Reduced traverse accuracy - why?

Tribrach, tripod, temperature shift, tired eye, short sites.......

Any slight error can change 1:75+k to 1:24k inside of 143meters.

Didn't even use my eyes - was set up with autolock onto the prisms.

Pins were well centred. I always rotate the carrier in a couple of positions to check they're centred.

Anything to do with not reading in both faces?

I think that is a likely suspect, although 1:24K is only 6 mm at that distance. What is the reason you did not take D+R shots? For us, D+R is standard procedure for anything other than a topo shot.?ÿ ?ÿ

You don't say what kind of traverse it is, closed loop??ÿ

Is it possible the collimation error of the instrument is only taken into account if you collimate at each setup?

There are no short cuts. you slacked off in your efforts and the results rewarded you.

Hi guys

It was a closed loop traverse. I normally take shots on both faces but read in the Trimble manual that its not necessary.

The manual states;

"The instruments design facilitates the ability to measure all points with a single pointing to the target in the face 1 position. In order to compensate for collimation errors the collimation routine allows the operator to accurately determine the current errors in the instrument and store the errors as corrections to be applied to all measurements made in a single pointing to a target."

As the instrument was calibrated recently I thought I may as well just measure in one face - like the manual says.

Cheers, Andrew

A "3 second gun" is pointing error.?ÿ An angle is 2 pointings, minimum. Measuring and angle in both faces means 4 pointings. It works out that in order to get 3 second angles out of a 3 second gun you need to do 2 sets of FR/FL readings. And this is assuming that your instrument is in perfect collimation. Turning sets balances out those inevitable imperfections. So there is certainly some degradation of precision by only turning one angle. And it might amount to as much as you are seeing.

The instrument specs and collimation is for the gun. It says nothing about the performance of your glass, the adjustment of the tribrachs, the flex in your tripods. Etc., etc., etc. It sounds like you have good equipment at your disposal but nothing is perfect. Averaging more readings usually has a moderating effect on those errors, too.?ÿ

Since you are running robotic sighting there is very little time saved by turning only on angle.?ÿ

?ÿ

Yeah they all say that! The instrument compensates, but only to it's know parameters. Do a calibration, then repeat 1 minute later, 1 hour later, 1 day later. It drifts.

So it is absolutely a false economy (of time) not to do at least 1 set face and reverse.

With a lower precision gun I would want to be doing multiple sets.

Comparing long linear traverses with GPS observations highlights how much difference a few seconds can make to traverse station positions.

Just seen your traverse is only 143m? not really a problem then.

The estimate of centering accuracy for your gpr121 is 1mm. If you break down that traverse of 143m into say five evenly spaced legs with, making it ten separate occasions of prisms positioned, and distribute your closing error throughout those ten separate instances (assuming 0 error in your instrument setups) , your errors are within the centering error estimates of the glass.

0.019'/10 = 0.0019' < 0.0033'

I do 4 sets' doesn't take very long because it's automatic.

Did the prism type (offset) somehow get change?

Many thanks for the replies. I suppose considering I only did one shot on one face, and the number of short legs, that the misclosure is pretty good!

Will go back to doing them on each face. Its only a few more seconds more on each set up.?ÿ

Cheers, Andrew

?ÿ

I have used many different brand/models of robots over the last 20 years and almost all of them required a D&R to get good results, especially for elevation work. However I bought a Trimble S3 about six years ago and have come to the conclusion that a D&R with this gun does not improve the results over a direct shot. This conclusion was made from a lot of testing and comparison over the last six years and I can find no difference, even on longer shots. I still turn a D&R on some of my longer control shots, just because I feel like I should.

I'm not sure what to attribute this to, but feel it must be the applied /stored collimation. I typically run the gun through the collimation routine at the change of seasons and hardly ever see a change in the stored values, which are also always very small. With robots I've used in the past the values were always much larger and I could run the routine back to back and get different results every time, which never gave me much confidence.

?ÿ

All that being said, check your rod and tripod. Winter time temps always seems to loosen my tripod head screws.

?ÿ

?ÿ

?ÿ

Compensators may not have settle on reverse shots. Slow the shot down and see if you get something different.?ÿ

The problem is this. Think of the error as an aspirin. it is an entirely different matter to dissolve that aspirin in a teacup?ÿ or you can dissolve the aspirin in the ocean. If you have short legs and/or the overall length is not very far, then you will not be able obtain a closure?ÿ much better than 1:24k.... not enough water to dissolve the aspirin in....?ÿ each tribrach has some error in it (centering and pointing etc) the manufacture specs for the instrument?ÿ is what 2mm?ÿ plus some parts per million??ÿ Not counting all the other error sources?ÿ what is my expected error in 28.6m with an instrument error of 2mm? I am using 28.6m as an estimate of the distance between 2 stations where the overall traverse is 143m.