@amksyr?ÿ
"Open" or "route" traverses are what we call a WAG or some derivation of such. IF you didn't have any need to be in a certain place at the end, it may not matter.?ÿ
You can get some idea of how close your WAG might be by running the traverse backwards, but geometry is not your friend on this sort of traverse. If you need it TIGHT, you have to use other options. Typically, we would run it forward and reverse, carrying vertical, and combine with a GPS observation on each end. If you can turn angles to a natural target from stations along the way, all the better. At that point Microsurvey's Star*Net is a great tool to analyze the network.
@rover83?ÿ
This is interesting - but when you switch backsight and foresight, do you use the recently created FS coords from the 1st setup as the BS coords for the 2nd setup?
Just wondering how you would make this work in data collector software like SurvCE?
Would the traverse point coords just float in the adjustment, so it wouldn't be constrained by those values??ÿ
Or do you book it all by hand and create a raw file manually?
Thanks
@cf-67?ÿ
That's right, I will use the computed position from the first setup FS as the BS for the second setup (closing horizon).
In the fieldbook I would record the final mean H/V/SD values from both setups, then compare for the horizon closure and deltas.
?ÿ
I'm currently running Access, which is super flexible for data collection. TBC for QC and network adjustment. So it doesn't really matter what I tag each point as in the field, post processing cleanup is simple.
It's been about 5-6 years since I used SurvCE, so I couldn't say what would be the best method for recording in the field. It might make more sense to record a second, different point ID and simply modify the point number in the raw file afterward.
And if I were using SurvCE, I would likely be using StarNet for my adjustment, which would make it easy to quickly modify any duplicate point IDs in post-processing.
As far as constraints go, something like this does require tight coordinates for the initial station position and backsight - instrument position would be constrained H+V and the backsight position constrained H. So the traverse positions would be floating, and we'll end up with error ellipses on our traverse points.
Sometimes I will throw in RTK observations throughout depending on whether we can get good satellite conditions along the way. Any additional observations will help tighten things up. Those vectors are used along with the TS observations, not the positions, as I would rather the vectors get assigned realistic weights in the final solution.
@rover83?ÿ
Thanks for the good info - pretty sure I could get a processible raw file in SurvCE now that I've thought about it.
I'm going to try this on an upcoming job - piers and anchor bolts for a ski lift...
Am definitely interested in only traversing one way...the downhill way!
Lets say your starting from a known point (X) and want to shoot features and topo to an unknown, but specific, marked position (Y) down the road and around a few corners.
Would this modification of Jed's bowtie give adequate info to close and adjust this one-way route?
In this example X and Y are the start and terminus with no known coordinates on at least one of them.
Would it be crucial to have set up on both X and Y?
?ÿ
I love this topic. I can t wait some nasty and bitter critics to come over on what I am about to say.
The longest the better , always . Even in gps ing, observe always the biggest line . Also instruments accuracy is a factory mater not land surveyors issue at all.
If we can limit ourselves to just press buttons and read many times instrument hights keeping our pole high unmovable all works will be 100% accurate , but we always need to do some funny things that bring us out of the main line to follow .
More, there is a big difference between the way to compute errors and the way to adjust your traverse to meet the ACCURACY required. In an ipothetical world where we do all by book and put same effort (max effort possible) in every each set up we have this:
Example: Error for long BL > Pitagora for angular and linear error squared / L,is small . This confirm that your idea is correct . Again this is precision , not accuracy.
No instrument can state "the far the distance the more accurate", but they can say...as it is build , after 10000000000000000000000 of 1000000000000000000 obs on this distance , we can state that this day in this time on this conditions THIS length has been measured precisely with instrument sensible up to bla bla .We have measured a second length, half of the first , and again bla bla bla. All of this build up some sort of data that altogether will let someone state something like this...On long distance X our instrument , when is assumed this and that, is able to provide this discrepancy, on half of it this discrepancy and bla bla. So we can say from this data that our instrument (builded like this, with this with that... ) is accurate up to 5 seconds one seconds half a second, it behaves better in this distance than in this other one and so on(plus ppm) . They will build up some gaussiana funny bells, apply some math and with some functions (linear and not, I don t know) will be able to say more about what u are using.
On our side, on field we observe the behaving of the instrument and , READING the speck we relay on it and feel confident all should be fine.
Back in the office , after we uploaded our data we relay on software algorithms to compute whatever adjustment we like depending on contract, field, site elevation and length.
About adjustment , I don t feel confident to say too much but yet when we start talking about weighting obs we need to be careful about it because weigh an obs is not as straight forward as it might look like and will affect how the algorithm has to behave computing or ADJUSTING your traverse to meet the required standard.
For me Adjustment is yet a huge mile stone to be more accepted (because seniors sometimes impose to you to accept it as it is) than understood.
I might suggest that the term "proportionally adjusted" equals the two cases :
1/2 of L (invProp) and 2 of L (Prop)
The longest the better , always . Even in gps ing, observe always the biggest line.
A great many network design standards documents would disagree with this statement. Relative accuracy between adjacent stations is high on the list of goals for quality geodetic or local networks.
Also instruments accuracy is a factory mater not land surveyors issue at all.
This is just patently false. Instrument accuracies are used to estimate positions and relative accuracies, which are generally required for boundary work. If you don't know your relative accuracy, you can't certify that you met standards.
@rover83 First , thanks for this replay , u can t even imagine how happy u made me right now!
Second, I beg you , spare some time for me and give me a sort of example of "A great many network design standards documents would disagree with this statement."< What does this mean ? who is saying that , rics ? what s this doc and in which practice will this apply ? Can I read one of them ?
Third, is my sentence wrong on the whole length of it independently of what are and how we use this coords even in relation to the survey extend?
"Relative accuracy between adjacent stations is high on the list of goals for quality geodetic or local networks."
Relative accuracy between stations (A-B-C) means just pure row angles and dist (000 - origin) ? also is this just a standard we accept without investigating it, or how to say a sort of "if this > that" , and skip the rest because this rules ?
This bit confuse me a lot.
"Instrument accuracies are used to estimate positions and relative accuracies, which are generally required for boundary work. If you don't know your relative accuracy, you can't certify that you met standards."
OK , so when u say instrument accuracies what u exactly mean with that ? trunnion axis check for trimble for example done by us in my opinion is just something to avoid completely. I can do it, I don t like doing it.
How do I check the exact required distance to carry this check out, if not required exact distance how do I check the system I am caring this check out is not affected by variables I am not aware off ?
"If you don't know your relative accuracy, you can't certify that you met standards." How do you know it if the producer is not giving that to you and relating this to the instrument precision ?
I might have lost something this years but I am still at the stage where the 1m CALIBRO is in Paris conserved under particular condition.
Anyway, thanks a lot for the consideration.
Last request, and most important . Can you give me some sources from where I can see the computations that are holding your statements , please ?
Again, Thanks 1000 times .
Your first post was barely coherent; I responded to the only part that made sense.
Not a single one of the dozen or so federal and state manuals/guidelines for control work that I have on my desktop make mention of "observe the longest lines". That's a general statement and doesn't mean anything in the context of control work.
As for ignoring instrumentation standard errors during processing and adjustment, I don't even need to respond to that. It remains patently false just as I noted in my first response.
It's generally understood that the burden of proof lies with those who make the initial assertion against best practices, so it's on you to prove your point.
I can t wait some nasty and bitter critics to come over on what I am about to say.
Some language barrier to work through here, but I'd like to make some comments.....
1. You speak in terms of absolute and perfect measurements. Surveyors know that there is no such thing. We understand that no measurement is perfect, but if done correctly will be within a predictable margin of error.
2. Long baselines are usually desirable, but not dogmatically always best. You have to consider the local conditions.
3. The quality of an instrument is, indeed, a major factor in the accuracy and precision of the measurements it makes. But not the sole factor. It must be well maintained and handled to perform to its rated capacity. Also, the quality and condition of all the parts of the measurement system - the tripods, tribrachs, rods, prism glass, etc. play a role and must be equally well maintained and handled. Even the monuments themselves are important. You cannot make measurements of equal precision between crosscuts in concrete that you can to pinholes in brass disks.
@amksyr You're doing what we call a "dead head" traverse. Without the second point, you cannot adjust your angles, and can only apply Bowditch/Compass, or Transit, or Crandalls, or Least Square to the final one. Your method WILL WARP your angles you turned, and it WILL get ugly at any corners you made, but will bring it to close. You'd be better off scaling your traverse and rotating between the beginning and ending values, but that won't expose a blunder but will keep the linear geometry the same as what you turned in the field.
Norman, would u use short or long length in traversing along massive fields?
Also, if I immagine it , wouldn't be like the bigger the dist the worst the approx to the "real dist(curved)" ? I mean , the only critics I can do to my self is to not have understood the ground and how to takle it.
Long Tangents = bigger errors for massive fields , when u process traverse with earth curvature corrections only?
Except for the massive fields , I continue to believe that the longer the dist in quite flat ground , the more accurate.
"I have" process a site where everything was cool , balanced and everything, and I find out that the loop is "better" with no trigs on the Base Line ( Fix-Constrained) than with tangent dist.
I agree that a single EDM measurement of a long baseline would be more precise than several incremental measurements, all things being equal. It is just that, in real life, all things are never equal. Sometimes practical considerations and local conditions render breaking up the traverse into shorter legs desirable.
As far as curvature correction on very long lines, that C/R correction is a well known mathematical relationship that all modern instruments can be set to correct for. In any case, for really long lines, the measurement is not likely to be made with the EDM. We use GNSS for that.
