Encountered one of those quirks that has happened very few times over the years. What is scary is how easy it would have been to have introduced an error of this magnitude.
Simple total station work traversing along a twisty, narrow county road. I had just put the rod man on a predetermined point and had him set a temporary nail. Let's say that distance was 290 feet. I also wanted to know the apparent location of the center line of the roadbed at basically the same distance from where I was at. He sidestepped roughly eight feet to his right. I took a shot. Very close to 322.8 feet. Not once, but several times. I told him to hold on while I tried something. I rotated a few degrees to the left and then to the right and then back to prism. Every reading was now very close to 290 feet, as it should have been. Somehow the internal brain of the total station was adding 10 meters to the distance.
Y'all be careful out there, ya hear?
this sounds similar to instructions for setting up an edm calibrated base line. something to the effect of; if it is not possible to set markers at the recommended
distances of 0, 150, 400, 1200 meters (i think those were the quoted distances), then set the marker at ten meter intervals nearer or farther than the recommended distance.
the idea is something about edm phase or phase shift. maybe someone else can elaborate on this.
as an aside, we had an edm that would drop 10 arc minutes on its vertical circle, following an abrupt change to cold weather. we would have to stop work, and run the instrument into our check and adjust/field calibrations. lose about 15 minutes of work, and all is good again
Um, That's like when the clock strikes 13 (or 25 for you military guys) time to fix the clock!
N
Mr. Cow, I'm curious what instrument you're using?. I've experienced similar type errors some years back using a Topcon 301. Once we had a 1,000 meter error that kept repeating on a shot across a canyon. We could take the instrument down, turn it off and set it back up and it would still give the same error. Finally brought in a different instrument and it shot the right distance. The other time was on a traverse through the woods where we got a 1 meter error on an approximately 300' shot. It also repeated itself in that it showed up on both the direct and indirect measurements. We discovered it when we did error analysis of the network and re-ran the suspect legs of traverse.
I've never had it happen on any other instrument to my knowledge, so I'm curious as to the instrument you experienced it on. By the way, the errors with the 301 were an anomoly as we used that instrument a lot over many years without any other instances. Our questions to Topcon about what happened were something to the effect about the phase shift and it be a very rare possibility.
Holy Cow, post: 342799, member: 50 wrote: Encountered one of those quirks that has happened very few times over the years. What is scary is how easy it would have been to have introduced an error of this magnitude.
Simple total station work traversing along a twisty, narrow county road. I had just put the rod man on a predetermined point and had him set a temporary nail. Let's say that distance was 290 feet. I also wanted to know the apparent location of the center line of the roadbed at basically the same distance from where I was at. He sidestepped roughly eight feet to his right. I took a shot. Very close to 322.8 feet. Not once, but several times. I told him to hold on while I tried something. I rotated a few degrees to the left and then to the right and then back to prism. Every reading was now very close to 290 feet, as it should have been. Somehow the internal brain of the total station was adding 10 meters to the distance.
Y'all be careful out there, ya hear?
It's a long time since I've seen this. On some instruments the software controlling the tracking used to save time by not reading the full set of phase differences, but only the 10 metre banding. This generally worked OK but occasionally enabled a 10 metre slip to occur. Once the first one occurred then often the instrument would remain convinced that it was at known distance and would continue to revise only the metre/centimetre part of the reading. Switching to "Standard measure" and then back to tracking reset the system and measurements returned to normal. If the problem could be identified then amending the distance by 10 metres used generally to give the correct result.
(NOTE: this isn't the same as wrong readings caused by getting two sets of reflections back - one from the prism and one from either wet leaves or from a window reflection close to the sightline.
The Geodieter 4400 units had an interesting variation, in that they would sometimes produce distances in the 97 kilometre range, or -3200 metre range. These are obvious to see and I learnt that when it occurred then, before resolving it, the "wrong" observation should be recorded, the system reset and the same point taken with the correct value. This then gave an "adjustment" figure which worked and could be used when you later found a few other funny values in the same obs. set.
This was an old Sokkia Set 2C that has been maintained routinely through the years. The same machine did this about 20 years ago in a case where the true distance had to be very near 30 feet but the number shown was closer to 62 feet. That was such an obvious error that it startled the operator. He was ready to toss it off a cliff if it couldn't produce the right number. The initial thought was that it was doubling the distance, but, that would not be consistent with how the measuring is done.
My purpose in posting this was to alert all users of all "black boxes" to the fact that glitches can occur at any time. Our work practices need to be such that these glitches get caught and corrected ASAP. There was a discussion here yesterday about outliers and justifiying their exclusion. The key to that is knowing that it is an outlier, by some method, any method. Pulling the trigger on gathering data does not mean that data is accurate.
I gotcha. Kinda today's dropped chain.
Reminds me of the 7 P's:
Proper planning and procedure prevents piss poor performance.
That interests me because I think I've experienced something similar with that type of equipment.
I was involved in a large interchange survey in eastern Oklahoma. It was an old cloverleaf design and had truck stops on each of the four corners. The truck traffic was brutal. We had an older Set at the time, I believe.
When we would look at our points up on a screen, there was definitely a gremlin at work. About one or two in every hundred shots would be "out of place". Eventually the consensus was random back-scatter returns from the reflective markers on the sides of the trailers that kept whizzing past. Although it seemed possible (we found out early in our EDM career that you could shoot a distance to a round 2" reflective highway delineator) I never bought the trailer reflector idea. If it was reflectors from the trailers, the bogey shots should have all wound up inside paving limits where the trailers were at. Most of them didn't.
It was a drafting tech that noticed the common denominator; they all seemed to either be about thirty something feet off, or multiples thereof. This idea was actually discarded at the time because it always seemed to be "a little over" thirty feet. We were looking for an exact repeated distance error and never could arrive at one. We were doing the job in feet...Maybe we were missing the "meter connection"...like not seeing the forest for the trees.
things that make you go hmmmmm...
Saturday I was setting a point with reflectorless just 30 feet away, and since I am solo I had to walk over to the rod several times, always getting closer. When I shot the last time I was suddenly off by 4 feet. The only difference was the sun came out from between some clouds. When I adjusted the crosshairs to be more nearly perpendicular to the rod the distance was again correct.
Holy Cow, post: 342799, member: 50 wrote: Encountered one of those quirks that has happened very few times over the years. What is scary is how easy it would have been to have introduced an error of this magnitude.
Simple total station work traversing along a twisty, narrow county road. I had just put the rod man on a predetermined point and had him set a temporary nail. Let's say that distance was 290 feet. I also wanted to know the apparent location of the center line of the roadbed at basically the same distance from where I was at. He sidestepped roughly eight feet to his right. I took a shot. Very close to 322.8 feet. Not once, but several times. I told him to hold on while I tried something. I rotated a few degrees to the left and then to the right and then back to prism. Every reading was now very close to 290 feet, as it should have been. Somehow the internal brain of the total station was adding 10 meters to the distance.
Y'all be careful out there, ya hear?
I've seen that a few times. Or I have had a parallel wall along a traverse line bend my line of site a couple of feet. All stuff you have to be careful of and experience helps you sort out. One of the reasons I always thought some guys were overreacting to the chances of a flier when using RTK. When we would traverse we had chances of abnormalities and we still do with gps.
IÛªll second ÛÏY'all be careful out there, ya hear?Û My 2 cents, Jp
I remember our old Sokkia SET2BII returning a weird distance now and then- usually due to a second prism within the infrared laser's cone messing things up (they were comparatively wide in those days), but occasionally no logical explanation could be found.
Sometimes our Leica TCRA1103 would record a goofy angle or distance on the ATA flash card but I usually attributed it to cosmic radiation corrupting the card itself.
I was gonna guess it was a Topcon 600 series. Our old 605 used to do that but it was always 18.5' or something like that.
SellmanA, post: 342836, member: 8564 wrote: I remember our old Sokkia SET2BII returning a weird distance now and then- usually due to a second prism within the infrared laser's cone messing things up (they were comparatively wide in those days), but occasionally no logical explanation could be found.
Sometimes our Leica TCRA1103 would record a goofy angle or distance on the ATA flash card but I usually attributed it to cosmic radiation corrupting the card itself.
There's an old thread on this phenomenon; I can't find it. There's something unique and special about the "10 meters". Perhaps Conrad or Kent remember the discussion. I think it had to do with harmonic frequencies of EDM's or something like that (way above my pay grade).
I found a similar error about 30 years ago with a top mount edm shooting across a road. The distance was short enough that the error was obvious. Turned out to be a flat 10 meters. Did some research and found out that if most edms blew a distance it was usually by 1, 10, 100, etc., meters.
Never had this happen with a "modern" gun but the old HP3800 would gain/lose 100 feet, depending on how you "tuned" it. If it close on the meter you'd better check it.
Andy
This might be the same instrument A deputy surveyor used in 1872 on a township line I'm working on!
Andy Bruner, post: 342859, member: 1123 wrote: Never had this happen with a "modern" gun but the old HP3800 would gain/lose 100 feet, depending on how you "tuned" it. If it close on the meter you'd better check it.
Andy
Only had this happen once with my HP3800 on the 100' setting. Did have a 10 meter bust one time with my HP3810A in a traverse leg. Also had readings of a 10th if a meter error when setting road offsets, would come up 0.32' short on one side and 0.32' long on the other, which would result in the correct overall distance between monuments. Really should have been using a chain (tape) in the first place. On one survey it also gave me an even foot error in one line measurement. Never was able to determine why these errors happened as they never occurred again.
Andy Bruner, post: 342859, member: 1123 wrote: Never had this happen with a "modern" gun but the old HP3800 would gain/lose 100 feet, depending on how you "tuned" it. If it close on the meter you'd better check it.
The old 3800s were bad about that. An operator had to have a "feel" when dialing in those integers. We had a procedure of double-checking the metered distances from both ends, BS-FS and FS-BS to try and eliminate that possible error. ahhh...the good old days.
Could it be that your gun was getting a return on a street sign or the reflective centerline striping in the background?
When we first purchased our Trimble 5603 I would get errant distances. In TRK mode, it would get a distance painted white wall, even though it was not in reflectorless mode.
Piss poor quality control on the firmware.
Eventually after several firmware upgrades a few years down the road, the problem disappeared.
Had a Spectra Precision 610 model. It also had major problems with random distances occurring about every couple hundred shots. They would only be 100' or so off. Nothing you could pin down.
The top of my head would have been the most reflective thing for a long, long way. This stretch of "road" has more dirt than crushed rock and comes to an abrupt halt about 600 feet further along at a barricade built of I-beams about 15 feet from where the bridge was removed about 40 years ago and never replaced. The top of the rod man's head would run a close second for reflectivity but that hat stays firmly attached at all times.
