Newsgroup sci.engr.surveying 3634

Directory

     The doubling-angles thread got me thinking about another instrument
practice.  In different regions it may be called repping, rapping, or
wrapping.  I refer to the practice of reading one angle turn, carrying
that same angle to the backsight, and adding subsequent turns to it.  I
question the continued use of this technique. 
     It is easy to see how it got started.  A transit with external scales
might have a least count of 30", but it was possible to point the
instrument with greater precision than that.  Turning six reps effectively
divides the least count by six, making it 5".  That is not true of modern
instruments.  The scales can be read with greater precision (not
necessarily accuracy) than the instrument's pointing. 
     Some electronic theodolites simulate this technique by carrying the
angle reading to the backsight.  With few exceptions, they do not carry
the plate itself, only the number.  By adding a constant to both backsight
and foresight, we are not gaining anything.  The end result is the same as
the mean of independent readings. 
     At the insistence of others, and at the expense of efficiency, I have
repped angles.  When more than one foresight is being observed, a new set
of angles must be started for each of them.  A single angle in the middle
of the set may deviate greatly from the rest, but will not be noticed if
the sum is within tolerance.  I have been able to find only one advantage. 
Some gunners think that the angle spread is a measure of their ability. 
Making the math slightly more difficult makes it less likely that they
will cheat. 
     Any comments?
     Kunkel

Jim Frame wrote:
> 
> Anyone know where I might find information about geodetic control
> stations in Costa Rica?
> 
> ---------------------------------------------------------------------
> Jim Frame   jhframe@dcn.davis.ca.us    (916) 756-8584  756-8201 (FAX)
> Frame Surveying & Mapping         609 A Street        Davis, CA 95616
> -----------------------< Davis Community Network >-------------------
You might try the Defense Mapping Agency (recently incorporated into the
new National Imagery & Mapping Agency) - they have done extensive work
throughout Central & South America.  They are at "www.nima.gov".
Good Luck,
Marty Shurr
North Pole, Alaska

Paul Kunkel  wrote in article
...
>      The doubling-angles thread got me thinking about another instrument
> practice.  In different regions it may be called repping, rapping, or
> wrapping.  I refer to the practice of reading one angle turn, carrying
> that same angle to the backsight, and adding subsequent turns to it.  I
> question the continued use of this technique. .......... 
	Your logic behind "repping" angles is lacking one important issue!
Don't forget that you must invert the telescope upon each repetition of an
angle.
This practice will also help average out the built in mechanical error in
even the most modern of total stations!
Eric

"Neil Giesbrecht"  wrote:
>With modern instruments....
>[...]
> Does the circuitry of these new guns make such a routine[doubling]
> unnecessary?
I have the double angle discussion with a fellow land
surveyor at least once in any given year.  My philosophy on
doubling has nothing to do with accuracy, as a well adjusted
and operated instrument will provide adequate precision and
accuracy for the majority of general surveys performed.
The effects of multiple angles on accuracy becomes academic
for most applications as error ratios of 50000 or greater
are routinely achieved on traverses with many legs and
perimeters less that 10000 feet.
The real reason to double angles is to provide an on the
spot check for measurement or recording errors.  Instrument
precision will never eliminate the potential for human
errors and blunders.  Taking multiple observations is the
only way to identify and minimize human error, as well as
identify an instrument that has gone out of adjustment (gee,
I had that total station adjusted last year).
There is not one good reason for not turning multiple angles
on traverse legs and other significant observations.  The
"time is money" argument never washes with me.  One busted
angle not identified on the spot in the field will destroy
the economic viability of any survey, and can ruin the
ability to meet tight deadlines, when time really is money.
The time savings is never worth the risk.
--
Bryan Bunch
bwbunch@skn.net
--

In article <56l715$44p@grissom.powerup.com.au>, dine@powerup.com.au (Peter & Carmel Spierings) says:
>
>The use of Least Squares in adjusting traverses is quite meaningless as
>the redundancies in a traverse are very small in number. The use of
>Least Squares assumes that, apart from small random errors in the 
>measurements, the number of redundancies is significantly large.
>Nonetheless, many surveyors keep on using least squares to adjust
>traverses. Bowditch would be much more preferable.
>
From a Texas surveyor and Least Squares zealot:
In a typical closed traverse there are at least three additional measurements
(two angles and one distance) beyond what is minimally necessary to calculate
the co-ordinates of points positioned by the traverse.  Plainly the fact that 
there is a need to adjust the measurements at all means that some redundancy
exists in the system of survey measurements.
The use of the method of Least Squares simply requires that there be something
to adjust, one more measurement than is needed to determine the figure of the
survey, and some realistic estimates of the standard errors of the survey
measurements.  I assume that there is no problem estimating the standard errors
of the measurements made in the course of a project, even if the manufacturer's
specifications are in doubt, since widely accepted procedures such as those of
DIN 18723 are available for that purpose.
In the back numbers of the British publication, "Survey Review", there is an 
article upon the relative inadequacy of Bowditch's Rule (oddly known as the
Compass Rule in the U.S. although Nathaniel Bowditch was American), P. Berton 
Jones' "A comparison of the precision of traverses adjusted by Bowditch Rule 
and by Least Squares" (No.164, April, 1972).
Am I right in thinking that you used Least Squares in adjusting the deformation
and control surveys you mention?  Is your aversion to Least Squares based in
part upon having used very awkward software?  If so, I can certainly sympathize.
Really though, not having a decent computer and software is the only reason I can
think of for not using Least Squares adjustments on every project.  Good programs
like Star*Net and Star*Lev are very easy to use and have very short run times,
unlike some old mainframe programs that were not and did not.
Kent McMillan, RPLS
Austin TX