Lights cause GPS issue??

[Dronebow]

Hi all. I've had these lights and landing gear for a while now but have until last night never used the lights.

https://www.amazon.com/gp/product/B071W1XY1L/ref=oh_aui_search_detailpage?ie=UTF8&psc=1

I decided to attach them and see how they do at night but very shortly after take off I lost GPS and was having obvious VPS issues as well. See the attached flight log screenshot. This was the second flight and the first was just fine but I did NOT have the lights on it for the first flight.

Shortly after takeoff it started drifting with the wind and I got the notice that GPS was lost and to switch to ATTI mode, which you cannot do on the Mavic manually, but it was in ATTI regardless. I was able to fly it back but while trying to bring it down to land it was moving exceedingly slow. I assume because VPS was thinking it was 1' off the ground. It was stating it was landing so I think the lights were throwing it off.

Just wondering if anyone else has these lights and have experienced similar issues.

 

[Deleted member 35657]

Do NOT put anything aluminum on your drone... Will give you cross signal

 

[Marko9219]

Do NOT put anything aluminum on your drone... Will give you cross signal

What is “cross signal”?

 

[Flycaster]

You really don't want to block the sensors underneath the AC, visual, and sonic. This can/will cause issues, as I have read a couple of folks with the light bracket seeming to block one of the sensors. Also, notice a compass error on you screen shot.
I would take off the lighting hardware, and try again without the lights, and landing gear, test that, make sure all is good, then install the landing gear, and retest. Then add the lights. And if the same location you do get issues with the lights installed, then you have your answer....

 

[Dronebow]

Like I said before. I had already flown this area without them and no issues. I’m pretty sure they were the cause.

My question was more around finding out if others had also experienced the same with these or similar lights. These are not that uncommon and I think many have purchased them.

Thanks

 

[NinjaNut]

Don’t have an answer for you, but that’s a shame, really cool looking and good for night flying.

 

[Callum]

You lost GPS due to a compass error.

Adding anything metallic to the mavic will affect the compass it is closest to and cause compass error as both don't agree.

Mavic sets satellites to zero during compass error as a way of switching into attitude mode.

Check the compass interference in DJI go and you will see the difference with it on and off.

I have a similar issue with a tracker and have to calibrate the mavic compass with it on.
I'm fine with that as I always fly with it so never need to recalibrate.

 

[HWCM]

Do NOT put anything aluminum on your drone... Will give you cross signal

Ahahahaha. This is incorrect. What is the bottom of the Mavic made of?

 

[Mike4884]

Magnesium on the bottom of mavic....so yea he is correct...really aluminum could block the signal or make it irradiate one direction

 

[Brojon]

I doubt the lights have anything to do with it.
The GPS/compass is on top and aluminum has little to no effect on electromagnetic fields.
That said many of the lights using CREE LEDs use PWM as a form of power control and those impulses could radiate outward and possibly interfere.
If you knew the bottom heatsink was connected to power common, you could in theory connect a wire from the light cases to the heatsink and reduce any signal being radiated.

 

[sar104]

You lost GPS due to a compass error.

Adding anything metallic to the mavic will affect the compass it is closest to and cause compass error as both don't agree.

Mavic sets satellites to zero during compass error as a way of switching into attitude mode.

Check the compass interference in DJI go and you will see the difference with it on and off.

I have a similar issue with a tracker and have to calibrate the mavic compass with it on.
I'm fine with that as I always fly with it so never need to recalibrate.

This seems like the most likely explanation. The "Weak GPS signal" message is not really accurate, since it clearly has 17 satellites locked. GPS health, on the other hand, has gone to zero with the switch to ATTI mode, which was almost certainly caused by the compass error and IMU exception. That, in turn, was likely due to the addition of ferromagnetic items to the aircraft. A new compass calibration should fix that as long as the magnetic field due to the new components is not too strong.

 

[Dronebow]

This seems like the most likely explanation. The "Weak GPS signal" message is not really accurate, since it clearly has 17 satellites locked. GPS health, on the other hand, has gone to zero with the switch to ATTI mode, which was almost certainly caused by the compass error and IMU exception. That, in turn, was likely due to the addition of ferromagnetic items to the aircraft. A new compass calibration should fix that as long as the magnetic field due to the new components is not too strong.

There were compass complaints due to the concrete parking lot I was taking off from and I had done a calibration but that calibration is probably a bit off because when you do the calibration it's not sitting on the concrete anymore. Then afterward it's back on the concrete. But I don't think I had the lights attached at that time.

Too bad the lights don't attach to the front legs. I think that would resolve the issue and also rather than have light coming from behind the camera it would be in front.

 

[sar104]

There were compass complaints due to the concrete parking lot I was taking off from and I had done a calibration but that calibration is probably a bit off because when you do the calibration it's not sitting on the concrete anymore. Then afterward it's back on the concrete. But I don't think I had the lights attached at that time.

Too bad the lights don't attach to the front legs. I think that would resolve the issue and also rather than have light coming from behind the camera it would be in front.

If you were taking off from a distorted magnetic field caused by rebar in the concrete then that alone could explain the issues. And you absolutely cannot calibrate to compensate for a magnetically distorted environment - all the compass calibration is doing is subtracting out the magnetic field due to the aircraft itself.

You need to do two things - recalibrate the compass away from any magnetic interference, and then attempt to fly by taking off from a magnetically clean location - not concrete.

 

[CaryTheLabelGuy]

Sub'd

 

[Dronebow]

If you were taking off from a distorted magnetic field caused by rebar in the concrete then that alone could explain the issues. And you absolutely cannot calibrate to compensate for a magnetically distorted environment - all the compass calibration is doing is subtracting out the magnetic field due to the aircraft itself.

You need to do two things - recalibrate the compass away from any magnetic interference, and then attempt to fly by taking off from a magnetically clean location - not concrete.

I've had compass errors before "while sitting on the concrete" but once airborne it will usually clear up ok. And in this case it did, while hovering. But once away from the "lit" concrete of the parking lot where all it had was the darkness and those lights shining, it lost GPS and compass and was drifting in the wind pretty bad. That's why I blame the lights and not the concrete.

I've found that this bird has been much less susceptible to rebar in the concrete than my old Phantom 3 was. It does mess with the compass but once you get airborne and hover for a bit it usually clears up. I may need to do some testing to prove this theory out but it's a bit scary because no one wants their bird to fly away on it's own. That and the fact that from 80' while trying to bring it down it thought it was "landing" the whole time and would only move about 1' per several seconds. The lights were clearly messing with the VPS.

 

[sar104]

I've had compass errors before "while sitting on the concrete" but once airborne it will usually clear up ok. And in this case it did, while hovering. But once away from the "lit" concrete of the parking lot where all it had was the darkness and those lights shining, it lost GPS and compass and was drifting in the wind pretty bad. That's why I blame the lights and not the concrete.

I've found that this bird has been much less susceptible to rebar in the concrete than my old Phantom 3 was. It does mess with the compass but once you get airborne and hover for a bit it usually clears up. I may need to do some testing to prove this theory out but it's a bit scary because no one wants their bird to fly away on it's own. That and the fact that from 80' while trying to bring it down it thought it was "landing" the whole time and would only move about 1' per several seconds. The lights were clearly messing with the VPS.

It does appear that the Mavic FC at least tries harder than the Phantom FC to correct for IMU exceptions after takeoff, so that is consistent with your observation. However - it will likely depend just how bad the magnetic interference is at the takeoff point, and that will vary with location on concrete.

Just to reiterate, it did not lose GPS - it dropped out of P-GPS due to the IMU exceptions that were almost certainly caused by compass errors, so the question is simply what caused those errors. It could have been rebar in the concrete at the takeoff point, ferromagnetic components in the lights (although you seemed to imply that you did a compass calibration after fitting the lights) or (thinking back to @Brojon's comment above), if the lights are drawing significant current that is routed near the compasses, possibly electromagnetically-induced magnetic interference. I would still put my money on the rebar.

The VPS issue may well be due to the lights, but that's a separate problem.

 

[Mossiback]

Have you tried hand launching? It might eliminate the rebar messing with the compass debate.

I like how the message tells you to "please switch to atti mode" when there is no way to do that with the Mavic.

I am surprised you have had no feedback from others that use the same lights.

 

[HWCM]

Aluminum is no magnetic. So, it's not an issue.
Also, the GPS is on top of the mavic and faces up.
If there is an issue, it is with the compass and the electronics inside the flashlight maybe causing magnetic interference. These Les flashlights have driver circuits in them.

Magnesium on the bottom of mavic....so yea he is correct...really aluminum could block the signal or make it irradiate one direction

 

[Brojon]

Aluminum is no magnetic. So, it's not an issue.
Also, the GPS is on top of the mavic and faces up.
If there is an issue, it is with the compass and the electronics inside the flashlight maybe causing magnetic interference. These Les flashlights have driver circuits in them.

Aluminum is still a (poor) conductor so it will have some shielding effect - thus tin-foil hats

 

[BossBob]

There were compass complaints due to the concrete parking lot I was taking off from and I had done a calibration but that calibration is probably a bit off because when you do the calibration it's not sitting on the concrete anymore. Then afterward it's back on the concrete. But I don't think I had the lights attached at that time.

Too bad the lights don't attach to the front legs. I think that would resolve the issue and also rather than have light coming from behind the camera it would be in front.

DJI tell you not to calibrate your Mavic’s compass anywhere near ferromagnetic material. This even includes your mobile phone which will contain tiny amounts, no money in your pocket, even a belt buckle will cause problems. So I’m not surprised that a calibration on a rebarred concrete structure caused problems. Read page 54 of the manual!