A Baseline Only Allows Checks Of The EDM
There are many ways to check the circles on an instrument.
I've used closing the horizon beginning at about 0-60 and 180(never at 60-120 or 180) to points at different heights.
I've also ran "distance" base lines, and set on measured intervals to turn 90's(right & left . . . not 90 & 270. . . reversed backsites) and measure out on the 90, while checking the previous 90's(both ways) I have turned.
I always ended up very close, but the patterns I often saw were interesting.
There's always a number of ways to do anything.
I don't think there is anything wrong with the numbers. You find and report the error more accurately (2-3 times) with the 1" gun. Those smaller errors are hidden in the results from the 3" gun because it is not capable of detecting them. The repeats should be within twice the least count (2" and 6"), and from a statistical standpoint you get closer to the true angle capability of the gun the more you turn (up to 7 times), but it's still a statistical number arrived at from an uncertain range dependant on the capabilities of the gun. Try the same thing with a 10" gun (recently calibrated) and I bet you come up with even better closure. The 10" gun should not be able to pick up the small errors present at all, from a practical standpoint. Course there could be something wrong with the gun too:)
Amen Perry.
Only one bad instrument. The 3 second is the good one. I've been using it for years and it's been excellent. The new 1 second is the problem.
I'm not really sure I understand you Duane.
The error of the traverse (in particular the one I ran both instruments over), without instrument error should be miniscule. It's a closed traverse, professional grade matched prisms and tribrachs, no setup changes (we had enough legs and tribrachs for every station) and done in perfect weather conditions.
You seem to be saying that the closed traverse I ran really has an error in it of, say, 6 - 10mm, and it's just a fluke that the 3 second instrument reports better.
And you seem to be saying that if I ran a closed traverse with a 10 second gun I'd close even better?
You have run the two instruments through exactly the same setups, but the only comparison you've given us is the total misclosure. What happens when you compare the two of them shot for shot and angle for angle turned at each setup through your traverse. Is there any consistent variation between them?
Look For A Circle Error
Today i did some more checking on the instrument.
I setup two targets - Leica GPR121s at about 100m and 70m with an angle between them of around 85º.
I then turned 5 sets of angles with the socket of the instrument pointing towards the furthest (rightmost) target and recorded the angle between the two targets.
I then removed the instrument from it's tribrach and turned the base of the instrument 120º and replaced it in the tribrach. Then I read the angle again (reading from a different part of the horizontal circle).
Then I repeated that a third time and read from the remaining third of the circle.
Then repeated the whole procedure a second time, to ensure repeatability.
Finally, I reshot the first set of angles again as I had a 2 second difference between my first and second set readings in that sector.
Results
84º07'13" (sector 1. socket facing approx rightmost target)
84º07'05" (sector 2. socket facing approx 120º to right of rightmost target)
84º07'02" (sector 3 socket facing approx 240º to rightmost target)
84º07'11" (sector 1. socket facing approx rightmost target)
84º07'05" (sector 2. socket facing approx 120º to right of rightmost target)
84º07'02" (sector 3 socket facing approx 240º to rightmost target)
84º07'12" (sector 1. socket facing approx rightmost target)
Looks like a definite issue with the horizontal circle of that instrument.
Looks Like You Found It
Whether it is fixable, is a different question.
Based on your test it is a 10" gun at best.
Remember that doing D&R you are reading from opposite sectors of the circle, so some of the problem will disappear.
Paul in PA
Look For A Circle Error
About 2 mm of radial centering error would explain your results. Is there anything strange about how the instrument fits in the tribrach? Although 2 mm should be pretty obvious.
Look For A Circle Error
Its a new tribrach and instrument.
The other instrument (which is good) shows no sign of having an issue using the same tribrach.
If we're looking for an error of 2mm, then the instrument won't actually fit in the tribrach. There isn't that much clearance.
There has been persistent funny angles with this instrument. That's why I've been checking it.
The instrument and optical plummets always match while traversing.
In short, it's not due to the instrument/tribrach interface.
I had a Leica give me those errors also. It ended up being a bad compensater. We would get errors of several minutes on doubled angles. This was back in 98 (roughly), when the ATR Systems were first coming out. I was turning angles manually, and always had good closers with our other instruments. A few times we'd break out our old Leitz, set it up on same tribrach, legs, etc. no adjustment, and get 2-5 seconds on the same points. I'm not sure your experiencing the same glitch, but my 2 cents.
Joe Glidden was fond of saying "Forced centering is fine if you never need to use the hub again" and I agree with him.
1:80,000 should be great for nearly EVERY facet of surveying, save possibly mine and or heavy construction.
However, if you feel it's bad, any traverse you've run, when run backwards, shouldn't cross, if it's as you say it is. If it does, even at 1:80,000, then your IMAN may have an issue with the optics.
In the East Texas hills, 1:30,000 is a fine closure. 1:80,000 begins to make most thing of pencil whipping, while it can happen though.
Buy Trimble. 😉
Why, do Trimbles never have issues?
Anyway, problems solved. We handed in our results and the instrument is being replaced. Should have our new one next week.
Look For A Circle Error
> Looks like a definite issue with the horizontal circle of that instrument.
Excellent method for determining the eccentricity of the horizontal circle. Definitely a manufacturing issue. Hope you enjoy that new instrument.
JBS
Look For A Circle Error
>removed the instrument from it's tribrach and turned the base of the instrument 120º and replaced it in the tribrach
I don't think I can do that with my old Topcon instrument and tribrach. There is a large slot in the tribrach and a plastic tab on the bottom of the instrument that only lets it go in one way.