New changes and options with new LS Plus

Darren Clemons

Well-Known Member
Darren: I have been following your posts. I received the LS Plus June 5. We had returned all 3 parts ( LS _00422, Triumph_1M20215, and the 35 Watt Radio). We have experienced your exact same problems, especially loosing the UHF at about 1,200' with line of sight of the antenna and the message to attach the external antenna (antenna is attached). We use the Matt Johnson (T1M radio issue) "For switching to the external radio" procedure. Did you ever come up with a solution that you would share?
Our main solution was to finally get several mifi’s and go to TCP corrections and finally ditch UHF completely (where we can of course).

From what I’ve been told it was an internal issue after the units were updated to the Plus. I have also now been told that the issue has been solved and that I will have to send back my 1st LS plus unit to make the necessary repairs.
 

Robert Jordan

New Member
I have just the LS...and I’m hooked on this rtpk. Based on above conversations I’m not sure if I need to update to the latest prerelease as I might lose it for now. Advice please.
Wes, I upgraded to the latest pre-release and lost the RTPK, so you'd better stay where you are for a minute. I miss it!
 

Shawn Billings

Shawn Billings
5PLS
Double check that Robert. From what I can see, the current pre-release version is 3.0.8.400 and it still has the checkbox for Enable Post-Processing on the device. (Accessible from compass button on action screen, Options, or Home>Setup>Edit under Action Group).
 

Wes Hand

New Member
Robert,

I actually upgraded late this afternoon and mine was working fine. I did not set up my base but I was dialed into our RTN in GA and it was only processing gps and glonass data but the speed In which it processed and gave a solution was noticeably faster than the previous prerelease. The previous prerelease took too long to process 400-600 records in my opinion to be called real time. Today after I updated it processed 512 records in less than 90 seconds. Before it would take forever. I look forward to testing it tomorrow once I add the other constellations.
 

APADDEN07

New Member
Our main solution was to finally get several mifi’s and go to TCP corrections and finally ditch UHF completely (where we can of course).

From what I’ve been told it was an internal issue after the units were updated to the Plus. I have also now been told that the issue has been solved and that I will have to send back my 1st LS plus unit to make the necessary repairs.
We've been having similar issues since we got our LS+ and T3, sent back one T3 already, curious if our LS+ will need to go back now as well. Ours wasn't an LS upgrade, just a fresh LS+ out of the box, but I wonder if we've got the same internal issue.

As for the RTPK, it's definitely a cool idea, but I still get results like this more often than I would like and there doesn't seem to be a way to differentiate between the solutions once the RTPK process has run. After DPOS my points still just show rtk and ppk options, no column for RTPK that I've noticed.
 

Attachments

Shawn Billings

Shawn Billings
5PLS
We've been having similar issues since we got our LS+ and T3, sent back one T3 already, curious if our LS+ will need to go back now as well. Ours wasn't an LS upgrade, just a fresh LS+ out of the box, but I wonder if we've got the same internal issue.

As for the RTPK, it's definitely a cool idea, but I still get results like this more often than I would like and there doesn't seem to be a way to differentiate between the solutions once the RTPK process has run. After DPOS my points still just show rtk and ppk options, no column for RTPK that I've noticed.
One engine fixed for the entire 62 second observation. In this case, RTPK is likely correct, but what it tells me is you need more data. Either you need more RTK observations to get the RTK to agree with RTPK or you need more raw data to get RTPK to agree with RTK or you need more of both because both are wrong. It tells you that you don't have enough information yet to know what is right. Was this under canopy? The engine RMS looks pretty rough at 0.305' and with only one engine fixed, it seems this must have been a poor location, in which case you CANNOT trust a single RTK fix.

In the point list, at present, the RTPK solution will be in the PPK column. Ideally both DPOS and J-Field RTPK will post-process and deliver the same results because both use the same processor. So it won't matter if you DPOS process from base to rover later, you should get the same answer. In reality, there could always be some minor version difference in the two post processing engines which means you might get a different answer from DPOS than you would from RTPK.
 

APADDEN07

New Member
Thanks for the clarification on the post processed results, that makes sense and I'm sure its the simplest solution to program, it just wasn't clear that that was what is happening.

That screenshot was part of some troubleshooting, and I was only able to get 1 engine fixes consistently, which is unrelated to the RTPK, it wasn't what I would call a poor location for GPS, fairly well daylighted dirt road running north-south, I would say fair to moderate.

Here's another one, which is obviously much closer to the RTK, but visually it looks like a real outlier, it's in the open in our parking lot at the office so no canopy issues.
20200727-14.05.45_3_COLLECT.png
.

It seems like most of the time the LS selects RTK over PPK in the points screen if you auto select the best points, so forgive me if I'm a little gun shy on the RTPK.
 
If I may chime in here.....I have been so reluctant to update to any of the new updates. I am somewhat lost with the new terminology that seems to come with the new releases and the thought of different results RTK vs RTPK vs PPK currently mystifies me.
I am still running 3.02.1151 and it provides rock solid results for me.
 

Michael Stazhkov

Developer
Staff member
5PLS
Here's another one, which is obviously much closer to the RTK, but visually it looks like a real outlier, it's in the open in our parking lot at the office so no canopy issues..
It seems like most of the time the LS selects RTK over PPK in the points screen if you auto select the best points, so forgive me if I'm a little gun shy on the RTPK.
Could you send the project to support for investigation? (Projects screen -> push "Send Project to Support" button). It can help us to improve RTPK.
 

Darren Clemons

Well-Known Member
If I may chime in here.....I have been so reluctant to update to any of the new updates. I am somewhat lost with the new terminology that seems to come with the new releases and the thought of different results RTK vs RTPK vs PPK currently mystifies me.
I am still running 3.02.1151 and it provides rock solid results for me.
John,
You are correct in that there are many new phrases and quite a bit of new terminology in all the new changes, but there’s really not that much different once you see it and use it a little while - except the fact that with either the LS Plus upgrade or using the rtpk feature, or both, you can and WILL get much quicker, redundant accurate results under canopy.

That’s what all of these features and the Plus upgrade are all about - time spent under canopy. If you’re getting good, timely results you’re happy with, that’s great. If you’re spending very much time verifying and validating those results, these upgrades and changes will decrease that time dramatically.

The rtpk feature is about as simple as any feature can get. When you are on a point and stop the session, the LS automatically starts downloading the data for the time period matching that session from your base (or RTN) and then internally in the LS (no internet connection required) runs the post processing and gives you the PPK solution right there on your point in real time. This process, as Shawn mentioned in his excellent writeup in another thread, has already been sped up tremendously and only takes about 3 to 4 minutes for a 500 to 600 second observation. With that, you are basically getting two shots in one and very, very often the rtk and the rtpk match and you have your redundancy. If they don’t match, you have two choices, either hit resume on your current session and the data will be added to the current session and can be reprocessed, or you can simply start another session and let the LS process another rtpk solution in another 300/400 seconds. If I have an rtk solution I feel good about, I like option 2 as I can use the DTL white box and will very often get data that will verify that 1st shot was good well before I complete an entire 2nd session.

Some of our tougher locations that were previously taking 3 or 4 fifteen minute PPK sessions, of which we wouldn’t know the results of until we were back in the office, now are taking a maximum of 12 to 15 minutes. The rtpk data with all four constellations is so much more robust that 4 to 5 minute observations are typically producing very good results. Instead of possibly spending an hour and having four 15 minute PPK sessions we hope of which two match, we have in 15 minutes either good rtk and rtpk matching on the very first point or at least 2 much more quickly acquired than in the past rtpk shots that match - and we get it right there on the spot - no wondering or waiting until you get back to office to process!
 
Thank you Darren for a very nice understandable presentation of the benefits. I NOW have a good idea of what RTPK actually means now.
I fully plan on getting the upgrades very soon. I do work under canopy and better and faster collection of data is important - no matter the environment.
I appreciate my more knowledgeable Javad mates working out the bugs for advancement and will wait until it reaches a rock stable state.


 

Darren Clemons

Well-Known Member
Darren: Are you using the Precise Topo action profile that Shawn recommends?
No, we aren’t using that profile. We’ve always created and made our own profiles for different circumstances. With the Plus upgrade, we had to tweak our boundary profile we’d used for several years with the standard LS. We’d always required a minimum of 2 engines during phases two and three, but now, with only the 4 “super” engines, getting 2/4 is much more difficult than previously getting 2/6 so we’ve lowered our minimum to only 1 engine.

This creates a few more jump/fail episodes than before, but the trade off is you’ll get many more epochs much quicker so you’ll get back to phase 2 quicker than before. Regardless of the profile you choose, it’s always been quite rare for us, in some of the very tough places we go, to ever get through phase 3 anyway. It’s just a process to get rtk with time separation over 240/300 seconds and now using rtpk solutions to verify.
 

APADDEN07

New Member
No, we aren’t using that profile. We’ve always created and made our own profiles for different circumstances. With the Plus upgrade, we had to tweak our boundary profile we’d used for several years with the standard LS. We’d always required a minimum of 2 engines during phases two and three, but now, with only the 4 “super” engines, getting 2/4 is much more difficult than previously getting 2/6 so we’ve lowered our minimum to only 1 engine.

This creates a few more jump/fail episodes than before, but the trade off is you’ll get many more epochs much quicker so you’ll get back to phase 2 quicker than before. Regardless of the profile you choose, it’s always been quite rare for us, in some of the very tough places we go, to ever get through phase 3 anyway. It’s just a process to get rtk with time separation over 240/300 seconds and now using rtpk solutions to verify.
What other settings are you using with your 1-engine profile? We've been having similar issues getting 2-engine fixes, but so far the 1 engine results have been spotty, and require some time consuming verification.
 

Darren Clemons

Well-Known Member
What other settings are you using with your 1-engine profile? We've been having similar issues getting 2-engine fixes, but so far the 1 engine results have been spotty, and require some time consuming verification.
The main thing that has to be changed is the consistency to zero if you’re allowing 1 engine fixes. It does not increase at all with only one engine. If you have it on 20 or 30 as was common in the standard LS it will never “finish” phase 2. The confidence, which is how many resets it does in phase 1, I’ve actually lowered to 5. Again, this is much lower than I typically used with the standard LS, but in almost all places the Plus just doesn’t get that many odd buckets anymore.

I typically am letting it run for about 350 to 400 seconds and whether it has any good rtk or not, I stop it and let the rtpk process and store. I then repeat that process as many times as needed until I have two or more rtk and/or rtpk that match.

Remember, you will need to go into the points screen and manually change the coordinate to the rtpk solution after it stores it as it will automatically hold the rtk coordinate until you do.

If you are a user of the DTL (distance to last) white box to aid in verification, which is as handy as a pocket on a shirt, it will always report the stored rtk of the last shot, even if you manually change it to the rtpk. You’ll need to use the Pdelta box and physically type in the point number of the last shot. That will then report you some nice 0.03’ish type numbers in that box if session 2 gets an rtk bucket that matches the previous rtpk coordinate. If it doesn’t after another 350/400 seconds stop and let it calculate another rtpk. Then go to points screen, change that coordinate to the rtpk and inverse the two points....

You'll likely get three points and two or all three will match and all this will be done in about the same time as it used to take for the very first of a possible three 15-18 minute PPK sessions.
 

APADDEN07

New Member
Thanks Darren! I hadn't noticed that the box was holding the RTK, that's a critical catch.

I also noticed this morning that before processing, if you auto select the best points, it's mostly the ppk/rtpk column, but after DPOS, the results shift to the RTK column, even though most of the points never got an actual RTK fix, just an RTPK fix.

Can someone confirm if the "Real-Time" coordinates are all shoved into the RTK column after processing through DPOS, and the new PPK column contains the post processed results if the raw data is stored after collecting an RTK/RTPK shot? I have several PPK coords saying float now, that were previously an RTPK fix. The RTPK procedure for letting it cook and process and then cook some more is working excellent for heavy canopy, but comparing the differences in coordinates is a little confusing still.
 
Top