therock003, post: 421284, member: 12282 wrote: Since data purchase is costly, and want to get best results avoiding possible misfires, i need to make certain of which way to go.
I am indeed situated inside a triangle of the three closest stations almost perfectly. So do you suggest i buy 10-12 hours of data that include different stations, or use the same station but under different circumstances. Like get 3-4 hours of data on day, then at night, and then at a different day thus checking the integrity of the one closest station, OR use a single time window and log 3-4 hours of date and solving using 3 different stations.
How much does 10 hours of data cost?
If you're centered between three stations I'd use all three. As many hours as you can afford but at 25Km distances 3 hours should give you great results. You also want to go to Computations under Project Settings and make sure it's set to Weighted Mean not Single Observation.
I would choose to use 3 CORS, 2 hrs AM & 2 hrs PM, same or different day.
Paul in PA
MightyMoe, post: 421289, member: 700 wrote: How much does 10 hours of data cost?
Its something like 10 Euro per hour. So 120 for the whole 12 hours and a 100plus something for registration since im going to be using the service for the first time. We're talking about 250E for grabbing those exact coordinates
Paul in PA, post: 421345, member: 236 wrote: I would choose to use 3 CORS, 2 hrs AM & 2 hrs PM, same or different day.
Paul in PA
What i'm still uncertain about is if and whether it matters the sum or the continuity of the measurements.
Lets say i have 3 hours of static data. Does it matter if its 3 hours non stop or 1.5 am & pm or 10 mins here 10 mins there, then another 25 mins and thus scattered throughout the day or the following days to conclude those 3 hours?
I wont be doing that of course, i just want to have an idea on what to expect on each case
Lee D, post: 421295, member: 7971 wrote: If you're centered between three stations I'd use all three. As many hours as you can afford but at 25Km distances 3 hours should give you great results. You also want to go to Computations under Project Settings and make sure it's set to Weighted Mean not Single Observation.
Ok noted, i will go ahead and do that. If CORS support GLONASS, should i go ahead and use both the GPS and GLONASS IGS orbits?
How does TBC handle orbits once their imported on the project? I did some testing on the roof having my base on the rod (autonomous reading) and the rtk 10 meters further just to see if baseline calculation is correct, and i did baseline processing both with and without downloading the orbits. I did not see any changes. I got 10.273 meters on both occasions
I would never use a sum of rather short observations without careful review of the data. If you look at your daily GNSS constellations throughout the day in a "Planner" you will find some 15 minute gaps with poor DOP, which may give you an outlier result that is better to exclude than to include in the mean. If you are using GPS and Glonass the gaps should be covered by the other constellation.
If you are looking to establish a base you will use over and over you want to verify your data time. Since you are purchasing data after your personal observations, over observe and then select your time periods. Start recoding data 15 minutes before the hour and then delete that 15 minutes for your precise processing.
You will want to have a GNSS Planner program you are familiar with, but that does not include unexpected satellite outages. For GPS you would want to get US Coast Guard Navigation Center reports on satellite availability. Go to their site and sign up for daily and outage reports. sign up for a free email subscription:
Paul in PA
Can you get your hands on a few more receivers? I'd like some more control points in the area if I'm paying for the data. You can tie them all together and that will allow flexibility down the road using your own data. I'd at least use the other one you have and set up a good point in another location.
I logged 3 hours of static data, and i'm about to purchase 3 hours each from 3 of the closest available National CORS stations. So i'll import 4 static files in TBC, i'll set 3 of them as control point quality and my home base as unknown and process 3 baselines, and set the weighted mean average on the options to get my fianl position? Wouldnt it be the same to create 3 different projects, each with one of the 3 cors station and my home base, get x,y from each project and then do a simple Xmean=X1+X2+X3/3 and y respectively, to get the same result?
therock003, post: 431974, member: 12282 wrote: I logged 3 hours of static data, and i'm about to purchase 3 hours each from 3 of the closest available National CORS stations. So i'll import 4 static files in TBC, i'll set 3 of them as control point quality and my home base as unknown and process 3 baselines, and set the weighted mean average on the options to get my fianl position? Wouldnt it be the same to create 3 different projects, each with one of the 3 cors station and my home base, get x,y from each project and then do a simple Xmean=X1+X2+X3/3 and y respectively, to get the same result?
It would be close but TBC will probably use weighting based on the calculated reliability of the three baselines.
If you want to see the effects of the the three baselines you can switch them on and off in the network adjustment I believe.
therock003, post: 431974, member: 12282 wrote: I logged 3 hours of static data, and i'm about to purchase 3 hours each from 3 of the closest available National CORS stations. So i'll import 4 static files in TBC, i'll set 3 of them as control point quality and my home base as unknown and process 3 baselines, and set the weighted mean average on the options to get my fianl position? Wouldnt it be the same to create 3 different projects, each with one of the 3 cors station and my home base, get x,y from each project and then do a simple Xmean=X1+X2+X3/3 and y respectively, to get the same result?
There is something you may consider,,,,,
If you have a newer version of TBC maybe you can work around the problem of paying for data.
Of course I've never used TBC for processing a file collected in Greece, but when I upload a file here TBC will automatically offer to process it using the Trimble processing software RTX-PP. I suppose it may not work in your area but that is one option. Also it offers options to use other processing services, probably other posters know more how this works in Europe and would have suggested giving it a try for you if they thought it would work.
Ok so i purchased the data from those 3 closest stations. I set them all as control point quality and left my base as unknown.
-Do i get accurate readings for elevations as well? Cause i also set height as control point quality and alongside setting the datum for my project i got a Z location for my base. Should i consider that an accurate orthometric position?
-I only processed selected baselines from stations to my base, not the baselines between the known stations, i assume this was the proper way to do this.
therock003, post: 433439, member: 12282 wrote: Ok so i purchased the data from those 3 closest stations. I set them all as control point quality and left my base as unknown.
-Do i get accurate readings for elevations as well? Cause i also set height as control point quality and alongside setting the datum for my project i got a Z location for my base. Should i consider that an accurate orthometric position?
-I only processed selected baselines from stations to my base, not the baselines between the known stations, i assume this was the proper way to do this.
I don't process baselines between known stations, I have done it just as an academic exercise, but when I process my control I fix them. The accuracy of your orthometric elevation is dependent on the accuracy of your ellipsoid height and the quality of the Geoid model that is applied to that height. I can't speak to the quality of the geoid model in Greece, the way I figure it out is to locate good elevation control from my GPS derived positions. If Greece is similar to my area you can expect 5cm or so of separation between the actual elevation and the GPS derived elevation, mileage will vary. You MUST apply the Geoid model or you can be tens of meters off the orthometric elevation by using the ellipsoid height.
Did some trials and experiments and encoutered various obstacles. Firstly i thought the known locations of stations were included on the rinex and recognised by TBC. This was not the case though since the ECEF of the stations given were different that the ones shown on TBC properties. So i converted the given ECEF to grid coordinates, i imported the CORS stations as unkown, then keyd in their grid locations that i calculated, in order to set them as control point quality, to finally recompute, and process the baselines.
I got 3 flags after processing
[GALLERY=media, 34]Flags by therock003 posted Jun 23, 2017 at 10:29 AM[/GALLERY]
And then i adjust network and i get a failed chi square test
[GALLERY=media, 33]Chi-square by therock003 posted Jun 23, 2017 at 10:29 AM[/GALLERY]
I followed this video as a guideline
[MEDIA=youtube]y-BGwwN17O8[/MEDIA]
It's not unusual to fail the chi square test on the first pass.
MightyMoe, post: 433467, member: 700 wrote: I don't process baselines between known stations,
I do, but only to verify the integrity of the data.
Lee D, post: 433784, member: 7971 wrote: I do, but only to verify the integrity of the data.
I should qualify, I will process the vector from each CORS to the unknown point separately which is checking the quality of the fixed points. If I like the resulting numbers I will then adjust the figure, I would suggest doing this for therock's project, it's a good way to get a feel for the numbers, generally I will see differences in the 4th decimal place of the seconds for the lat, long displayed in DMS format. Also it allows you to check on just what TBC is doing during the adjustment, I've had some strange results so I like to watch.
Do you allow CORS stations to float at all or do you hold them all absolutely fixed in an adjustment?
Lee D, post: 433798, member: 7971 wrote: Do you allow CORS stations to float at all or do you hold them all absolutely fixed in an adjustment?
I fix them as long as everything looks good, although I will say I don't live anywhere with velocities that coastal areas deal with. I will however throw one out if it looks wonky and grab something else.
That doesn't seem to happen much anymore, the bigger issue is always the elevation.
Guys i may suspect what is goping wron on my project. ECEF refer to specific projection system. On my Case ECEF of the CORS refer to Hepos but the static i logged refers to WGS since when logging a static fine you do not define projection setting or anything. So i think i have to convert these ECEF wgs84 to ECEF Hepos, and then get the grid hepos, before starting to process baselines. Or did i think of that worng?