Any suggestion on how to include back sight data in an adjustment. 2-1-2 just does not cut it
> Any suggestion on how to include back sight data in an adjustment. 2-1-2 just does not cut it
Sure. Use a DV line for the backsight measurements, either SDist Zang HI/HT or HDist DeltaH HI/HT
Example:
[pre]
.Units FeetUS
.Units DMS
.Order FromAtTo
.Sep -
.3D
.DELTA OFF
# At Spike 12
DV 12-13 191.5440 88-08-17.00 5.315/4.945 'SPIKE.WASHER
M 13-12-137 240-53-20.50 113.9440 88-53-03.00 5.315/4.945 'LO.BT
[/pre]
Note that the above measurements were right out of the data collector, hence the extravagant precision of expression.
[sarcasm]I don't know how well you measure, but your data collector sure does a top notch job.[/sarcasm]
> [sarcasm]I don't know how well you measure, but your data collector sure does a top notch job.[/sarcasm]
Yes, it would be quite a trick to measure an angle to any bearing tree with an apparent uncertainty of better than 0.5 arc seconds, let alone one 19 inches in diameter and less than 114 ft. distant. :>
> [sarcasm]I don't know how well you measure, but your data collector sure does a top notch job.[/sarcasm]
At the same time the raw data is showing angles to the hundreth of a second, elsewhere in the programs settings, it is being told that the standard error of the angle is +/-3 seconds, 5 seconds, 30 seconds, or whatever is appropriate. Same for distances and every other element of the data.
> At the same time the raw data is showing angles to the hundreth of a second, elsewhere in the programs settings, it is being told that the standard error of the angle is +/-3 seconds, 5 seconds, 30 seconds, or whatever is appropriate. Same for distances and every other element of the data.
Yes, I should have pointed that out. The values logged by the data collector were simply estimates based upon the raw measurements. Those values are weighted in the adjustment by the standard errors specified by the user to characterize their uncertainties.
The program concludes nothing from the fact that the mean of two angles had a 0.50" fraction, nor that the mean of two distances output by the total station to the nearest 0.001 ft. could have a 0.0005 ft. fraction. The superfluous precision was completely harmless.
> 2-1-2 just does not cut it
You can use the M data type, but use a question mark for the angle.
Example:
[code=html]
M 2-1-2 ? 100.00 90.0550 5.12/4.88 'CONTROL
> > 2-1-2 just does not cut it
>
> You can use the M data type, but use a question mark for the angle.
I'd expect that the various conversions to Star*Net format from the different DC formats would convert a backsight record to a DV line, though. At least, that's how the Star*SDR converter handles it.
IT SEEMS THAT SOMETIMES THE BACKSIGHT IS REPLACED BUT SOMETIMES NOT???? i'LL HAVE TO LOOK MORE CAREFULLY,BUT I THINK SOMETIMES THE REPLacement is an SS record. So I have been using DV.... I have tried to use an M record AND THE SOFTWARE IS JUST NOT HAPPY WITH 2-1-2 359-59-59.......
THANKS
> IT SEEMS THAT SOMETIMES THE BACKSIGHT IS REPLACED BUT SOMETIMES NOT???? i'LL HAVE TO LOOK MORE CAREFULLY,BUT I THINK SOMETIMES THE REPLacement is an SS record.
When I want to renew the backsight from the same setup, I do it as a new occupation of the same station at the same HI. That way, all subsequent directions from that point in the DC file will be reduced to the directions taken to that renewed BS.
Otherwise, I'd expect that the backsight checks will be AVERAGED and the average value used to reduce the angles taken to objects from the setup. That may be fine for many cases.
> I'd expect that the various conversions to Star*Net format from the different DC formats would convert a backsight record to a DV line, though. At least, that's how the Star*SDR converter handles it.
Same with the Star*TDS converter... At least MicroSurvey's version of it. I always had problems with StarPlus' version. Never worked right unless you edited the raw file to only include terrestrial data, and even at that, it still had it's hiccups.
> Same with the Star*TDS converter... At least MicroSurvey's version of it. I always had problems with StarPlus' version. Never worked right unless you edited the raw file to only include terrestrial data, and even at that, it still had it's hiccups.
As I recall from conversations with Ron Sawyer, who originally developed Star*Net, the problem that they had was getting information out of the various vendors with their own data collection formats. Since that time, the various formats have probably settled into more or less static entities, I'd suppose.
I'm pretty sure that Michael Sage, who worked with Ron on the various converters and other programming projects, was also working on the various revisions/alterations that MicroSurvey impressed upon Star*Net.
In order to use the M record you need to replace the BS horizontal angle value with a question mark, as in my example above.
oh thanks Jim, I thought you were making a joke. The converter changes the back sight check into a DV record as a standard default. But I frequently shoot and record shots on the back sight when I traverse and these are converted to SS records. So I guess if I use the back sight check routine that the converter will turn the record into a DV .
> The converter changes the back sight check into a DV record as a standard default. But I frequently shoot and record shots on the back sight when I traverse and these are converted to SS records. So I guess if I use the back sight check routine that the converter will turn the record into a DV .
One practice I follow is to manually edit the Star*Net file to thin out the population of repeat distance measurements to the same prism from the same setup. Basically, all you're measuring is the repeatability of the EDM, which is maybe good for two or three at most. If you don't do that, the adjustment will treat all of the backsight checks as independent measurements from fresh instrument and target setups and, since they actually all have the identical centering errors since they are to the same prism from the identical instrument setup, the distance will be treated as better measured than it really was.
GOOD POINT kENT
Good point, Typically, when I traverse I might re set-up over the same pair of points., checking to see if I am still level and still over the points and most of the time this means tweeking the level at the TS....
