Well, that's all well and good for theory. Can't hold a candle to years of acutal experience in a proven techinique repeated thousands of times with hard, consistent results. Skill with a plumb bob makes all the difference, and for those who don't even carry such at all times, best stick with your bubbles and do the error analysis instead of amazing closures in every case.
And do note the question was about prism ple BIpods, not tripods. In fact, tripods will decrease the accuracy and time factor of bipods and plumb line, not to mention weight and cost of gear.
Ancient tech still trumps new tech in many basic ways. Plumb line is far older than the wheel and for very good reason.
> Well, that's all well and good for theory. Can't hold a candle to years of acutal experience in a proven techinique repeated thousands of times with hard, consistent results.
Well, the point is that one can test the claim in about half an hour just by using the technique you describe to plumb the rod by eye with a plumb bob string and then measuring the error of plumbing with a total station set up about 5 ft. away (at a measured distance from the ground mark that the pole is being set up upon). The procedure is very simple and can easily measure the plumbing error at sub-millimeter accuracy. Repeating the set up 15 or 20 times gives a very good means to characterize the random errors inherent in the plumbing method.
> And do note the question was about prism ple BIpods, not tripods. In fact, tripods will decrease the accuracy and time factor of bipods and plumb line, not to mention weight and cost of gear.
Well, that extra leg of a prism pole tripod weighs maybe a pound, but it is such an improvement in the functionality over a bipod that the bipod is something not to even think about using as long as prism pole tripods are for sale. Again, that same centering accuracy test can be applied to plumbing a prism pole in a bipod to compare the results. The test takes maybe half an hour or less and is quite impersonal in that all that is done is the pole is plumbed and the angle between point of pole and target is measured to compute the plumbing error. The same procedure repeated 15 or 20 times will give a very good assessment of the random errors of the method as well as showing any systematic bias for that particular pole bubble.
The value of +/-0.4mm s.e. that I quoted was from an actual test of a target (HT = 5.0 ft.) on a prism pole in a prism pole tripod. The claim that something can be plumbed by eye, 5 ft. or more over a ground mark, with a standard error of less than +/-0.4mm using a plumb bob string at 15 ft. simply goes against both theory and experience. However, I'm willing to keep an open mind if someone wants to demonstrate by testing that it is actually possible.
1980 or thereabouts. Got my first prism pole and bi pod. Set it up. Checked it with the theo for plumb. Close by no bananna. Adjusted with plumb bob tech. Whao, much closer. Spend a good bit of time adjusting bubble. Still proven inferior to plumb line. Used the plumb line tech for 30 years. Results excellent. No further theory required. High winds addressed with foot weights made of logging chains. A mm with this tech stands out like a sore thumb.
You go right ahead and do all the proofing you wish. Play with your statistics for as many hours as you will. I will stick with what is both simple and gives best results for me. That is proven as bipod and plumb line too many thousands of times to guess at. You will be long dead before you ever get a set of data as extensive.
> 1980 or thereabouts. Got my first prism pole and bi pod. Set it up. Checked it with the theo for plumb. Close by no bananna. Adjusted with plumb bob tech. Whao, much closer. Spend a good bit of time adjusting bubble. Still proven inferior to plumb line. Used the plumb line tech for 30 years. Results excellent. No further theory required.
Oh, you've been speaking entirely about bipods, not the prism pole tripods which should significantly outperform the bipods when the bubble is adequately sensitive and in fairly good adjustment. Sure, pretty much anything is better than just plumbing a prism pole using only the rod bubble in just one position, even probably eyeballing plumb with a plumb bob string.
I guess I don't understand the resistance to just testing the random centering errors of the method you describe. If they're less than +/-0.4mm, that's what they are. And if they're not, they're not.
I use bipods and prism pole on every job I do - have been for 14 years. I check the pole for plumb once a week. Never have any trouble meeting ALTA/ACSM requirements. I also don't do huge projects. Shopping centers and apartments are the biggest jobs I do except for the occasional "big" boundary. 100-600 acres.
You go right ahead and do all the proofing you wish. Play with your statistics for as many hours as you will. I will stick with what is both simple and gives best results for me. That is proven as bipod and plumb line too many thousands of times to guess at. You will be long dead before you ever get a set of data as extensive.
> You go right ahead and do all the proofing you wish. Play with your statistics for as many hours as you will. I will stick with what is both simple and gives best results for me. That is proven as bipod and plumb line too many thousands of times to guess at. You will be long dead before you ever get a set of data as extensive.
Well, correct me if I've missed something here, but I don't think that your set of data includes anything as fundamental as a measurement of the uncertainty of centering using eyeball and plumb bob string. You may not have considered this, but in modern surveying that is basic data. It's ridiculously easy and quick to make that determination, but I'm not going to run out and buy a bipod which I'm unlikely to ever use, just for the purpose of testing a method I'm unlikely to use. I suppose I could take one of the legs off of one of my prism pole tripods, but I have hopes that one of the younger surveyors will realize that this is something he or she can easily do to answer this and other similar questions.
Naturally, statistics are involved as they are in all modern survey measurements since every measurement contains at least small random errors and statistical methods are very well suited to characterizing them.
My method is proven by decades of repeatability giving excellent results. Your iron skulled attempt to suggest otherwise is for you to address as you will. Do as you will, as will I.
> My method is proven by decades of repeatability giving excellent results.
Well, you apparently don't realize it, but in modern surveying practice, one needs to have, among other things, a fairly realistic estimate of what the uncertainties in the target (and prism) centering are. "Excellent results" is not a measurement of uncertainty but a statement of optimistic belief. With realistic estimates of target and prism centering uncertainties, a surveyor can modify the uncertainties in the measured angles and distances and weight them accordingly when the survey is adjusted and the errors in the work are analyzed. That way, he or she can document that, to a reasonable certainty, results really are as good as specified.
I realize it's an entirely new way of approaching the question of survey accuracy to have to actually put numbers on the uncertainties of survey measurements instead of the subjective assessments of an earlier era, but it just isn't that difficult and it's very useful.
You have a great deal to learn about the practical application of the ancient plumb line. But such learning is not available to those with cast iron skulls. Sell your theory to someone who prefers that to empirical proofs of efficacy.
Mr. Stephens, I Take It You Have A Lot Of Time
to step at least 10' away in 4 directions, wait for your plumb line to steady, observe, repeat, adjust the rod and reobserve 4 times, repeating if neccessary.
More often than not in my work it is not possible to step 10' off line in either direction so I have to rely on properly adjusted equipment.
By having a properly adjusted rod bubble, I set it up and proceed with my work. When I go to retrieve it I can tell instantly if it has moved.
Kent, I have 3 SECO tripeds and it is a joy to adjust the rod bubble with them since the rod can spin at will. I have 1 SECO bipod and have on occassion removed the 3rd triped leg when I knew I would need extra bipods.
However it is not neccessary, a triped stays steady with the 3rd leg hanging loose and if you are concerned it can be velcroed from flapping. The 3rd leg hinges close to the rod, while the quick release legs are farther from center, promoting 2 leg balance. The fact they are at 120° and not 90° does not affect stablity.
So Kent, go ahead and do that 2 legged adjustment to confirm that it is indeed a fumbly-mumbly way to work.
Paul in PA
Mr. Stephens, I Take It You Have A Lot Of Time
> go ahead and do that 2 legged adjustment to confirm that it is indeed a fumbly-mumbly way to work.
Well, you and I recognize that the practice that T.P. describes is inferior to just centering a prism pole in a prism pole tripod with a sensitive bubble in good adjustment (and verifying the bubble adjustment at the same time).
Oh Great!
> > go ahead and do that 2 legged adjustment to confirm that it is indeed a fumbly-mumbly way to work.
>
> Well, you and I recognize that the practice that T.P. describes is inferior to just centering a prism pole in a prism pole tripod with a sensitive bubble in good adjustment (and verifying the bubble adjustment at the same time).
[sarcasm]Now I have to replace all my prism pole bubbles with sensitive bubbles (in good adjustment) to use Kent's method![/sarcasm] Kent I believe what you have here is another solution in search of a problem.
In all seriousness, If I was having a problem getting good closures, I would investigate your methods, but if your closures are consistently good, then it would be more prudent to invest your time elsewhere.
Oh Great!
> Now I have to replace all my prism pole bubbles with sensitive bubbles (in good adjustment) to use Kent's method! Kent I believe what you have here is another solution in search of a problem.
Perry, you evidently aren't aware of this fact, but the manufacturers of prism poles sell them with a variety of different bubble accuracies. You've done carpentry. I'll bet you didn't use a torpedo level on a door frame. Same idea. Oh, you need to actually level your total station and center it over the ground mark that you're trying to set up on, too. Bummer, I know. :>
> In all seriousness, If I was having a problem getting good closures, I would investigate your methods, but if your closures are consistently good, then it would be more prudent to invest your time elsewhere.
In modern surveying, we don't really care so much about the error of closure of a traverse as we do about more important stuff like the relative positional uncertainties of things. That is how survey specifications describe survey accuracy increasingly commonly and it's an answer that you need to have realistic estimates of things like the uncertainties of angle and distance measurements, GPS vectors, and centering of targets, prisms, and GPS antennas to get at. Saying you got a good feeling about the error of closure of a traverse doesn't get you much.
Oh Great!
This week I am successfully running the robot with nothing but a plumb bob hanging from a prism.
When I put it in star*net it is just as accurate as the equivalent number of sets with a handheld rod. Slightly tighter in the horizontal (the forced-centering effect) and slightly looser in the vertical (it is a bob, after all), but still overkill for anything short of anchor bolts or a particle accelerator.
Oh Great!
> This week I am successfully running the robot with nothing but a plumb bob hanging from a prism.
>
> When I put it in star*net it is just as accurate as the equivalent number of sets with a handheld rod.
Okay, I'll bite. What are the values of the standard errors of centering for the plumb bob/mini prism (I assume) combo you're using for both handheld prism pole and bob/prism?
Oh Great!
With the handheld pole, I use 0.003' in H at both the instrument and the target.
With the handheld glass/bob combo I had to crank the H at the target end to 0.02'
Oh, the horrors! It might compound to 0.04'!
Oh Great!
You might wonder how the horizontals are "tighter" if I had to loosen up the standard error for the centering. Here are my thoughts about that:
With the plumb bob, by the time you get the bob steady, you have the glass VERY steady. It wiggles less in the horizontal than the handheld rod (for me, anyhow). All the wiggle is at the bottom and the glass is at the top. The trouble is in eyeballing whether the tip of the bob is truly over the ground point, and this is where the centering error gets larger. So the method is ever so slightly more precise but somewhat less accurate.
But hey, less stuff to carry. More than enough accuracy.
Oh Great!
> With the handheld glass/bob combo I had to crank the H at the target end to 0.02'
Yeah, that's about what I'd expect. It sounds reasonable.
Oh Great!
> You might wonder how the horizontals are "tighter" if I had to loosen up the standard error for the centering. Here are my thoughts about that:
>
> With the plumb bob, by the time you get the bob steady, you have the glass VERY steady. It wiggles less in the horizontal than the handheld rod (for me, anyhow). All the wiggle is at the bottom and the glass is at the top. The trouble is in eyeballing whether the tip of the bob is truly over the ground point, and this is where the centering error gets larger. So the method is ever so slightly more precise but somewhat less accurate.
Unfortunately, that's inconsistent with what we mean by "standard error". The standard error is a measure of the random errors inherent in a process. What in effect you're saying, if I've understood you, is that the random errors in the process to which you assigned the smaller standard errors are larger than those in the process with the larger standard error assigned to it. That is a fundamental inconsistency, I'm afraid, that indicates the values of the standard errors were wrongly chosen.
