There has been some extended discussion about the angular accuracy of a Topcon GTS 203 total station, built ca. 1994. In going back and looking over prior threads, I didn't see anyone actually mentioning what the manufacturer's specifications for the instrument are.
Does anyone know any of the following:
- is the horizontal circle absolute or incremental?
- does the horizontal circle have a single reading point or a diametrically-opposed pair?
- does the instrument have single or dual axis compensation?
- what is the manufacturer's claim for the accuracy of a horizontal angle measured with the intrument?
If this information appears somewhere on the web outside of a paywall, I've yet to find it, but all of it is relevant to the continuing discussion.
Here's what I got:
I'm almost certain that, as with other Topcon series, that the third digit in the model number (3, vs. 2 or 1), is decreasingly accurate.
That is, the gts 203 is a 20"/10" instrument vs. the 10"/5" 202 or the 5"/1" 201
That said, Here's how I read the specs:
.- is the horizontal circle absolute or incremental?
No idea. Any way to determine this with the instrument in front of you?
- does the horizontal circle have a single reading point or a diametrically-opposed pair?
Single reading point.
- does the instrument have single or dual axis compensation?
Single axis compensation
- what is the manufacturer's claim for the accuracy of a horizontal angle measured with the interment?
I leave this to your more knowledgable reading of the spec. My guess is that your determination in the other thread of 10" is correct, compared with the 201's 5", and the 202's 6"
Any of that help?
Extrapolating to your GTS-203, then, it looks as if it only has one circle reading point on the horizontal circle and uses an incremental reading system. That is, it measures angles by counting interference fringes detected by the circle reader as the circle is turned. I haven't thought in detail about the implications of that, but at first impression, I think it means that you don't want to reset the instrument when you measure Face Rt, but simply read the circle as it is, subtracting 180°00'00" from that direction when you figure the mean of Face Lt and Rt.
The one angle reading point (as opposed to having one on each side of the circle) means that it is very necessary to measure all angles as the mean of Face Left and Face Right directions in order to cancel circle eccentricity errors.
It looks as if the instrument has only single axis compensation, so directions on both faces are also definitely necessary to correct for tilting axis errors and leveling the instrument is critical.
The accuracy spec quoted for the GTS-202, i.e. 6", is quoted as being determined from the DIN 18723 test procedure. What that means is that the standard error of a direction taken as the mean of Face Left and Face Right is +/-6" or less and that the standard error of an angle determined from two such directions is therefore +/-8.5" or less.
If you have a GTS-203, presumably some other specification applies, however. My guess would be that the GTS-203 is basically a GTS-202 with possibly a less well machined center and different software settings to interpret the circle reading. If there was high demand for the GTS-203, the centers may be from one of the higher-grade lines such as the GTS-202.
> Extrapolating to your GTS-203, then, it looks as if it only has one circle reading point on the horizontal circle and uses an incremental reading system. That is, it measures angles by counting interference fringes detected by the circle reader as the circle is turned. I haven't thought in detail about the implications of that, but at first impression, I think it means that you don't want to reset the instrument when you measure Face Rt, but simply read the circle as it is, subtracting 180°00'00" from that direction when you figure the mean of Face Lt and Rt.
One easy test that comes to mind for the circle reading system would be to set an engineer's scale at 34.4 ft. from the instrument and at right angles to it. If you point the instrument at a division on the 50 scale (with graduations every 1/50 inch), and zero the instrument, does the instrument display an increment of 10" at every graduation? That is, after zeroing the instrument, turn it until the horizontal angle changes by 10" and observe the scale to see whether the instrument is exactly pointing at a graduation or not.
I'm not certain how the incremental reading system works on the GTS-203, but if it is generating a square wave signal with a period of 10" from the interference pattern between a grating on the circle reader and the circle, the angle that the instrument turns between 0-00-00 and 0-00-10 may actually be less than 10".
Would the ability to read out either degrees or gons imply that they need a finer resolution on the circle to be able to convert to either with the advertised accuracy? You might try a test like Kent proposed with readout in both systems and see where the least digit changes.
>
> One easy test that comes to mind for the circle reading system would be to set an engineer's scale at 34.4 ft. from the instrument and at right angles to it. If you point the instrument at a division on the 50 scale (with graduations every 1/50 inch), and zero the instrument, does the instrument display an increment of 10" at every graduation? That is, after zeroing the instrument, turn it until the horizontal angle changes by 10" and observe the scale to see whether the instrument is exactly pointing at a graduation or not.
I haven't even finished round three of the other test (five targets at 150' plus four at 10', for input to Starnet. I'll throw this on the list.
I'm still not sure my GTS 203 will measure in Gons. There may be a parameter setting somewhere but I haven't found it yet. I'll keep looking.