Hi Guys.
I'm having an issue with a total station. It's not doing a great job closing traverses.
Scenario: Closed traverse. Constrained centres. Matched Leica prisms and robots. 4 sets turned per stations using ATR. Both instruments recently calibrated.
We have an older 3 second instrument that will reliably close a traverse to 1:150,000 or greater in horizontal (and typically better than that in vertical, which is unusual, but I'm cool with that.
We have a new 1 second instrument that has never achieved better than 1:80000 and seems to be a bit "fuzzy" with horizontal angles. Left and right face readings check ok, but the angles still seem off.
I'm certain it's flakey, but I'd like to prove it and find a way to demonstrate it's bad, before sending it off to be fixed.
Question 1: What misclose ratios are you getting with your traversing.
Question 2: Any ideas on how to prove that there's somethig funny with this total station.
one thing to consider is that sometimes the prism offset is getting applied twice-once in the gun and again in the data collector.
Have you checked the edm on a calibration line?
It's definitely not an EDM issue. It's a Leica system end to end. The distances are fine and match between both instruments. One instrument is too new to have gone over the baseline. The other went over the baseline not too long ago.
When was the last time you calibrated your tribrachs?
What is wrong with 1:80000 closure?
Are are your foresites and backsites balanced?
Are you inverting the scope...2 direct, 2 reverse readings?
Old school, but still apply.
> ....Closed traverse. Constrained centres.
By "contrained centres" I presume you mean what I call "forced centering".
> .... 4 sets turned per stations using ATR.
What are the splits like on these four sets? And how does the same angle turned with your various instruments compare?
> Question 1: What misclose ratios are you getting with your traversing.
I've been doing Least Squares for years so I haven't formally calculated a closure for all that time. But, I think, 1:150,000 is pretty hot. I wouldn't expect to get that every time. 1:80,000 is not all that bad.
> Question 2: Any ideas on how to prove that there's somethig funny with this total station.
I'd want to check the tripods and tribrachs being used. Are the tripods wobbly or the head loose? Does the instrument seat firmly on the 3 nubbins in the tribrach plate? Have the tribrachs been adjusted in the last week?
Could be the user, also. Have the 1:150,000 crew run the 1:80,000 instrument and see what happens. Is the crew getting the feet set firmly when they set up? Is somebody laying hands on the legs after the instrument is set up?
I'd kill for 1:80,000 in this hilly country.
I hear ya Perry...CT class "A-2" (industry standard) survey is 1' in 5000'.
> Question 1: What misclose ratios are you getting with your traversing.
what is the actual closing error?
and how does that compare to your shortest leg?
ratios can be very misleading.
I remember a case where a very long traverse (miles) had a real good ratio closure but there was a bust of Several Feet in a short leg.
For almost all survey work, I'd consider 1 in 80,000 pretty good. I've gotten 1 in 1,000,000 before, but at that point, its more luck than skill.
tribrachs and legs
we just replaced our tribrachs as the 3 pins under the instrument were not locking tight from wear. we noticed the issue very quickly so it was not an issue.
when i took the Wentworth PLS program we got everyone to bring in equipment. we has a 26-point loop with 26 legs and 26 tribrachs. my lab partner and i spent the day breaking down every setup and tightening the legs, most were wobbly as a weeble. the class results were interesting and you could see the difference between our data and the groups ahead of us. the group behind us matched very well... even if they griped that we took time to do all that work!
Collimate the intrument and see what that tells you. I'm guessing your tripods & tribrachs are probably fine as evidenced by the older instrument's closure capability.
We use the Least Squares adjustment on everything now. Your ratios are excellent considering you are turning 4 SETS by hand. I would do four sets if the situation required with a robot maybe, but by hand no thanks. Further more you are kind of splitting hairs. On a 5000' traverse you are talking about .035 ish to .070 ish out. Seems to matter very little in the majority of the surveying work I have done.
You should run both guns thru the same three or more leg traverse and compare the data
Even a new tribrach can be out of adjustment along with any other instrument or prism setup that has a level bubble and/or crosshair to line up.
Optics may vary too.
Basically, the difference between 1:80,000 and 1:150,000 is very little and beyond the expectation of any requirements I know of unless you are doing machine alignment.
The quality of your instruments will not be the same, even if they are the same brand and model. Instruments are built within specs and not identical. Use and handling can change their quality.
How are you closing angles checking?
I believe you can readily get 1:150,000 closures with good procedures. I have done it consistently. All the recommendations are good, but it sounds like you know that, because you have instruments that do better than this new one.
Off the top of my head, I say set up a three-point intervisible traverse course. Set up solid tripods (whatever it takes) with adjusted tribachs. I like the topcon tribachs that have an built-in optical plummet and a see-through optics that you can use the instruments optical plummet on. I would traverse through the tribachs (don't tear anything down until your done, and clip the instrument in to the fixed tribachs. Turn angles with one gun, unclip and turn them with the other close to the same time. Fine-level should look good on both. Instrument should appear over the point through the guns optical plummet on both....etc.
That comparison, I would hope, should convince a good vendor who stands behind their equipment.
Oh, and by the way, I say take a brand new instrument out on a good calibration baseline first thing. You want to make sure it doesn't come out of adjustment.
Just some thoughts.
I don't use Lica but with my S6 Trimble says that as soon as you get it back from the shop you need to run it through the adjustment routines. Didn't make much sense but that's what we do. Maybe running all the adjustments on the total station will tighten things up. No doubt you've done them but if you haven't I'd do that first thing.
Your 3 sec Leica might be a 1.5 sec gun and like wise your 1 sec gun might actually be a 2 second gun. Have you checked them out via ISO standards?
Ralph
When I worked in the mountains of western NC, we would have killed for a 1:80000 closure especially with all the wild verticals and un-balanced front and backsight distances.
I remember one setup that gaves us fits for well over an hour before we decided the gun was bad so we called it a day. There was 3 of us that day. One guy on fronsight, one on the back and me on the gun. I regularly turned sets under 10" so no one could figure out why I turned 4 sets all of which were way out by 1'30". We all checked our heights and levels and all was good. Still no joy. Beginning to doubt myself at this point, I had the other guys have a go at it. Both were far more experienced gunners than me but I regularly turned better sets than anyone in the office. Their turns were far worse than mine at well over 2' error. We decided something must have gone wacko on the gun.
Back at the office we checked out the gun on some known nearby points. It was fine. Next day, we took the same gun and another just like it and tried again. First setup and set of turns, I hit under 5". Baffled, we just continued on the rest of day with similar results even with some wild verticals on the following setups. The weather conditions were almost identical.
No one had an explanation except for: "sometimes that just happens". The LS at the office said he had seen it happen before. If any others involved back then remember, I'm sure they are still just as baffled then as now - as am I.
> When I worked in the mountains of western NC, we would have killed for a 1:80000 closure especially with all the wild verticals and un-balanced front and backsight distances.
>
> I remember one setup that gaves us fits for well over an hour before we decided the gun was bad so we called it a day. There was 3 of us that day. One guy on fronsight, one on the back and me on the gun. I regularly turned sets under 10" so no one could figure out why I turned 4 sets all of which were way out by 1'30". We all checked our heights and levels and all was good. Still no joy. Beginning to doubt myself at this point, I had the other guys have a go at it. Both were far more experienced gunners than me but I regularly turned better sets than anyone in the office. Their turns were far worse than mine at well over 2' error. We decided something must have gone wacko on the gun.
>
> Back at the office we checked out the gun on some known nearby points. It was fine. Next day, we took the same gun and another just like it and tried again. First setup and set of turns, I hit under 5". Baffled, we just continued on the rest of day with similar results even with some wild verticals on the following setups. The weather conditions were almost identical.
>
> No one had an explanation except for: "sometimes that just happens". The LS at the office said he had seen it happen before. If any others involved back then remember, I'm sure they are still just as baffled then as now - as am I.
That happens all the time, if you take the gun out and don't let it acclimatize it will happen, likewise if there is some uneven heating and cooling cycle causing the gun to expand and contract it will happen. That's why Thomas McGrath always carried a Parasol through all the ridicule.
Ralph
Tribrachs - recently.
Depends on application.
yes.
yes.