Sudden Loss of Workable Fix in Open

James Suttles

Active Member
For a fairly good while, around 2:30PM till around 4:00pm ET, we struggle with getting a fix or decent RTK. We recently purchased an S3 for the office base. Today we again were in the wide open, and the LS was at float..... Another crew and there LS similar situation, they could get a fix, one minute locate a point and try to replicate the point, and it as 1' different. So I started turning off constellations. I went back to standard GPS, Glonass, and it went to working, it was slow but working. Turned on Galileo and it still worked but still not anywhere near 100%. So I looked at the S3 and it had flagged Galileo Satellite #18 as unhealthy. I turned off the unhealthy satellite and added back BDU and the LS went to working again.

It happens about the same time everyday, so I am guessing that since the 24 hour orbit, that is why the consistency. I will have a chance tomorrow to try and see if my theory worked, I will update once I know more. If others are having similar experiences, it would be good to collaborate to see if there are any similarities.

Thanks
 

James Suttles

Active Member
I retuned to the site this AM, and relocated several of the points collected during yesterdays weird float situation. The results were mixed, with a couple of points staking out about 0.08' from yesterdays RTK, one point was 0.60' from yesterdays collected an RTPK's point. So not sure what may have been the issue. Still have a few hours to see if this afternoon has the same issue.
 

James Suttles

Active Member
Well 2 O'clock is here and the problem is still here. The problem of floating in the wide open still exist, but not as much as yesterday. It will fix, but all indications are that the info from RTK is incorrect. An example is we were trying to set a corner and RTK got us within a 0.10' thought all was good, so we were letting it collect a shot to be sure before we set the corner and it suddenly jumped to move 7' then invalid jump, then to 2 float', then 4' then 0.20' etc. The RTPK indicated move 0.04' and it repeated the same solution 3 times, while RTK still was bouncing around. We used RTPK to set the corner, to within 0.02', RTK would never get a solution that would agree consistently, it did get a closer solution of about 0.06' for about 20 seconds, then it bounced to 3', but it never completed the shot on its own. This issue of around 2 or 3 O'clock has existed for quite sometime, with slowed or diminished responses from RTK, but the last few days it has been significant and the results from the RTK is very questionable during this time. As mentioned earlier we re-located some corners from yesterday afternoon, this morning and the results cause pause.
The differences are as follows:
#2009=0.11'
#2010=0.19'
#2003=0.08'
#2004=0.64'
#2005=0.04' so you can see the differences are not consistent. Point #s 2010 and 2009 are wide open sky, #'s 2003,2004 and 2005 are in light to medium canopy.

I have removed from the correction stream satellites that are flagged as unhealthy, Galileo 18 and Galileo 14, and the problem still exists. Up until 2 O'clock, you could not ask the the LS to perform any better, but after that until about 4 or 4:30, its hit and miss, and really makes you think its best to just stop during this time of day, and our schedule and workload will not support that. The level of uncertainty during this time is scary.

Any ideas?

Thx
 

Matt Johnson

Well-Known Member
5PLS
The raw data is needed for this to be investigated. Please share it.

Are you using AltBOC signals? They have occasionally caused issues in the past. Try turning them off if you are using them.
 

James Suttles

Active Member
Those signals were on yesterday, I turned them off, so that could have been some of the issue. I will send the project in tonight, once the crew gets back into the office.
 

James Suttles

Active Member
The project file is 64mb, so here is a link to download the file.


The points labeled #fj ( example=2005FJ, 2004FJ) indicate # found with Javad Also is Day1
The points labeled #fj2 (example =2005FJ2, 2004FJ2) indicate #found with Javad Day 2

If the link doesn't work, let me know.

Thx
 

John Troelstrup

Active Member
For a fairly good while, around 2:30PM till around 4:00pm ET, we struggle with getting a fix or decent RTK. We recently purchased an S3 for the office base. Today we again were in the wide open, and the LS was at float..... Another crew and there LS similar situation, they could get a fix, one minute locate a point and try to replicate the point, and it as 1' different. So I started turning off constellations. I went back to standard GPS, Glonass, and it went to working, it was slow but working. Turned on Galileo and it still worked but still not anywhere near 100%. So I looked at the S3 and it had flagged Galileo Satellite #18 as unhealthy. I turned off the unhealthy satellite and added back BDU and the LS went to working again.

It happens about the same time everyday, so I am guessing that since the 24 hour orbit, that is why the consistency. I will have a chance tomorrow to try and see if my theory worked, I will update once I know more. If others are having similar experiences, it would be good to collaborate to see if there are any similarities.

Thanks
FWIW,
I have the same issue here in Florida and within the same window as you describe.
My LS is standard issue with intent to upgrade it soon.

I was curious to see if the upgrade to more satellites would alleviate this problem but it appears it will not.
I am not a GPS scientist but I assume that the constellation is not fully packed yet, to provide constant stellar results and the window of time that we are both experiencing is evident of a gap in proper satellite geometry?
 

James Suttles

Active Member
RTK, will either not fix or when it does fix, the data that is collected is incorrect. Also during collect, there is lots of invalid jumps. We tried to stake out a few points, and RTK would get within say 0.10' and you decide to collect some data to be sure you are in the right place, and then it jumps to 4' away, saying fixed, and then 7' away, so its just really a frustrating time of the day. We relocated the points the next morning and found the differences to be significant.

If the sats that are unhealthy are the issue is there anyway to automate the LS to ignore those sats that are unhealthy?
 

Matt Johnson

Well-Known Member
5PLS
If the sats that are unhealthy are the issue is there anyway to automate the LS to ignore those sats that are unhealthy?
I believe that they should already not be used. You can check the satellite status screen and look for code 06 which means they are unhealthy and not used.

GAL 14 is visible now and already flagged unhealthy.
65FD3D3A-0575-45B6-9D0C-E5940C3747BC.png
 

Matt Johnson

Well-Known Member
5PLS
They are obviously causing an issue even if they aren't used.
It is just speculation that the issue is caused by unhealthy satellites. Alexander is going to investigate the data. If the issues happens again it would be good to capture screenshots of the RTK engine screen and satellite status screen so we can see what satellites and signals are being used.
 

James Suttles

Active Member
Update: The time of 2:00 to 4:00 came today and everything went smoothly, the difference today is that sat #'s GLO#11, GAL#14 and GAL#18 were not being transmitted by the S3 base. It maybe a coincidence, but that is what is currently different than the last 2 days of trying to nail down what has been plaguing us.

I will update tomorrow, to see today was just a coincidence.
 

Adam

Well-Known Member
5PLS
I can second that. I was set up using your base from 2 to 4 and it worked well.
 

Adam

Well-Known Member
5PLS
I spoke with Shawn a little bit ago and he experienced poor performance today then turned off Galileo and it fixed right up. The problem is certainly a moving target.
 

Phillip Lancaster

Active Member
Update: The time of 2:00 to 4:00 came today and everything went smoothly, the difference today is that sat #'s GLO#11, GAL#14 and GAL#18 were not being transmitted by the S3 base. It maybe a coincidence, but that is what is currently different than the last 2 days of trying to nail down what has been plaguing us.

I will update tomorrow, to see today was just a coincidence.
Are you talking about the Delta 3S or the T3? What is the S3?
 

James Suttles

Active Member
You are correct its the New Delta 3S. We have it as a permanent base at our office, which helps eliminate trying to find a good secure base point location that has a good view of the sky. No more having to find a cemetery, school, or onsite location, seems to work great even in canopy if we are within 12 or so miles from the office. Anyway, Netview & Modem allows us to see which sats are unhealthy and we were able to turn those off with better results yesterday, that may be different today, just have to see.


Thx
 
Top