Playing with a pro GPS

[Bill93]

Last year I acquired a Trimble 4000sst receiver and accessories. It has gotten me some interesting measurements, but doesn't qualify for the NGS program for GPS on Benchmarks because it doesn't process enough parts of the GPS signal.

 

Recently, I bought a slightly newer model, the 4000sse, which uses all the old accessories and does qualify for OPUS shared solutions which can be used for the GPS on BM program. These receivers can be bought on eBay occasionally for a few hundred dollars, but you need to know what all is needed in the kit and to distinguish the reasonable models from the totally obsolete.

 

Here's the setup on a tri-station with old receiver. The newer one doesn't look much different.

 

 

For those who don't know about OPUS, here's a summary: The receiver records measurements every 30 seconds, and stores them essentially as a time delay from each satellite in view. You download those into a computer file and after some waiting time for CORS and orbit data to be collected, you send it to a web site at NGS. They automatically process this against CORS station data. A few minutes later it emails you a report that tells you where you were and some measures of performance.

 

The orbit data is either "Ultra Rapid" meaning relying on predictions and available within a few hours, "Rapid" meaning updated with real measurements the next day, or "Precise" which has been processed to the best possible estimate but 2 to 3 weeks after the fact.

 

There are two processing options, OPUS-S or static for 2 hours to 48 hours, and OPUS-RS or rapid static for 15 minutes to 2 hours. The RS algorithm is better in some ways and lets you get away with the shorter sessions for some uses, but OPUS-S with at least 4 hours is needed for any NGS use of the data.

 

I'm hoping that when the weather warms up I can do some OPUS shared solutions on benchmarks that NGS can use as they check their gravity model. I want to have checked my results on a couple HARN stations first.

 

Today I got the results using final "precise" orbits for a 7-hour session on a point in my back yard, processed by OPUS-S and also split into 7 sessions of 1 hour each for OPUS-RS. Here are some plots I put together to compare results. There are effects that the processing doesn't know about, so the supposed 95% confidence limits only apply to some of the error sources and can't be taken at face value. But note that the horizontal grid is 1-cm squares and the vertical grid is 2 cm. All of the horizontal position outlines would fit comfortably on a US 5-cent piece.

 

 

Edited December 20, 2016 by Bill93

[astrodanco]

Hi Bill. I've also collected a few old GPS L1/L2 survey grade receivers.

 

Are you using a regular surveyor tripod and a tribrach? It might be easier to use a two meter fixed height rover pole (with level vial) and a bipod (or tripod).

 

I've found local HARN/HPGN stations to be fairly useless as anything other than a rough check. They're all HPGN 1991.35 and it doesn't appear possible to convert between HPGN 1991.35 and NAD83 (2011) Epoch 2010.0 with high accuracy here on the West coast. The best the NGS tools can do is 4-5 centimeters. Perhaps there's some unmodeled motion in there somewhere. In any case I can obtain more accurate positions than that by resurveying the station.

 

In addition to OPUS, have you tried submitting your RINEX data files to:

 

1. NASA APPS

2. CSRS-PPP

3. Trimble CenterPoint RTX Post-Processing

 

I've also tried SCOUT, but can't seem to get it to actually process uploaded RINEX files.

 

Have you signed up for OPUS Projects yet? If not, NGS sometimes have upcoming OPUS Projects training online webinars available that you can sign up for and gain access at zero cost. OPUS Projects is the bomb.

Edited December 21, 2016 by astrodanco

[Bill93]

Yes, I should check out the other processing options.

 

Doesn't OPUS Projects require simultaneous measurements at multiple sites? I only have one good setup.

 

I'm using a tripod and tribrach. I measure up on 3 places to the ground plane to check centering/leveling, and calculate back to ARP height above the mark. I've pasted a spreadsheet in the back of the field book to quickly give me the difference. I don't think I have any problem with repeatability of setups, because session results seem to match as well as the pk-pk values of a session. Fixed height would avoid the calculation and chance of blunder there, but would require me to get more hardware and figure out how you make it stable. It seems like you would either need a sliding sleeve for the rod if using a tripod (4 legs aren't stable), or have to be careful at balancing and carry weights with a bipod.

[astrodanco]

Doesn't OPUS Projects require simultaneous measurements at multiple sites? I only have one good setup.

 

The other observations can be from CORS or PBO or what have you sites. They don't have to be your own. I would have a hard time explaining it, but it lets you arrange all the stations and how they are used by OPUS during processing into in a hub and spoke arrangement that improves your statistics. It's a better way of processing on their end though move involved to set up and perform on our end.

 

See ftp://geodesy.noaa.gov/pub/opus-projects/

ftp://geodesy.noaa.gov/pub/opus-projects/intro.pdf

ftp://geodesy.noaa.gov/pub/opus-projects/create.pdf

ftp://geodesy.noaa.gov/pub/opus-projects/upload.pdf

ftp://geodesy.noaa.gov/pub/opus-projects/process.pdf

ftp://geodesy.noaa.gov/pub/opus-projects/manage.pdf

etc.

 

or have to be careful at balancing and carry weights with a bipod.

 

Yes indeed, I carry weights with me when using a bipod and rover pole for long static sessions. Not a problem for RTK though. I use the same weights one would use for a canopy. I bought them at the local Dick's sporting goods store. They're inexpensive (I think it was $20 for a pack of four) and the slide in opening on the weights fits the bipod legs nicely. Quick on and off. I also have a more expensive standalone fixed height SECO tripod that I picked up off Ebay. Even it sometimes must be weighted down when used on concrete or asphalt. The legs were too big for the openings on canopy weights, so I bought some leg clamps off Ebay and use my ADM Accessories (telescope) weights. I can also substitute non-skid rubber feet (from Home Depot) for the leg spikes when needed.