Emlid

Community Forum

Difference in elevations in fix mode

(Stefko321) #1

Hello everyone,
I recently experienced something strange with Reach RS+. I’ ve made a stake out and have surveyed each point. The results were surprising because since point sovit11 elevations were rising and they should be decreasing.

image
image.png1257×379 17.8 KB
.
i have finished surveying the street and it crosses with recently surveyed street and elevations differ with about 80 cm.
image
The elevations in second pictures are transformed with a licensed program from ellipsoidal to Baltic height system, and they also differ in ellipsoidal height.
image
image.png1281×141 8.67 KB

image
image.png1280×144 8.3 KB

(Christian Grüner) #2

Do you have the raw ubx?

(Stefko321) #4

Yes, i am uploading all the ubx files for this date, i am not sure how to view them.
*.raw_201901230646_UBX.zip (8.0 MB)
raw_201901230744_UBX.zip (1.5 MB)
raw_201901230758_UBX.zip (726.2 KB)
raw_201901230807_UBX.zip (5.0 MB)

(Stefko321) #5

Any ideas?

(Christian Grüner) #6

Ill try take a look at the raw files later. Thanks for uploading.

(Christian Grüner) #7

Can you also provide the survey csv file?

(Stefko321) #8

Yes, here it is:
NoviPazar-CSV.zip (2.7 KB)

(Jurijs Ješkins) #9

You were measuring each point only for 3-5 seconds. I would suggest measuring it for longer time (at least 10 seconds). Points are measured in Novi Pazar, Bulgaria?

(Stefko321) #10

Yes this is correct.

(Christian Grüner) #11

With the right (optimal) conditions, you dont need more than 2-3 secs.

(Jurijs Ješkins) #12

We can try to post process your data, but I can’t find any RINEX data from Bulgaria. Maybe you know any NTRIP caster? Which one were you using?

(Stefko321) #13

I was using 1Yocto network. The interesting thing is that all points are in straight line, meaning that X and Y are ok, even elevations collected by 2019-01-23 08:25:10 UTC are ok even with 2-3 sec. I have previous data collected by a surveying firm with gps, that matches elevations measured with emlid.

(Jurijs Ješkins) #14

They have RINEX only for sale, unfortunately.

(Andrew Yushkevich) #15

Hi @sdobrev85,

Can you share your RINEX base log here, please?

(Stefko321) #16

Hi,
I had the base corrections set for collecting RTCM data. I have attached them in case they can be in any use.
Can you share how to open these type of files?base_201901230646_RTCM3.zip (572.3 KB)
base_201901230807_RTCM3.zip (386.5 KB)

(Andrew Yushkevich) #17

Do I understand right that you expect the height of the soviet17 and so10 to be the same?

(Stefko321) #18

Yes, it is one street, with 2% slope max, the difference should be about 6 cm. And i was on sight, there are no holes this deep.

(Stefko321) #19

The problem with the height starts with point: sovit11, it should be lower then sovit10.

(Andrew Yushkevich) #20

I can assume, that the issue is related to the following factors:

  • As @jurijs.jeshkins mentioned, your point collecting time is a way to short. I’d recommend to increase it to a few minutes
  • Also, your NTRIP station is quite far from the rover. The distance is around 20km, however, the recommended baseline is 8-10km
  • Another reason is that you’ve got strange signal breaks on base logs

base1

base1
base1.png1919×1050 53.4 KB

base2
base3
base3.png1919×1050 46.5 KB

Do you have any plans to do more tests?

(Stefko321) #21

Thanks for the help. I am using Reach Rs+ in my practice, and if i encounter something else which is strange, i will let you know. I have made an initial test when i bought Reach and compared points measured with Trimble R4 and Reach RS, and results were satisfying, but i was using a pole holder and were gathering observations for longer periods (couple of minutes).