlight shows the deleted points.You can see in the upper parts where slope is very even it has little to no effect.
Lower down it does play with them which is understandable as its flattening out and I (deliberately) took out some where the grade starts to change.As said this one is so even a slope to make it somewhat simple, but don't be fooled by 'even looking slopes'. They can be in reality a large rolling slope rather than an even grade slope.
Nice. What is the structure in the lower left hand corner of the plat? Did you code those breakpoints/lines in the field?
Its all about the field work. I collect everything in 3D, starting with features and breaklines, then I fill in with points where needed. I have developed my technique from years of data collection and surface model building. I feel one is so dependent on the other I require anyone that works for me to learn to do both.
Sent from my iPhone using Tapatalk
[USER=8882]@rfc[/USER] yes I coded the breaks in the field, but did add some extras by hand later, (in office).
BT, BB, bank top, bank bottom for egs, get linked as breaks.
Coding, as others indicate, join lines in DC, and if they're 3D then easy to select as breaks.
The strucure(s) bottom left is an open sided shed with the walls shown and 'verandah' (uncovered wall sections) shown individually.
The shed sits on a bench cut into the hill, and next to it is a very odd small excavation like a loading ramp with 3 sides.
Other odd looking 3 sided affair to north was a wrongly entered small shed which I tidied up later by hand.
As client asked for 1 metre contours I took this as an exercise to evaluate my field times given my statement on not (mostly) much longer to achieve greater accuracy.
When it comes to drains and other similar features, these need locating horizontally anyway so little extra in a few vertical points added.
This job took about 4 hours, including looking for 2 'lost' boundary pegs, and tidying them up with new paint and markers.
I could only save about 1 hour max, if that, to get only 1 metre contours. Given the features that needed to be included regardless of need for contours.
I've shown 0.5 interval, but survey could easily support 250mm contours.
With regards to feature only surveys without contours. I generally included sufficient to account for them later if required. I only charge for features, contours as required.
Been surveying too long to know someone, sometime will say 'oh. Now we need the levels'. Saves me effort and clients expense.
Don't often get lumbered with worthless survey for contours when not required initially.
Often its the next owner that asks and I can oblige and quickly.
That's all bit off topic stuff.
https://babel.hathitrust.org/cgi/pt?id=wu.89071917272;view=2up;seq=10;skin=mobile
Begin with the basics
DDSM
when i survey all i see is triangles laying on the ground ...... thousands of them
triangles on a busy toe going to more shots than required on a not so busy shoulder a weeding process less busy break lines
i usually leave the triangles on a no plot but are visible in cad but not pdf
Contours are a result of triangles.
Just imagine laying various shaped triangles of plywood on the ground to approximate the round earth
please review attached pdf
Peter Kozub, post: 438498, member: 375 wrote: when i survey all i see is triangles laying on the ground ...... thousands of them
triangles on a busy toe going to more shots than required on a not so busy shoulder a weeding process less busy break lines
i usually leave the triangles on a no plot but are visible in cad but not pdfContours are a result of triangles.
Just imagine laying various shaped triangles of plywood on the ground to approximate the round earth
please review attached pdf
You and me both brother, sometimes I feel like I'm in the matrix!!
Ron Lang, post: 438500, member: 6445 wrote: You and me both brother, sometimes I feel like I'm in the matrix!!
Please review title block on PDF
