Static GPS

I'm sure you'll get 50 replies telling you the "best way"...

I've done a ton of static processing in TBC, as well as network adjustments, it is more than adequate for the task. You'll want to bring as many receivers to the party as you can, since your number of non-trivial vectors for each session is N-1 where N is the number of receivers in the session.

For a network adjustment to be meaningful you should have fully independent redundant observations of each vector to be included in the adjustment. If possible you should have ties to more than one primary control point as well, even if just for a check.

Have the crews take detailed field notes including, at bare minimum, point number, receiver S/N, start and end times, and antenna height. Have them verify the height, level, and plumb after ending the session but before breaking down.

Feel free to e-mail me with any questions, especially on TBC. Hopefully your Trimble dealer has someone on staff who's qualified to train on static networks and network adjustment.

Why not use your receivers for OPUS & OPUS-RS? Get both types of observations on each station on a couple of different days. This should give you results you can be confident with and probably surprise you with the results. You may have to convert to the specific datum you want after you get the solutions, but that should be simple.

What is the distance between proposed adjacent stations?

>I am figuring we will need to purchase Trimble Business Center to properly post process the data, is this correct?
I'd be looking hard at CarlsonGNSS before I bought TBC. The raw data files can be converted to RINEX using free utilities, and any baseline processing software can use RINEX, so you aren't necessarily restricted to TBC.

> Why not use your receivers for OPUS & OPUS-RS?
That's not the proper way to establish a control network, for various reasons. The least of which is that he needs to tie his survey to local NAD83(91) control, and OPUS would yield NAD83(2011).

Static GPS, To Answer Your Questions

Using 1 CORS and 2 receivers you get 3 vectors per observation period. A third receiver would give you 6 vectors in that same time. A big improvement. A fourth gets you to 10 vectors. The formula is (#receivers) Factorial. Since you are only looking to do 4-6 points, I would say time & cost wise, there is not much difference in using 2 or 3 field receivers, and unless you have other work there is no payback on the fourth.

Since you are within a municipality, another alternative is a couple of L1 receivers. Much better payback on static work in a smaller area. Today I used 2 L1/L2 and 2 L1 receivers. I had a third L1 along, but I found out a long time ago that the fifth unit with added setup and breakdown time did not improve efficiency and parts of each observation set were longer than needed. If you needed 5 control points, then it is once and done.

I disagree with Norman, OPUS & OPUS-RS result in well defined coordinates and any Least Squares program can use vectors and/or coordinates. OPUS-RS coordinates are based on a network solution of up to 9 CORS using twice as many observables. What I recommend for 4 to 6 control points are 3 hour morning and 3 hour afternoon sessions on different points, 3 days and you are done with 6 double occupations. Send off to OPUS as a data check. Break into 3 1 hour observations, send to OPUS-RS and exclude your nearby CORS from the solutions. Submit comparable files from the nearby CORS to OPUS-RS. Use a post processor with the local CORS and your receivers and adjust. Compare the results. Adjust all your OPUS-RS coordinates against the vectors. Once you are satisfied with your network quality you can transform it to whatever you like. Trust me, you will very likely ignore your OPUS Solutions in a CORS rich area.

If you buy just 1 L1 receiver you can tie your base into another or new control point every time you RTK in the future, easily 2 points a day or you can static check 4 or more of your RTK points a day.

Paul in PA

Static GPS, To Answer Your Questions

> I disagree with Norman....
The fact that NGS has developed OPUS-Projects should be proof enough that they agree with me.

> If you buy just 1 L1 receiver...
Our man indicates that he has a base/rover pair and wants to establish 4 to 6 control points. I think that will be plenty of equipment to do the job.

How Does NGS Agree With You?

Did you make a suggestion that I fail to see?

Paul in PA

With only a few stations you can afford to re-observe the whole network, in a different order to the first observations. This will pretty much guarantee that you will eliminate (or at least identify) any height or centring errors. It's surprising how often a height can be measured wrongly, twice! TBC provides good processing, as long as you carefully check through the data and discard any short sections. Read all the graphs carefully.

Static GPS, Number Of Vectors

I miss-wrote the formula for vectors, it is (#receivers = 1) Factorial.

Another vector concern, Chris Mills said "discard any short sections".

If you are setting up and removing receivers as the day goes by you may get some short in time vectors, most likely not as tight in precision. In my software the term is to "exclude" that vector from the LS solution. Only that station to station vector is excluded. In my recommended 3 hour sessions you should be able to get 1.5 hours or more of completely common vectors.

Paul in PA

Static GPS, Number Of Vectors

Yes, Exclude if the overall record is long but only a short timescale is common to a particular vector.

Discard if the observed time at a specific set-up is short or broken into small segments. The first and last few minutes of any SV being visible tend to hold most of the problem signals, so unless you are very short of SV's then those parts rarely help the solution.

Sat in wide open plains it won't be a problem, but if you are based in a well populated, well tree covered part of the world clear skies are often a luxury.

Check to see if your State has any guidelines that would be helpful. Here is a link to the NYSDOT Guidelines/Procedures (see Ch8).

here

How Does NGS Agree With You?

> Did you make a suggestion that I fail to see?
Apparently.

NGS-Projects resolves simultaneous observations into a networked solution which includes vectors between the adjacent stations. If NGS believed that a local control network like the one proposed here could be adequately established with reference to OPUS only why would they develop this new product? Further, why would they design it to work in the way it does rather than the way you described?

Lots of local control networks are being established with reference to OPUS only, including the one in Norman, Oklahoma. If close enough is good enough not much bad is likely to happen as a result. But it just isn't the right way to do it. Having direct vectors between the adjacent stations will tighten up the local networks enormously. That is going to be more apparent in a future with over a hundred satellites yielding autonomous positions at the centimeter level.

Static GPS, To Answer Your Questions

So you do not eliminate dependent/trivial baselines? I am genuinely asking, some people do - some do not. I was taught that not doing so will skew the statistics to look much better than they truly are. When you discard trivial baselines the formula is Baselines=Number of receivers - 1.

1 Cors and two receivers means two baselines; three receivers means three baselines...

Given that he has a monument on project datum that he has used in the past and two receivers I would recommend an old-fashioned (but tried and true) GPS loop. I would do it twice and check for blunders with dual RTK observatoins...

Also, the procedures needed would be dependent on what is meant by "tight". You can go so far as the NGS route and check the plumb of your fixed-height tripod with your total station from two 90 degree apart setups, etc.

I think this has been linked to on here before, but it is good reading for someone new to static GPS.

Trivial Baselines