Just been getting some more hours in with the P4RTK and seeing some very disturbing take off behaviour. Everything is dialled in, plenty of satellites, strong RTK lock, I’m not trying to take off too soon, everything is green and good to go.

After initiating take off for a 30m height flight, the aircraft lifts 3 or 4 metres normally, then climbs rapidly at about a 45 degree angle in a heading not consistent with the mission. It then starts to circle at about a 50m radius, for about 300 degrees or so, then settles and goes into the mission as normal.

It has done this three times today fairly consistently.

I suspect it has something to do with the site, as it’s behaved flawlessly everywhere else

Anybody ever seen anything like this?

I did a spectrum analysis on the 2.4Ghz and it’s clear, plus with an uploaded, waypoint mission I don’t see this a control issue, more a Sat issue perhaps?

Ideas? Similar experience anyone?

Does it look kinda like this (move ahead to the 1 min 30 sec mark)?  

Sounds like a compass error to me, perhaps theres metal in the ground or around your??? have ya checked the flight log?
https://phantomhelp.com/logviewer/upload/

This is very similar to what happened to me when i took off from my car roof

Bashy Posted at 6-1 18:40
Sounds like a compass error to me, perhaps theres metal in the ground or around your??? have ya checked the flight log?
https://phantomhelp.com/logviewer/upload/

I’ll take a look at the flight logs next. I’m sure DJI used to claim that RTK avoided (or reduced) all the issues with steel structures? Although I don’t see that anymore on their website. Hmm, maybe there’s a reason for that!
I calibrated the compass before flight, and didn’t see any compass errors on the TX, I was in area with plenty of steelwork so it sounds likely.

nickedw Posted at 6-2 21:26
I’ll take a look at the flight logs next. I’m sure DJI used to claim that RTK avoided (or reduced) all the issues with steel structures? Although I don’t see that anymore on their website. Hmm, maybe there’s a reason for that!
I calibrated the compass before flight, and didn’t see any compass errors on the TX, I was in area with plenty of steelwork so it sounds likely.

Your first mistake was calibrating the compass just before the flight. Contrary to what most people believe, the compass calibration has little to do the the area you are in and is NOT necessary unless there has been some work done on the aircraft. You might post your flight log to here: https://www.phantomhelp.com/LogViewer/Upload/

Then others can take a look at the information.

I had an issue w/ the "toilet bowl" effect right after calibrating right after automatic take-off when it needed me too.  This never happened before or again.  The controls sometimes always feel upon the first say 30 seconds of flight like it's not doing exactly what I want.  I assume the camera gimbal is facing the wrong direction as the take-off performance G-forced the camera into an awkward position.  It settles after though and is ok but probably after 5% of battery was eaten After hovering.

I think the combination of the slightly tall grass overloading the gimbal, the right after calibration, the SETI satellites near me made it confused for a bit so it had to go into slack mode making the threshold gain limits further apart.

Geebax Posted at 6-2 21:35
Your first mistake was calibrating the compass just before the flight. Contrary to what most people believe, the compass calibration has little to do the the area you are in and is NOT necessary unless there has been some work done on the aircraft. You might post your flight log to here: https://www.phantomhelp.com/LogViewer/Upload/

Then others can take a look at the information.

I didn’t make a mistake. The aircraft behaved erratically, I have a clearly defined take off procedure which I’ve used thousands of times in all kinds of hostile environments including moving ships. Operating from steel structures (like ships) it is common for the aircraft to force a compass calibration in order to take off - I don’t do them for fun.
Can I just emphasise this is a phantom4 RTK - it is not the same as a phantom4. I have both, and they are very different, it’s not a cosmetic thing, they have different hardware and completely different software, so if you’re reading this without having seen an RTK don’t assume it’s like a normal phantom.

Getting the logs off an RTK is also slightly more involved and I am in the middle of a very large job with the aircraft currently, so will put them up when I get time.

I had over 7 hours in the air with this yesterday and it performed flawlessly, so what ever the issue was, it was certainly connected with site I believe.

nickedw Posted at 6-4 21:31
Where do I say I calibrated the compass? And I didn’t make a mistake. The aircraft behaved erratically, I have a clearly defined take off procedure which I’ve used thousands of times in all kinds of hostile environments including moving ships. Operating from steel structures (like ships) it is common for the aircraft to force a compass calibration in order to take off - I don’t do them for fun.

Can I just emphasise this is a phantom4 RTK - it is not the same as a phantom4. I have both, and they are very different, it’s not a cosmetic thing, they have different hardware and completely different software, so if you’re reading this without having seen an RTK don’t assume it’s like a normal phantom.

"I calibrated the compass before flight, and didn’t see any compass errors on the TX, I was in area with plenty of steelwork so it sounds likely."

Those are your words I believe. The P4RTK is indeed different to other Phantoms, but the compass is no different. I won't argue with you, since you clearly know better.

So I will leave you with this, should you choose to take notice of it. A quick search on this forum will turn up dozens of situations where taking off from a surface containing ferrous metal caused loss of control of the aircraft. If you take off from ships, then you need to take precautions to avoid the aircraft being influenced by the metal in the ship.  And, as has been pointed out to flyers many times, calibrating the compass has nothing to do with being in the influence of local ferrous metal, it is performed to identify the metal parts of the aircraft itself from the earth's magnetic field. Finally, a common misconception is that the aircraft 'forces' you to calibrate the compass, it does not. The correct message presented to you is telling you to move away from the influence of the nearby magnetic field, not to perform a calibration.

nickedw Posted at 6-4 21:31
Where do I say I calibrated the compass? And I didn’t make a mistake. The aircraft behaved erratically, I have a clearly defined take off procedure which I’ve used thousands of times in all kinds of hostile environments including moving ships. Operating from steel structures (like ships) it is common for the aircraft to force a compass calibration in order to take off - I don’t do them for fun.

Can I just emphasise this is a phantom4 RTK - it is not the same as a phantom4. I have both, and they are very different, it’s not a cosmetic thing, they have different hardware and completely different software, so if you’re reading this without having seen an RTK don’t assume it’s like a normal phantom.

“Operating from steel structures (like ships) it is common for the aircraft to force a compass calibration in order to take off - I don’t do them for fun.”

I’m surprised you don’t have more problems if you calibrate compass beside large steel structures.

Not trying to pile on, but yeah.

In the old days, we didn't get compass warnings on the app.  You launch and watch the lamps and study the behavior of the AC as it hovers before flight.  

The app warning on the more modern systems I think sometimes does more harm than good.  

You have a good compass the previous flight?  In most cases, the worst thing to do is calibrate it before the current flight.  

If you fly from a steel structure without calibrating, AND your compass was good previously minus the steel structure, there's a good chance your AC will behave normally once it clears some distance from the steel - regardless of whatever warnings you were given in the app.  

But when you return, you may have compass problems as you get close.  Makes sense, right?  

I don’t know if you’ve ever actually seen this behaviour, but the aircraft will not lift without a compass calibration. It’s not a warning, it will refuse to take off. I don’t choose to do this, if I want to get the job done, I am forced to do this.
This isn’t recreational flying, it’s paid work, if you want to survey an offshore wind turbine, you need to lift from a metal ship, same thing with a Quay - you get as far from steel as you can, but can’t avoid it.

Generally holding  the aircraft as high as possible while calibrating works. I have never had a compass issue in flight, nor have I seen a compass issue when returning near steel structures.

I’ve done a lot of this stuff with non-RTK all without issue. I regularly operate from quaysides with steel reinforced concrete, railway lines and dockside cranes, I spent all day yesterday doing it, and yes I was forced to do several calibrations - every flight damn near on my Mavic air (which I use as a tape measure for getting the height of the tallest structures for mission planning with the RTK) And never have any problems like the one on Saturday.

I have pulled the logs from the RTK now, which involves putting it in developer mode and scratching around in the file system to find them.

here is the log
https://www.phantomhelp.com/logviewer/SN6QPGECNYZGN5Y29VNU/


it doesn't look correct and doesn't help IMHO  the speeds, for example are absurd and clearly wrong. I suspect this isn't set up for RTK yet.