Instrument Constant question

> I'm checking the instrument constant of my ancient TS (AC+BC-AB).
>
> AC is 139.035
> AB is 263.344
> BC is 124.308
> (All three measured 6 x 3(instrument averages)for a total of 18 measurements each.
>
> That = -.0010'
>
> The Instrument Constant set in the instrument is -31.1mm. That's also written on the sticker inside the battery compartment. The Prism constant is currently set to -30m.
>
> -.0010', which is -.3048 mm certainly seems fine to me, but the book says:
>
> "If using above procedure and no difference is found from the instrument at the factory or a difference of over 5mm is found, contact your Topcon dealer".
>
> Difference between what and what? Confused.

I truly doubt your EDM is good to 0.3 mm. It looks like you could increase your error 15 fold and still not take it to the dealer.

It just came back from the dealer lol!

The real concern I have is whether, with the settings I'm using, the instrument is operating within tolerance, or whether all my measurements have been off by 30mm. I've re-read the section of the manual for distance measurement and it says to set the prism constant value to zero. I think I've been screwing up all my distance measurements by the prism constant.

What the heck is an Instrument Constant, anyway?

The instrument constant is the extra distance measured from the instrument axis to the measuring device. While it may look similar to the prism constant it does not offset the prism constant. The prism constant is applicable to the prism way out there and the instrument constant is applicable to inside the instrument, here. You need both.

Prove it to yourself, tape out two points, set up on one and measure with the EDM to the other. If you have prisms that can be changed to zero offset, do that without changing the prism constant in the instrument and note the differences in distance. Then change the prism constant and remeasure. Testing is the best way to assure yourself you have it right.

Paul in PA

It depends on the instrument. With Topcon, I believe there is a correctable instrument error from calibration independent of the prism correction. Why this is near the prism constant, I can't say. My Topcon GTS4 has the number written inside the screw-on lid on top of the telescope, about 8, if I remember.

The measurement process calculates AC+e + BC+e - AB+e = 0. Solve for e, which is the measurement error incorporated in each measurement due to the prism offset and other errors. The other errors are supposed to be corrected by instrument calibration.

If you are using a Topcon instrument with a -30mm prism, setting the prism offset in the instrument to 0 and doing your test, the calculated error should come up about +30mm (I think). You would then set the prism constant to minus the calculated error, probably around -30mm. If you set the prism constant to -30mm, or the calculated correction and make the measurements, you should come up around 0, which you did.

Looks to me like you are good to go.

I don't have wide experience, but I thought that Leica was the only one to hide the constants and pretend they are zero. My old old Topcon needs to be set to -30 for common prisms.

How many prisms did you use in your test, and if they are reconfigurable, do they appear to be set in the nearer or further position? Misinterpreting the prisms could confuse the results.

You might try one of those red reflectors that look like a taillight, that people set beside their driveway to help find it at night. Some brands are good past a hundred yards with my old instrument. They will have a very small prism constant, and give you a way to determine by comparison whether your prisms are set to 0 or -30 position.

The classic test is set up three tripods/tribrachs (A, B & C) in line with A to B being about 50' and B to C about 50' and B on line between A and C.

Put the instrument in tribrach A and prism targets in B and C.

Measure the distance to B and C.

Put the instrument in Tribrach B and prism targets in A and C and measure the distance to A and C.

A-B + B-C should equal A-C. If not then you have your prism constant is Not set correctly.

> I'm checking the instrument constant of my ancient TS (AC+BC-AB).
>
> AC is 139.035
> CB is 124.308
> AB is 263.344

> (All three measured 6 x 3(instrument averages)for a total of 18 measurements each.
>
> That = -.0010'
>
> The Instrument Constant set in the instrument is -31.1mm. That's also written on the sticker inside the battery compartment. The Prism constant is currently set to -30mm.
>
> -.0010', which is -.3048 mm certainly seems fine to me, but I think I've read that if you're using Topcon prisms (I am) with a Topcon TS, the prism constant would be set to zero in the instrument.

Why not just examine the prisms to see whether they are -30mm offset prisms or not? It should be marked on the prism assembly somewhere. Otherwise, take a closeup photo of the prism from which it can be identified and post it. 0mm offset prisms have the nodal point of the prism forward of the centerline of the prism assembly mount which a sideview should show well enough.

The Standard ABC test is incapable of determining the instrument constant unless you use a correctly identified prism. It is also imperative to use the three tribrachs to get the most benefit. You all but eliminate centering errors...

That seems too high for an instrument constant. Perhaps someone entered the prism offset in the wrong place? If you are using a prism that is -30 then the prism offset in the topcon gun should be -30 (nikon +30) unless you have the controller(dc) offset the difference. If in doubt as suggested above tape distance between 2 points on flat surface such as a parking lot and check it out.

We laid out a building a few years ago. I was showing the crew a few tricks. One of them suggested we forgo worrying about tension, sag and plumb bobs. I humored him and laid the tape on the asphalt. The 100 foot building suddenly appeared to be 99.84 feet...

Depending on where you are you might want to use an NGS calibration baseline. It will provide an accurate value for comparison. If you make observations of all the points you can also calculate and/or verify your instrument's performance. There is a nice little DOS program, CALIBRAT, on the NGS to perform the reductions.

Identifying a wrong prism constant would seem straightforward given a standard for comparison.

NGS site for calibration baselines is: http://geodesy.noaa.gov/CBLINES/calibration.html

> It depends on the instrument. With Topcon, I believe there is a correctable instrument error from calibration independent of the prism correction. Why this is near the prism constant, I can't say. My Topcon GTS4 has the number written inside the screw-on lid on top of the telescope, about 8, if I remember.
>
> The measurement process calculates AC+e + BC+e - AB+e = 0. Solve for e, which is the measurement error incorporated in each measurement due to the prism offset and other errors. The other errors are supposed to be corrected by instrument calibration.
>
> If you are using a Topcon instrument with a -30mm prism, setting the prism offset in the instrument to 0 and doing your test, the calculated error should come up about +30mm (I think). You would then set the prism constant to minus the calculated error, probably around -30mm. If you set the prism constant to -30mm, or the calculated correction and make the measurements, you should come up around 0, which you did.
>
> Looks to me like you are good to go.

What was intended to be an Instrument Constant thread is turning into another Prism Constant thread, lol.

I put all three of my prisms side by side with a reflector with near zero offset. One of the prisms is clearly marked -30, One,(a Nikon) says "30 mm offset", but it's similar to the other one that says -30; both have sliding tubes, and both have the nodal point almost over the center of the tribrach. So too with the third, a Leitz.

Bottom line...All three measure 30mm+/- MORE compared to the flat reflector, and MORE than the tape measured distance. With the prism constant set to -30mm, they all measure the correct distance (as measured with the tape).

So, now that that's all said, I believe I am "good to go", confirmed by actual careful (prisms in tribrachs) measurement.

Paul in PA is correct

Years ago, the early EDMs had a variable instrument constant from one time turning it on to another. NGS discovered this and developed a lens cover with two holes in it and expoxied two right-angle prisms so that there was a constant distance from transmitter LED to receiver. Based on the difference between the known constant of the cover and what the instrument gave as a distance, that was the instrument constant for that period of time until it was turned off. Next time the instrument was turned on, the same procedure was repeated to determine the "new" instrument constant.

It had absolutely nothing to do with the retro-reflector offset constant distance.

If you have a prism with a known offset (normally 0, -30mm or -40mm), set the instrument prism offset at that same value, have the correct temp and baro pressure set and the Instrument Constant is correct, e will be 0mm. If Prism manufacturers generally have a pretty good record of placing their prisms correctly with respect to the plumbing point.

I think the point of the manual instructions is that if you have the prism offset in the instrument set correctly for the offset of the prism you are using, you should contact the dealer if the calculated error e is greater than +/-5mm. The difference is from 0.

There is a variable factor which I'm pretty sure is part of the the Instrument Constant and that is the cyclic error which depends on the distance between the instrument and the prism. The distance error looks like a sine wave as the distance increases. When the instrument is put on a calibration rail, the error is measured over a range of distances and the proper correction is applied to balance the maximum plus and minus errors so that there is minimum error at any distance.

> Depending on where you are you might want to use an NGS calibration baseline. It will provide an accurate value for comparison. If you make observations of all the points you can also calculate and/or verify your instrument's performance. There is a nice little DOS program, CALIBRAT, on the NGS to perform the reductions.
>
> Identifying a wrong prism constant would seem straightforward given a standard for comparison.
>
> NGS site for calibration baselines is: http://geodesy.noaa.gov/CBLINES/calibration.htmlbr >
That's cool. There's one about 30 miles from me. I'm going to give it a spin. Thanks.

Yes, Paul is indeed correct

And if I remember correctly this is also the same reason the original Wild/Leica prisms were designed to have a constant of -34.4mm - because this directly offset the EDM constant.

Thus all of the ensuing confusion over the years because Leica calls -34.4 a 0mm offset.

Sort of the exact same thinking behind their tilting prism concept (the one where the prism was mounted above the horizontal tilting axis. Because the EDM was mounted the same way. Really, both are beautiful in their simplicity.

With newer EDM technology and better distance measurement I wouldn't think Leica would still strive to construct a total station with the EDM at a 34.4 offset internally, but rather compensate with a derived constant (the same as the other manufacturers).

Messing with the EDM constant requires more consideration than prism constant. There have been several discussions about verifying prism constant rather quickly and absolutely on this board. Investigating whether or not your EDM constant is correct should include a visit to a CBL, taking an atmospheric gauge and thermometer with you, and truthfully considering how accurately you can center over a point. At least this is how I would do it. If you're not chasing that last half millimeter for critical projects, you probably don't need to worry about your EDM constant at all - especially if it just came back from the shop.

Hello. I use a Pentax r425vdn and it has written inside the offset 20.2 mm?ÿ What does it mean exactly ? Thank you?ÿ?ÿ

http://www.pentaxsurveying.com/en/pdfs/R400V-MAN-BASIC-EN.pdf

A reasonably robust method to determine the combined constants for instrument and prism is to set up and measure among four points, each about 15 or 20 meters apart, on a straight line and as closely as possible at the same elevation. The objective is to minimize the differences in height between prism centers and instrument horizontal axes.

Given four points
... A_____B_____C_____D,
measure [AB], [AC] and [AD]
... the square brackets [ ] indicate the [displayed] distance values
then measure
... [BA], [BC] and [BD]
... [CA], [CB] and [CD]
... [DA], [DB] and [DC]

note that each [displayed] value includes, but does not disclose, the combined constant "k" so:
... [AB] = (AB + k), [AC] = (AC + k) and [AD] = (AD + k)
where AB, AC and AD are the respective actual distances and, in theory, AB = BA,
AC = CA, AD = DA, etc.

Calculations, once all distances have been measured are:
AB = (AB + k)
BC = (AC + k) - (AB + k) = [AC] - [AB] ....... I
CD = (AD + k) - (AC + k) = [AD] - [AC] ...... II
BD = (AD + k) - (AB + k) = [AD] - [AB] ..... III

BA = (BA + k)
BC = (BC + k)
AC = (BA + k) + (BC + k) = BA + BC + 2k ...... IV
BD = (BD + k)
AD = (BA + k) + (BD + k) = BA + BD + 2k ....... V
CD = (BD + k) - (BC + k) = [BD] - [BC] ....... VI

... repeat for measurements from C and D and note that where there are differences, the actual distances between points are the differences in displayed distances. Conversely, where the distances are additive, the differences are 2k. That is, when distances are differences, the "k's" cancel out and when additive, the "k's" are doubled.

The process yields 12 total equations that allow solutions for "k." Not all values will be equal but, the average should be reasonably and sufficiently valid. For those so inclined, the equations may be solved using least squares that in turn result in a single value for "k" along with residuals and standard deviation, etc. For more redundancy, the method may include five or more points instead of just four.