Error due to object near line of sight

> To detect diffraction, move the focusing knob slowly back and forth while watching the target. If the target appears to move relative to the obstacle and the cross hairs, the obstacle is causing diffraction. Offset the line to correct this condition.

That's how to detect errors from diffraction on grazing or obscured lines.

There is another source of angular errors that occurs along grazing lines and that is from lateral refraction caused by an abnormal horizontal or vertical temperature gradient near a surface that is much hotter or colder than the ambient air. The rule I follow in that case is to just avoid grazing lines, particularly if target blurring or scintillation (image dancing) is present as seen through the instrument telescope.

Convenience and economy of effort are not adequate reasons to do a job poorly.

This thought comes to mind somewhat frequently when doing total station work through foliage. How do we know when we are getting faulty data? Any number looks good compared to the dreaded screen reading indicating no number can be obtained.

I've often encountered this and wondered, and yet apart from long grazing lines across hot bitumen haven't had an issue to date.
I'm not refuting any comments here, merely sharing my experience.
There are times when focus to target I can see it, focus on obstacle and it's partly blocking the target.

Recently I traversed a town with forced centering, about 10 set ups.
One line I wasn't happy as when I sighted the target it was evident] a small ornamental tree was on line.
When I changed focus the central stalk (about 1.25" diameter) was dead on line.
I then reobserved whilst the tree was pulled to the side and got exactly same bearing and distance.

Similarly I have observed through car windows and measured distances.
My approach is caution. If I have a doubt I re-observe by an appropriate method.

We do a lot of Structural Deformation surveys. We are using a T-3 in a forced centering tripod, siteing a lighted target at night, then measuring the offset of various points using a target mounted in a machanist micrometer, measuring to 0.001 inch. We have noticed over the years that as things encroach upon our line of site, the closures get worse. If anything is within a foot of the line of site, it casues errors. We have a 1500 foot long line that is totaly open, and we often get a closure of +- 0.010 inches, but others with many encroachments where we are happy to get a closure of +- 0.100 inches. We do our observations at night, and in January February and March so it is normally cold. We find that after an overcast day, we get the best closures. We suspect that it is casued by less heat waves from sun warmed concrete over our line of site.

Focal parallax can be very frustrating. It is true if the target moves while focusing through it, the location (observation) is useless. I would bet that most us that have experienced this were either looking directly down a fence line or directly down a structure wall. Photons are our friend, but can misbehave, as predicted by Albert Einstein...

This environmentally induced error is not a function of the instrument and no amount of mechanical manipulation will reduce the effects.

Although quite a few surveyors still use optical instruments, their use has declined. Probably in decline also is the general concept and implementation of observed redundancy and proper procedure. Observing a target or station from one setup is a sideshot..I don't care how many angles are wrapped or distances recorded. Sideshots don't mean much in the world of reliable positioning...you get what you pay for.

The only way to actually fix a foresight's position with optical observation, with any predictable reliability, is from multiple known stations.

Instrument Operations 101....make sure there is nothing in between you and the backsight or foresight. If there is, move the gun...;-)

> Convenience and economy of effort are not adequate reasons to do a job poorly.
>
> This thought comes to mind somewhat frequently when doing total station work through foliage. How do we know when we are getting faulty data? Any number looks good compared to the dreaded screen reading indicating no number can be obtained.

[sarcasm]I guess this means you are now a member of the "Put a Dimple on Your Cap" club?[/sarcasm]

😛

We have several replies saying diffraction and refraction are serious concerns for precise work so should be checked for and avoided, and some anecdotes that don't clearly differentiate diffraction versus refraction effects.

Does anyone have an equation to predict what C clearance is needed at D distance for less than A deflection of the sight line?

> We have several replies saying diffraction and refraction are serious concerns for precise work so should be checked for and avoided, and some anecdotes that don't clearly differentiate diffraction versus refraction effects.
>
> Does anyone have an equation to predict what C clearance is needed at D distance for less than A deflection of the sight line?

You're asking specifically about angular errors from refraction, right? I'd think that would be very difficult to model since the zone of the air producing lateral refraction effects near an object would depend upon various factors, including:

- temperature of ambient air away from object,
- temperature of object,
- wind speed and direction,
- condition of flow near object (laminar or turbulent).

Probably, the most reliable means of gauging effects would be via seeing. Lateral refraction should produce either some observable lateral elongation of a target, I'd think, or significant blurring of it as seen through the telescope.

Makes me think of relativity humor:

Yo momma so fat, observations in her general direction are distorted by gravitational lensing.

yo mamma..

..so big, her apron got 2 area codes!

> Makes me think of relativity humor:
>
> Yo momma so fat, observations in her general direction are distorted by gravitational lensing.

LMAO!!! Thanks Andy. I'll be sure to remember that next I'm around other physicists. Might be a while. Last physicist (got his Masters from GA Tech) I worked with was over 20 years ago. He also might very well be one of the worst computer programmers I've ever known. Apparently, pure logic is not for everyone.

As I recall, Guy Bomford discovered that while serving as a Royal Engineer in the Survey of India. He called it the "Hot Rock Correction." See Bomford's "Geodesy."

Interesting topic!!! Surveying, physics and humor all in the same thread. Awesome!

As to simple diffraction I was always told to avoid grazing objects in my line of site. I never knew there was a way to actually detect it while running the instrument.

As Kent mentioned about the target "dancing" due to heat. No one ever told me that could/would be a problem. When I saw that happening, I would just watch the object dance a while, due my best to focus on a mid-point and take the shot. If it was really bad, I'd take several shots and look at the results to see if it was consistent and then store the shot.

For my reply to Andy's reply about the last physicist I worked with... I was wrong. At the last engineering/surveying firm I worked at(8 or so years ago), our survey manager was a physicist by training who was going for his surveyor license. He was a physics major with a mathematics minor and I was a computer science and math major with a physics minor. He's the one that told me diffraction running the instrument(s). One can imagine we had really interesting conversations on rain days.
I "miss our conversations" (as Ken Watanabe said in 'The Last Samurai').

A high school drop out's take on the subject:

Being one that detests formal instruction almost as much as a prostate exam I have a hard time splainin' what my brain is thinking. I cheated my way through Calculus (as my brain approached zero)... however, pictures do help...

I don't think the phenomenon of objects appearing to 'move' when focused through a theodolite is a result of atmospheric refraction, or reflection for that matter. I believe it has to do with the way the telescopes are constructed.

While you can view a small amount of parallax with binoculars, it does not approach the intensity or appearance that is evident through a theodolite. I believe the difference is that in a most surveying instruments it is the objective lens that moves when focusing. Most binocs focus at the other end.

While I may fall short being able to utilize the Greek alphabet in labeling angles...this diagram is what I think is happening when one experiences focal parallax:

With all those little lines and angles in there I think one might be able to come up with some sort of formula to predict the actual location of the range pole. I also think that the focal length AND size of the objective lens would be a factor.

I also think that is a lot of trouble to go through when you should really find a better place to set up the gun...:snarky:

PS - Has anybody ever held a half dollar coin centered over the objective lense, then looked through the gun? You can still "see" the entire field of view, but it distorts as you focus. The ONLY light that is getting into the barrel is through the 3/16" sliver that the coin isn't covering up..

Everybody says, "just avoid blocking the view" but you have to be alert to that tree branch or other obstacle that isn't visible in the scope unless you rack the focus through its range. Having a feel for the range of the effect seems worthwhile.

I set up an experiment to explore the effect (minus thermal refraction). The instrument is in my garage and pointing at a screw that mounts the light on the neighbor's garage about 200 ft away. I clamped a piece of aluminum plate to the guide on my table saw and set it 1.5 meters away from the instrument (close focus limit). A scale taped to the bed of the saw lets me measure the distance off axis.

So long as I didn't mess with the focus, I found no sighting error as I slid the plate in increments across the view until it got too dark to see the target.

What happened was the focus got very sloppy.

With the plate centered in the view there is a range of maybe 15 seconds pointing depending on the focus chosen, and the view of the target screw is clear enough you might not notice something was partially blocking it if you weren't being very careful. As more of the view is blocked, you notice image degradation, but might be tempted to take a reading that could be anywhere in a 20 second range.

At 2 cm off axis the view is quite clear, but I can notice the image shift relative to the crosshairs a few seconds while the focus is in a tolerable range.

With the obstacle at 3 meters, I again saw no shift before messing with focus. A hole in the blocking plate was letting some light through, and I saw a 3 minute change in pointing over the focus range on a very degraded image, and again 10 or 15 seconds with usable focus.

Paden's diagram might help explain this, if I can digest it. My knowledge of optics is usually limited to the simple model that would not take this sort of thing into account.

Thermal refraction, of course, can be worse than that.

From someone with too much formal instruction

Paden Cash is right on. It is not refraction causing this. I suppose refraction is theoretically possible when sighting through a total station but it would require looking through a very small hole or slit. An object like a pole or a branch wont cut it.