I've been reading up on nodal prism assemblies (I'll call them that to, distinguish from just a prism without a 2-axis pivoting mount) on these forums and elsewhere, and I think I understand the implications. Not meaning to beat a dead horse here, but I can't help but think : why aren't prism holders (the rotate+tilt variety) *all* nodal ? It seems to me there is no disadvantage, and eliminates any (significant) error if the prism isn't perfectly lined up with the station's line of sight.
?ÿ
I mean, look at this one. The prism's "node" is waaay beside the rotation axis. And it's not like it would've been hard to solve : just make a "deeper" holding bracket that moves the prism back a few mm. I just don't get it, I must be overlooking something.
?ÿ
I understand those few mm are meaningless in most surveying jobs given the other sources of error, but still...
Thanks for any insight !?ÿ
?ÿ
Careful study of your picture shows the nodal point of the prism on line with the prism bracket center, but the target above the prism is not aligned with either.
Paul in PA
"Because that's the way we've always done it..."
My theory, which is just speculation, is that they used 0mm offsets first because - no correction! It made things simpler.
But that leaves the prism hanging way out in front and off balance, so someone reversed it and got the -30mm offset to make the assembly more compact. Problem solved.
It seemed to be quite a long time before anyone ever thought of the nodal prism. I vaguely remember seeing a paper years ago complaining that surveyors were doing the prism thing wrong. Maybe no one noticed until then?
Aside from my 360 prism for the robot, nodals are the the only ones I use today.
It's nice to have a discussion about prisms. Sometimes it seems like everybody here does that magic sky surveying thing... ?????ÿ
The root of the issue is that light travels slower in the glass than in air. So the effective point of reflection is around 30mm (Leica a little more) behind the physical rear point of the glass on the larger prisms.
If you mount it so the effective reflection point is over the pivot and over the point to be measured (zero offset) then all is fine for shots dead on the frontal direction. But if the prism is pointed somewhat off direction, there is an error.
Mounting the glass so its physical point is over the pivot gives consistent readings despite the pointing of the prism being off somewhat.?ÿ
The price of this benefit is that the effective reflection point is now 30 mm or so behind the point you want to measure so you have to crank in an offset correction.
I wish someone would provide a definition of "nodal" as I only guess from context.
Either way, the prism assembly needs to be pointed at the instrument.
After I spent years collecting all the components that I wanted to have to have identical tribrachs on the instrument and for all sites on tripods, they come out with prismless capabilities.
In recent years I use 40mm x 40mm reflective card targets for all kinds of things and they stick to anything and have 0mm offset.
These are what I wish I had more of
Thanks for the replies so far!
your picture shows the nodal point of the prism on line with the prism bracket center, but the target above the prism is not aligned with either.
I'm not sure I follow - isn't the nodal point the visible intersection of the lines within the prism ?
?ÿ
It seemed to be quite a long time before anyone ever thought of the nodal prism
that's what bothers me most. Clever people have developped and used total stations and prisms for *decades* , manufacturers with world-class optical heritage and expertise, and nobody ever thought of this until "recently" ?
?ÿ
I understand the 0mm offset thing as a convenience when calculating was expensive. But it seems now we need to punch in offsets and corrections anyway, so typing in 40 instead of 30 shouldn't be very taxing !
In recent years I use 40mm x 40mm reflective card targets
Cool, would those work with an old-school Wild TC1000 ? (in early testing I was shocked to see that its EDM doesn't work on a regular mirror, unless the alignment is way more critical than I thought ? or my unit is defective )
Cool, would those work with an old-school Wild TC1000 ? (in early testing I was shocked to see that its EDM doesn't work on a regular mirror, unless the alignment is way more critical than I thought ? or my unit is defective )
A flat mirror reflects with angle of incidence equal angle of reflection, so if you are 1 degree from perpendicular, the beam comes back 2 degrees away from you.
I suspect those targets would work for you.?ÿ My old-school Topcon works well to some brands of 3-inch red "driveway reflectors" with the cube-corner pattern in them, and flat reflective sheets that have a similar pattern, with a range up to 100 yards/meters. Hillman brand works best for me, a competitor not quite as well.?ÿ It does not work on the 1-inch reflectors commonly found in the store because their pattern is not cube-corner, nor on the fine-grain reflector tape.
The offset of a nodal prism is dependent upon the diameter of the prism. So if you're using different sizes of nodal prisms (e.g. a 62-mm diameter prism on your backsight and a 25-mm diameter mini-prism on your foresight), then either you or your data collector need to keep track of the different offsets. Not at all impossible to do, but it's yet another thing of which you have to be aware. There's a comforting simplicity if all your prisms have a 0 mm offset, if you're willing to take care in aiming (e.g. using targets that have collimator sights) ?? or if they all have a -30 mm offset, if you're less willing to take care in aiming. I assume that the common -30 mm offset is a convenient compromise across prisms of different diameters, to reduce the sensitivities to aiming errors while not forcing the user to keep track of multiple offsets.
Maybe this is wrong but I have assumed that Leica chose -34.4mm because that is effectively nodal but most of the other manufacturers weren't quite as Swissly obsessive-compulsive so they chose -30 to have a nice round number.
I am OK with obsessive-compulsive behavior when it comes to precision measurement...and Leica does well in this regard...
...but I do take issue with a product that claims -34.4mm = 0 mm. With all the possible flubs and mis-entered numbers that could happen in the field, it was (and remains) a terrible idea to label prisms with offsets that are literally incorrect.
The first time I did a small traverse with a Leica instrument of any kind the closure was terrible.?ÿ I heard how great Leica was and this is the best it can do? What? What happened?
I took over the setup from some guy that recently retired, he had ancient AGA zero offset traverse targets. The Leica salesman told him to just put zero offset in the instrument! They surveyed like that for a few years.?ÿ Someone on the old RPLS forum told me about the Leica minus 34 point 4 mm error, Wow, couldn't believe it.
After I corrected the data the traverse closed great as expected.
Now I'm using a Trimble S7 with a TSC3 with Access and you put into the controller whatever the actual offset is, what an idea! The thing with Leica prisms you have to be careful because the number printed by Leica on the prism is wrong, subtract 34.4mm from that number to get the correct number.
So if you're using different sizes of nodal prisms (e.g. a 62-mm diameter prism on your backsight and a 25-mm diameter mini-prism on your foresight)
Ah ! That's a good reason, I hadn't considered that.
I think my TC1000 also uses the unholy Leica convention of "-34.4 = 0"?ÿ ??ÿ I did some tests on a calibration range with a Seco -30mm mount, entered "-30mm" as correction in the instrument, and surprise, my measurements are off by at least 34mm + some ppm (around 20, I blame user error and my half-assed atmospheric correction )
In short, good practice is not to mix accessories: keeping to one manufacturer minimises errors down to those you make yourself. Surely that's enough errors for anybody.
Depending on the instrument/software combination, a double offset may occur.
At a previous employer, we had some older Leica TC1101 instruments which applied a prism constant in the instrument and then sent the result to the DC. As our DCs were Trimble, they assumed that the raw data was unadjusted and applied an additional constant per the project/setup settings.
And as we had projects that required us to book every traverse measurement, it was not uncommon for crews to ditch the DC and change the prism constant in the instrument when traversing...only to have the next crew use the DC without checking the instrument settings first.
Also, depending on the software, the prism offset may be assumed to be negative, i.e. if you enter "30" it computes a -30 offset.
I guess it's a good argument for using a complete TS package from a single manufacturer...
