Float in the whole flight

Hi my friends
I did 2 flights yesterday, one of which floated 75% and the other 100% of the time marks.

1 Like

Hi, if you take a look at rover files you would see alot of noise.
I would check my power supply and radio interference
Test you drone before takeoff, and turn everything on. If noise is there, try turning/move one thing off at the time and you`ll find the source of this noise

4 Likes

Hi @rdezhalon,

I confirm what’s @TB_RTK saying. You can start with checking the antenna placement requirments. If it’s ok, then you probably need to isolate the receiver’s body from internal electronics.

If you need help with raw data analysis, I’m ready to help!

Hello.
Thanks to @TB_RTK and @ svetlana.nikolenko for the tips.
I achieved 100% fixed results with RTKPOST ver.demo5 b31 and changing the elev mask.
The results were very different.
And I agree with you, there is a noise inside the body and it causes some time marks to float.
Is there a solution to identify the source of the noise?


image
Of course, by changing the mask in flight 2, the fixed results did not get better than 73%.
And that’s confusing to me
image

The power supply is the easiest one to check. Do the receiver share power with anything?
If so, use a separate one and check if the noise is gone.
Radio interference is cumbersome to troubleshoot, lets proceed with that after the battery.

Hi @rdezhalon,

Once again, I agree with @TB_RTK. I’d not say there is one solution for all such cases. It really depends on the hardware setup. But battery and radio are good to start.

Also, I want to suggest Emlid Studio for data processing and analysis. It’s free but optimized for our devices and has a much more straightforward interface :slightly_smiling_face:

1 Like

so what did you change to get 100% fix? very interested as I have the same issues

2 Likes

Hi Joe,

All cases are different. The settings that helped Reza may not help you. If you share your logs, I’ll take a look at them and help to find the best settings :slightly_smiling_face:

Link to files: Stage 1
Stage 7
Base Data
Thanks!

Hi Joe,

Give me some time to check the data and play with the settings, and I’ll share my findings.

Hi Joe,

Sorry for the late reply!

It seems you have an interference issue as well. Here’s the plot showing the data quality (SNR) of GNSS signals:

SNR is low overall, and sometimes the signals are lost completely. Pay attention to the very beginning of the log: all signals are heavily disturbed simultaneously. I’d say the drone was armed there.

So, make sure the antenna and receiver itself are isolated from electronics.

I also noticed you have GPS only enabled on the base. I’d recommend enabling more constellations. It won’t be superfluous.

20220313055143.zip (8.5 MB)
Hello friends
Thanks to svetlana.nikolenko for the answer.
Please look at the attachment file.
At the end of the flight, a number of time marks have become float and single. What could be the reason?

1 Like

The next question is:
In Emlid Studio, in the base section, is there a way to copy and paste the coordinates from a Notepad file into the ECEF coordinates?
Or must it be typed?


Thank you so much

1 Like

Hi Reza,

It seems that the interference issue is still there. You can see spaces in the signals at the end of the log. SNR of L1 signals changes almost every second there, and L2 signals have a lot of cycle slips. Did you manage to find the RF noise source?

In Emlid Studio, in the base section, is there a way to copy and paste the coordinates from a Notepad file into the ECEF coordinates?

I’ve just checked that it’s possible to past values into these fields. Do you have difficulties with that? Do you work on Windows?

1 Like

Hello svetlana
I have not yet found the source of the noise.
I checked the wiring and saw no problem.
That this happens on some flights.
And this time I have a single time mark.
Last time it was just a float.
Is there gammer in the area?

1 Like

Yes it is Windows.
It is not possible to copy and paste with the right and keyboard

Hi Reza,

You can try to isolate Reach’s body with a foil. It should help. But in general, I suggest placing it as far as possible from radio components and a battery.

I’ll try to reproduce this.

I had EXACTLY the same problem foil reduced it dramatically. Which kit are you using?

I’ve just checked Emlid Studio Beta 10 on Windows, and it works fine. Maybe the coordinates are in the wrong format? For example, you use a comma as a separator, while Emlid Studio requires a period.

Hi svetlana
Thank you, my problem was solved.
Yes, I use dots to separate decimals.
To remove mistakes in typing coordinates
I realized that I had to copy and paste the trend part first and then the decimal point