Test parameters for comparing TLS+ to TLS

oajioka

Member
Hi all,

We recently received our upgraded Triumph-LS+ and we have set up a quick comparison between the upgraded unit and our un-upgraded TLS.

We set up our Triumph-1M base station on our office roof with the following settings:
Protocol: simrx
Format: MSM3.2
Ch. Bandwidth: 12.5 kHz
Modulation: D8PSK

Both units were set up side-by-side under a tree by our office. The TLS+ is using the action profile suggested to us by Michael Glutting named "TLS+ Min2Eng Medium" except it is set up to auto-restart 1000 times. The TLS is using the standard "Boundary" action profile again set up to auto-restart 1000 times.

The idea is to see how many epochs both units record in the same amount of time, and what the horizontal spread looks like.

I will post in this thread again when we have the results of the test, but in the mean-time we're looking for suggestions as to other ways to compare the new and old units.
 

oajioka

Member
I'm back with the results of our test.

The test took place between ~1:15pm and 4:00pm. Checked up on them at ~3:00pm.

Un-upgraded TLS using standard "Boundary" action profile with the following settings.
9993


9994


Three shots were taken with this unit:
#101.1
When we checked on it at 3:00pm it had not stored any shots. This shot didn't verify or validate, and was manually accepted.
9990


#201.1
This shot verified and validated and was auto-accepted.
9991


#201.2
When we came back at 4:00pm it was working on this shot. It verified but did not validate and was manually accepted.
9992


Scatter plot shows all shots landing within ~0.13'
9995
 

Adam

Well-Known Member
5PLS
Looks like you are using the standard GPS and GLO rtk firmware with this one. I must say it's not a fair test if you are comparing gps/glo to multiconstellation regardless of the chip.
 

oajioka

Member
TLS+ using the "TLS+ Min2Eng Medium" with the following settings:
9996


9997


Between 1:15pm and 3:00pm it stored 12 shots.
The first 10 shots were taken quickly, all completing within the first few minutes. The last two shots took much longer, with #100.11 completing around 1:30pm and #100.12 was manually accepted when we came out to check on it at 3:00pm.

At 3:00pm we decided to alter the action profile. Settings were changed to the following:
9999


10000


#200.1
When we returned at 4:00pm to take it down, it had not stored a shot. This didn't verify or validate, and was manually accepted.
10001


Scatter-plot showing all shots landing within ~7'
10002


Scatter-plot throwing out #'s 100.11, 100.12 and 200.1, shows all shots landing within ~0.25'. However, these shots were all taken in the first 5 minutes from the beginning of the test.
10003
 

Attachments

  • 20200528-10.58.51_00719_2_REVIEW_COLLECT.png
    20200528-10.58.51_00719_2_REVIEW_COLLECT.png
    21.3 KB · Views: 242

Adam

Well-Known Member
5PLS
Which rtk firmware are you using?This test looks like gps and glo only as well.
 

oajioka

Member
Hi Adam, thanks for the reply.

Just talked to Michael Glutting, he also said it looks like GPS and GLO only and he is helping us figure out our settings.
 

oajioka

Member
Michael Glutting helped us sort out our base station issues and we are now receiving corrections from GAL and BDS. I will post in this thread again when we have a chance to set up a new test (possibly as soon as tomorrow). Thank you Michael for your excellent support!

Adam: You said it would not be a fair test to compare multi-constellation with GPS+GLO, which is a fair point. Next time we will set up the un-upgraded unit with the 2 or 6-engine multi-constellation firmware.
 
Last edited:

Duane Frymire

Active Member
Michael Glutting helped us sort out our base station issues and we are now receiving corrections from GAL and BDS. I will post in this thread again when we have a chance to set up a new test (possibly as soon as tomorrow). Thank you Michael for your excellent support!

Adam: You said it would not be a fair test to compare multi-constellation with GPS+GLO, which is a fair point. Next time we will set up the un-upgraded unit with the 2 or 6-engine multi-constellation firmware.
What did you have to do with the base?
 

oajioka

Member
Back again with the results of our second test:

This test took place on 5/29/2020 between 9:37am and 2:49pm.

This time the test was broken up into roughly hour-long intervals, with the TLS+ switching between the "Min 2 Eng Fast" and "Min 2 Eng Medium" profiles and the TLS switching between the "GPS+GLO+GAL+BDU 2 Eng" firmware and the "GPS+GLO+GAL+BDU 6 Eng" firmware.

1) 9:37am to 10:37am
TLS+ "Min 2 Eng Fast"
TLS "GPS+GLO+GAL+BDU 2 Eng"

2) 10:40am to 11:37am
TLS+ "Min 2 Eng Fast"
TLS "GPS+GLO+GAL+BDU 6 Eng"

3) 11:39am to 12:43pm
TLS+ ***Incorrect Action Profile No Data Collected***
TLS "GPS+GLO+GAL+BDU 6 Eng"

4) 12:46pm to 1:44pm
TLS+ "Min 2 Eng Medium"
TLS "GPS+GLO+GAL+BDU 2 Eng"

5) 1:48pm to 2:49pm
TLS+ "Min 2 Eng Medium"
TLS "GPS+GLO+GAL+BDU 6 Eng"
 

oajioka

Member
Un-upgraded TLS "Boundary 2 Eng" action profile settings:
10013


10014


First period, 2 engine firmware and action profile, two shots were taken, neither verified or validated and both manually accepted:
10015


Second period, the TLS was changed to the standard Boundary action profile (see post #2) and the "GPS+GLO+GAL+BDU 6 Eng" firmware. Three shots were taken, first two verified and validated, last one manually accepted.
10016


Third period, the TLS settings remained the same (standard boundary profile, 6 eng firmware). Four shots were taken, first 3 verified and validated, last one manually accepted.
10017


Fourth period, the TLS was changed back to the 2 engine firmware and action profile. One shot was taken and manually accepted.
10018


In the last period, the TLS was changed to the 6 engine firmware and action profile. Seven shots were taken, all verified and validated except the last which was manually accepted.
10019
 

oajioka

Member
TLS+ "Min 2 Eng Fast" action profile settings:
10020


10021


First period, TLS+ "Min 2 Eng Fast", 121 shots were taken:
10022

There is a cluster of 77 points with a much tighter spread.
10023


Second period, TLS+ remained on the "Min 2 Eng Fast" action profile. For unknown reasons, only 3 shots were taken in this period.
10024


Third period, TLS+ was supposed to be switched to the "Min 2 Eng Medium" profile, but the wrong action profile was selected. No data.

Fourth period, TLS+ was switched to the "Min 2 Eng Medium" profile. 33 shots were taken.
10025


Fifth period, TLS+ remained on the "Min 2 Eng Medium" profile. 11 shots were taken.
10026

The last two shots, 500.10 and 500.11 were outliers. If we ignore these the horizontal spread is closer to the previous period.
10027
 

oajioka

Member
Forgot to mention earlier that the units were set up in the exact same location as the first test, under a tree near our office.

With the un-upgraded TLS, the 2 engine firmware struggled but the 6 engine firmware was able to take a few shots per hour.

The TLS+ with the "Min 2 Eng Fast" profile took a huge number of shots in the first hour, but several bad fixes seemed to slip through. I am very curious why it did not continue to take shots at the same rate in the second hour.

The TLS+ with the "Min 2 Eng Medium" profile was the best performer in these conditions, able to take more shots than the TLS with 6 engine firmware and achieving a comparable horizontal spread.
 

Shawn Billings

Shawn Billings
5PLS
Some variation will occur due to the ever changing satellite constellations. Your test is significant, but testing from one hour to the next has some variability that a longer test would mean out. Also even though both receivers were under a tree, doesn't mean the multipath each receiver is working through is the same. Even a slight variation could advantage one over the other.

Regarding the two engine standard LS action profile, I would turn off consistency as it only advances when two or more engines are fixed.
 

Adam

Well-Known Member
5PLS
Yeah, I never use consistency when using verify with resets. It's not a fair test with the LS set to 10 and the LS plus set to 0.
 
Top