Instrument height, rod height, prism pole, depending on the type of survey mark, do you notice small differences between measuring with a tape measure, rod or prism pole, due the contact interface with the survey mark? Or what if you need to tighten your resected positions vertically, but don't want to mark them?
Topcon's GLS scanner uses a survey workflow with prisms, similar to a total station, it's very accurate on distance and horizontal angles, but vertical angle measurement isn't anything to write home about.
Vertical errors pollutes least squares analysis, a level traverse addresses it nicely but that normally requires marking all occupations. Why is vertical less accurate on the scanner? I'm not sure why, maybe it's something to do with how it recognises the prism, maybe the telescope is designed for spinning at high speed on its trunnion axis and some compromises are made when stationary collimating with the prism? The scanner likely uses a dual axis compensator and horizontal angle resolver common with total stations, but the telescope definitely isn't, yet vertical errors don't appear to occur during scanning.
So I came up with the crazy idea of cutting a slot in a prism target board hanging a tape from it and measuring the differences between stations with the level.
Observations in images are from the GLS scanner, two static GPS positions and a Sokkia PL1 Level, a total station wasn't used.
We have carefully measured the differences in height between our different types of prisms, total stations, scanner, target spheres, and GNSS receivers. We measure one HI, typically to the measuring mark on the side of the SX10/S6 or to the measuring mark on a rotating optical plummet (which is the same offset, etc as an S6). We will also measure a check HI when possible (GNSS antenna, scanner mark, etc). If using an unmarked station, then all of the HI's are with respect to the tribrach plate (i.e. where the instrument/prism sits.
We do something similar using Leica's height hook, so we know the height off the tribrach, when wer're marking ground points. But now, for resected positions, we're not marking ground points, but we're also levelling and measuring the height from the prism centre, treating the prism centre (195.5mm off tribrach) and total stn centre as the occupation point.
The GLS trunnion is 226mm above the tribrach, so we're using an instrument height of 30.5mm for consistency relative to the prism occupation.
Today, I'll set up our optical level to check our prism heights and slots cut into target boards. I'd like these spot on too.
Results so far suggest better vertical accuracy also improves distance and horizontal angle std deviations in least squares analysis.
Cheers,
Peter.
the height above tribach plate is 0.196 m for the Wild T2 (which I started on), Zeiss S10, Trimble S6, and Trimble SX10. The only instrument I use that is NOT 0.196 is my Trimble TX8.
I measured the SX12 and discovered it was 195.5. Interestingly the Sokkia prisms and carriers I'm using are also 195.5.
The Leica prisms I have are 196mm
I know, it's only a minor difference, but the first order level I'm using measures reliably down to 0.02mm it seems odd to round instrument height to the mm when the level is so accurate.
Cheers,
Peter.
"...rounding instrument height to the mm when the level is so accurate."
Especially since you're measuring a slant/slope distance with a pocket tape.
Forest for the trees.
Leica makes a nice little tape for height measure ups. It consists of a bracket that mounts to a tribrach. You hook the tape on the bracket and measure away. The tape is calibrated to for slant and the offset to prism/instrument center. Even with this method you are probable only good to 1mm but that is about as good as you are going to get with any method. The nice thing about the Leica's system is who all of the parts play together. All of the total stations and prism post have the same vertical offset from the tribrach.
I'm familiar with said Leica equipment/methods.
The photo showing "a pocket tape hanging from a cut slot in a prism target" ain't quite the same thing.
The ultimate irony of the thread is the pursuit of a more efficient workflow (to improve precision/accuracy due to contact with the interface with the survey mark) at the expense of overcomplicating the situation, additional imprecise measure-ups/downs, additional potential user error (book keeping), and then leveling to 0.02mm and calling it the bees knees.
Was it a fun experiment? Maybe.
Just use your equipment the way it was intended to be used, and adjust your LSA error estimates to account for each different source of error.
Understanding the "truth" of your data capture is more important than trying to fabricate additional precision/accuracy to a system where there is none.
If you need better precision/accuracy, then maybe different equipment/methods should be employed.
Chasing your tail for a millimeter or two has been the downfall of many.
"Leica makes a nice little tape for height measure ups..."
Even better, on the latest Lieca total station models (TS-16, MS-60) there is a built in EDM to measure the instrument height, which then automatically populates that field in the dc. Slick.
Another neat feature of Leica accessories is that if you use any Leica glass on any Leica rod you read the target height directly - and correctly - on the rod's scale. Slick.
I agree that height measure ups are a weak point in the system. Most just accept it and deal with it. Leica has gone further than any other manufacturer to address this weakness.
For the record, I've long felt that the common box tape is way too floppy to accurately measure the instrument height. Especially in any kind of breeze. For that reason I've long used a folding wooden ruler, which is much stiffer.
My MS60 is in the shop to have the HI measure up calibrated, it is consistently 4mm high.
The AP20's ability to update the rod height in the CS is great.
I took delivery of an AP20 on Wednesday and tried it out on a job yesterday. I agree, major game changer. Once you have it you will never want to be without it.
Trimble had these poles, which were nice.
However, you could get the point down in a dimple quite a ways, so I often wondered how many survey points were mis-measured over the years vs leveling to the top of the dome.
They always reminded me to check that the tips of the rover/stakeout rod were sharp/pointy.
How often do you check your rod points?
Leica height hook, true vertical measurement...
Topcon GLS also has a height measurement plummet laser, but it's usually 2 to 3mm out compared to the height hook.
I uploaded these images previously, they don't display when I checked, so just uploading them again. Least squares suggests improving vertical accuracy also improved distance and horizontal angles, most observations have std error deviations of zero.
I set up two tripods, with a tribrach on one and the Sokkia PL1 level on the other and measured the relative height difference between the Sokkia AP01 prism holders. The prism holder connect two ways, one via a V groove and shoulder into a tribrach carrier, the other via a 5/8-11" UNC thread.
So it appears I can significantly improve the vertical centring error by selecting the most accurate holders for the traverse kits, and shimming or stoning the remaining for pole mounting.
So it appears I can reduce my prism vertical centring error within 0.1mm or +-0.05mm, instead of +-0.5mm by calibrating my prism holders.
Photographs of level calibration of prism holder height. Note I've used one tribrach, one Topcon carrier (with thread attachment) and Sokkia AP41 carrier with v-grove and shoulder attachment.
Factors affecting instrument height v's levels; tape measures in windy conditions, prism pole points pentrating marks, worn points, different tools used to measure instrument height.
Level differences are included in least squares, instrument heights errors aren't.
Maybe a rod tip design that consists of a flat, used as a height reference datum, with a spring loaded tip that drops / retracts to align to the centre of the mark, the weight of the prism pole suffienient to cause the tip to retract. It would likely fill with compacted dirt and become stuck.
If I can use fixed height rods over control points, it minimises transcription of height errors, then level resected instrument positions, it saves the hassle of measuring instrument height and marking positions, it's much faster levelling from a tape (or small wood measuring staff) attached to a prism target board.
Cheers,
Peter.
@ Norman_Oklahoma
For the record, I’ve long felt that the common box tape is way too floppy to accurately measure the instrument height. Especially in any kind of breeze. For that reason I’ve long used a folding wooden ruler, which is much stiffer.
That's funny because I think it's just the opposite. I routinely subtract 0.01' from all height measurements.
What first order level are you using to measures reliably down to 0.02mm?
