Optical
Old value H 00.00'00" V 00.00'00"
new value H 00.00'02" v 00.00'13"
It seems like no matter what I do I can never get the new v under 10"
trunnion tilt axis
old value T 00.00'32"
new value T 00.00'02"
tracker test results
H 00.00'09"
V 00.00'12"
These values are well below the "take it to the shop" values.
What is a proper expectaion for performance with these types of collimation values?
thanks for your help
Collimation means to adjust the line of site. So by collimation of the instrument one is simply adjusting the error from the instrument. In electronic instruments we are reprograming the instrument, 13 seconds = zero. As long as the instrument is within specifications the operator is good to make the adjustment in the field by "collimation of the instrument", and the performance should be as good as new! On older mechanical instruments we were doing the same function by doubling the angle. Or turning direct and reverse and setting the point at the physical split of the two sightings.
This discussion is one that I have had with many of my cohorts. Many of us think that we have to send an instrument to a guru to make sure it is working correctly. Since I can remember instruments have come with their own tool packet and we were expected to make field adjustments too level bubbles cross hairs and more. Now we just do it electronically. First thing I do when I get my instrument back from service is put it through the collimation process because 1. I know how mail is handled 2. I don't know the technician that worked on the instrument. My 2 cents. Jp
Those numbers look typical of those I saw- over 8 years and many collimations -on 3 different 5603's my former company owned.
For me, the proof is in the pudding. What kind of splits are you getting on sets of F/R sightings and angles after collimating?
Collimation means to adjust the line of site. So by collimation of the instrument one is simply adjusting the error from the instrument. In electronic instruments we are reprograming the instrument, 13 seconds = zero.?ÿ
I was taught that a TS collimation routine collimated the TS to whoever was using it. In other words, everyone points (sights) differently, therefore you're adjusting the TS to whoever is using it. When I was a party chief in charge of an FDOT project way back in 1994, the EFB software dictated a collimation routine at the start of each day. If I remember correctly, we had to sight an object at least 300' away, and the object had to be somewhat elevated. It was said that the EFB made corrections to observed angles based on the collimation routine results. I never made any inquiries to confirm that. What do you mean by 13 seconds = zero?
That brings us to the present. We are using a Topcon ES-105 TS. It's a 5" instrument, and it has an onboard collimation routine. I tried it out in the office this afternoon, but I need help with interpreting the results and with deciding when to keep or reject the correction based on the TS specifications.
I assume that V. Off. means the average of the vertical angles is 7". Would that be an acceptable tolerance for this instrument? I'm quite sure I could get better results by sighting a more distant and elevated object. I don't know what EL is. Any idea?
The ES series specifications appear above for your scrutiny.
Another old post concerning this is?ÿ
Total Station Check and Adjust routines
https://surveyorconnect.com/community/surveying-geomatics/total-station-check-and-adjust-routines/
I was actually in on the discussion! Different instrument manufacturers have different check and adjustment routines. The terminology can be confusing.
?ÿ
I was taught that a TS collimation routine collimated the TS to whoever was using it. In other words, everyone points (sights) differently, therefore you're adjusting the TS to whoever is using it. When I was a party chief in charge of an FDOT project way back in 1994, the EFB software dictated a collimation routine at the start of each day.
It's weird that it was explained like that, since collimation is a function of the instrument rather than the operator.
Collimation adjustments are done to account for the angular discrepancies between F1 and F2 pointings. If two individuals separately point at an object in F1 and then in F2, it doesn't matter that they point differently, their pointing is still going to be consistent between the two faces. The angular discrepancy will be within a second or two between each individual, making it a waste of time to run collimation adjustments every day - unless the instrument was really drifting out of whack every single day.
The (primary) purpose of a collimation check is to come up with an angular correction for F1-only measurements rather than to calibrate the instrument to any particular individual.
?ÿ
That brings us to the present. We are using a Topcon ES-105 TS. It's a 5" instrument, and it has an onboard collimation routine. I tried it out in the office this afternoon, but I need help with interpreting the results and with deciding when to keep or reject the correction based on the TS specifications...
I assume that V. Off. means the average of the vertical angles is 7". Would that be an acceptable tolerance for this instrument? I'm quite sure I could get better results by sighting a more distant and elevated object. I don't know what EL is. Any idea?
Those values are not inherently "bad" just because they exceed what you would think is a "good" accuracy when measuring.
They just show how much the instrument is "correcting" or modifying single-face measurements, not how accurate your measurements are.
I don't know the Topcon terminology, but it looks like F1 horizontal angles get a 1" correction while the vertical angles get a 7" correction. Presumably this means that during collimation the deltas between F1 and F2 were 2" H and 14" V. Not necessarily a problem, and not an indicator of "accuracy" either.
However, as others have said upthread, there is a limit to how much automatic adjustment an instrument can make and still meet manufacturer's specs. Most instruments will tell you when that limit is exceeded. For instance, I seem to remember that the Trimble S series can compensate for more than 20 minutes of discrepancy before needing service.
We run collimation/check & adjust routines monthly, or when the instrument has been cargoed somewhere, or right before precision work, or when we are getting odd results and want to make sure nothing has drifted out of whack.
But we don't do it daily for typical topography/mapping work. We definitely don't do it when switching operators - even for manual, non-robotic work.
?ÿ
The (primary) purpose of a collimation check is to come up with an angular correction for F1-only measurements rather than to calibrate the instrument to any particular individual.
Some 50 odd years ago when I first started, when we were observing primary control (with a T3) we used to swap observers after every pair of FL/FR just to remove the "observer bias".
How much bias did you typically find between observers??ÿ What about their method of observing or eyesight causes the bias?
Typically it was very small, perhaps a couple of seconds (which, I suppose, is quite a lot with?ÿ T3). Not a problem if it was consistent, but we did find that some surveyors had different biases looking "with the light" or "against the light". Where it did matter was on those set ups where the survey pillar was built into a wall or hedge, so you needed an observer on each side. Then you needed to swap over at intervals to remove that bias.
Edit: I should have added, always make the final adjustment to pointing to each target from the same direction.
It's weird that it was explained like that, since collimation is a function of the instrument rather than the operator.
Thanks for clarifying that.
They just show how much the instrument is "correcting" or modifying single-face measurements, not how accurate your measurements are.
Maybe how much the instrument will "correct", since the final step is either to apply the correction or not.
I don't know the Topcon terminology, but it looks like F1 horizontal angles get a 1" correction while the vertical angles get a 7" correction. Presumably this means that during collimation the deltas between F1 and F2 were 2" H and 14" V. Not necessarily a problem, and not an indicator of "accuracy" either.
It would be great if the Topcon manual explained their terminology. What would you consider an exceptable vertical delta for a 5" instrument? I can't find any data relative to that in the attached Topcon ES-105 specifications.
But we don't do it daily for typical topography/mapping work. We definitely don't do it when switching operators - even for manual, non-robotic work.
FDOT was very clear concerning a daily collimation (axis) test.
?ÿ
It's weird that it was explained like that, since collimation is a function of the instrument rather than the operator.
Thanks for clarifying that.
They just show how much the instrument is "correcting" or modifying single-face measurements, not how accurate your measurements are.
Maybe how much the instrument will "correct", since the final step is either to apply the correction or not.
I don't know the Topcon terminology, but it looks like F1 horizontal angles get a 1" correction while the vertical angles get a 7" correction. Presumably this means that during collimation the deltas between F1 and F2 were 2" H and 14" V. Not necessarily a problem, and not an indicator of "accuracy" either.
It would be great if the Topcon manual explained their terminology. What would you consider an exceptable vertical delta for a 5" instrument? I can't find any data relative to that in the attached Topcon ES-105 specifications.
But we don't do it daily for typical topography/mapping work. We definitely don't do it when switching operators - even for manual, non-robotic work.
FDOT was very clear concerning a daily collimation (axis) test.
?ÿ
If the operator chooses not to apply the collimation correction after properly performing the test, then the split between F1/F2 measurements is going to be larger than it should be, and F1-only measurements are going to have systematic error built into them. Maybe it??s significant for the work, maybe not.
Like I mentioned above, the only unacceptable vertical delta in the collimation correction is one that is too much to compensate for internally. Whether it??s 5 seconds or 5 minutes, if the instrument can apply the correction, you??ll get almost the exact same results after applying it. 1? or 5? instrument, doesn??t really matter. It's still critical to get enough F1/F2 observation pairs during collimation, so the instrument has enough to work with.
Maybe FDOT just wanted to be absolutely sure they wouldn??t have to deal with that particular problem. Spending extra time on unnecessary procedures is sort of a hallmark of public sector work. (Not to ding public sector, since spending less time on critical procedures seems to be the hallmark of private sector work.)
But if they were claiming it was to adjust the instrument to the operator, that??s just incorrect. Wouldn??t be the first time I have seen someone at the top insisting on a procedure for a totally bogus reason. Once had a PM that insisted we take about a dozen "Here" positions when setting up an autonomous base position. Totally pointless, but he was convinced that it was somehow "better" and could not be reasoned out of it.
?ÿ
One more thought ?? collimation error is why I have crews turn F1/F2 observations to the backsight on every single setup. If we start seeing the splits drift too much, it??s time to do a collimation. If we ignore the problem, we start seeing odd biases in our topo data, and our mean turned angles/sets will start failing angular tolerances during QC/QA.
?ÿ
One more thought ?? collimation error is why I have crews turn F1/F2 observations to the backsight on every single setup. If we start seeing the splits drift too much, it??s time to do a collimation.
Very sound advice!
Instruments have feelings too.
I just notice the instrument has an extra spring in it??s step after the check and adjust routines.
@rover83?ÿ
I found an interesting webpage concerning collimation.
https://www.seilergeo.com/total-station-collimation-report/#
What I find interesting about it is that the software generates collimation reports. I assume this level of concern would be for a higher order of surveying than what would commonly be used.
Also, there are recommendations for target distance and target deviation from a horizontal plane.
"Note ?? Do not use Autolock during Collimation or Horizontal tilt axis tests.
Test?ÿthe instrument as follows:
For both collimation tests ?? Instrument should be at least 100 m (~328 ft) from the target.
HA VA collimation ?? should be on a horizontal plane- target should be less than 5?ø higher or lower than the instrument.
HZ Tilt test ?? target should be at least 14?ø above or below?ÿthe horizontal plane.
You must take a minimum of one observation on each face. Seiler recommends three or more.
Final collimation values must be within 0.02 gon (= 0?ø01??05?) of standard values. If not, adjust the instrument mechanically." Send to?ÿa Seiler Service department.
I wish I knew how they came up with their final collimation values. What do they mean by "standard values?"
I believe standard values would be what the instrument would read if it were perfect. For example a total of 360 degrees when face 1 is added to face 2.
