Lost my P3A in the drink!

Josecoelho · Feb 18, 2017

folp said:
Hi all -

I lost my P3A in the Oakland harbor this morning and I don't know why. I was flying a mission with litchi and lost my connection over the harbor. Last thing I saw was what looked like my drone hitting the water... Here's the link to the log:

Phantom Log Viewer - PhantomHelp.com

It looks like it starts going down at 12 min 6.6 seconds. If somebody could take a look and offer some insight as to what happened I would be much obliged

Thanks!

Mike

Hello
Sorry to hear about your drone.
Any chance the drone entered a NO FLY ZONE perhaps a military base or installation designated as NFZ ?
From what I understand in a case where it enters a NFZ it will land where it is .
I have had mine since Xmas and have not had the chance to fly temps in the - 20 C but hey I have already managed to break my antenna on the remote.
Cheers
Jose

folp said:
Hi all -

I lost my P3A in the Oakland harbor this morning and I don't know why. I was flying a mission with litchi and lost my connection over the harbor. Last thing I saw was what looked like my drone hitting the water... Here's the link to the log:

Phantom Log Viewer - PhantomHelp.com

It looks like it starts going down at 12 min 6.6 seconds. If somebody could take a look and offer some insight as to what happened I would be much obliged

Thanks!

Mike

dwallersv · Feb 18, 2017

Wessex said:
@dwallersv thank you for the reply to meta4. As a moderator I was disgusted with his response.
I do know what I'm talking about. I am a licenced radio operator (G4TMF) and an electrical engineer.
If the assumption that the battery is going to shield radio signals from reaching the receiving antenna on the drone then ask yourself why the drone will still receive radio signals inside a building? Not just drones but mobile (cell) phones and SatNav units.

This is really a matter of degree, if you think about it.

If a car were constructed of 2" thick metal, it would be a most excellent Faraday Cage, and you'd never get any RF inside except through the windows. Let's make them lead crystal glass 2" thick too, and you'll be safe in that cabin from any RF intrusion!

If you have 2 P4 batteries, you can perform an experiment in your living room to demonstrate this. Set your cell phone down on a table, check on your location, note that you are getting GPS.

Now place the two batteries side-by-side over the phone and note that you lose GPS pretty quick, and it doesn't come back as long as the batteries are there.

2" of lithium ion electrolyte looks to RF like a 2" thick chunk of conductive metal.

Wessex · Feb 18, 2017

But rf doesn't just come directly from above but from 24-32 satellites. They are not all stacked above one another are they? The receiver antennas are omnidirectional.

Sent from my STV100-4 using PhantomPilots mobile app

Wessex · Feb 18, 2017

I'm not saying there will not be loss of sight of some satellites but it has been established that it was not the craft going into ATTI mode that caused the dive into the drink per se.

Sent from my STV100-4 using PhantomPilots mobile app

Wessex · Feb 18, 2017

Josecoelho said:
Hello
Sorry to hear about your drone.
Any chance the drone entered a NO FLY ZONE perhaps a military base or installation designated as NFZ ?
From what I understand in a case where it enters a NFZ it will land where it is .
I have had mine since Xmas and have not had the chance to fly temps in the - 20 C but hey I have already managed to break my antenna on the remote.
Cheers
Jose

It hit the water at around 53-54mph Jose which points to the craft flying upside down power descending into the water. A controlled landing is around 4mph.

Richard R · Feb 18, 2017

Wessex said:
It hit the water at around 53-54mph Jose which points to the craft flying upside down power descending into the water. A controlled landing is around 4mph.

Makes sense. Consider that the barometer would be reading a decrease in altitude so the controller would be applying power to halt that. Being upside down, probably went to max power and with the aid of gravity ..... Our own version of 'controlled flight into ground'c

Wessex · Feb 18, 2017

Again, I went into my garden to double check the satellite reading given in Litchi on my tablet. I live in the South of the UK Latitude 50.87537000 and to be honest the average satellite count is usually 9-11 in my County (Dorset). Perhaps someone else from the UK could post their satellite count and approximately their location.

So, flying the A/C I was getting 10 satellites showing. Catching the A/C and inverting it revealed that the satellite count remained constant at 9 with just the occasional dip to 8 and back again. Switching off the motors and suspending it upside down and standing away from the A/C showed a satellite reading of 9.

Why Mike's A/C lost 66% of the satellite count I can't answer for sure. However what I am certain about is in my case my A/C only lost 10% which is a big discrepancy to Mike's A/C.

There is an App for the Android phone called GPS Test. It shows a few things of importance to this discussion and the main one that concerns us is the Signal to Noise Ratio (SNR). This shows that there is a big difference between the Signal strength of each satellite received. Although the app shows 22 satellites in view from my location, only 9 are in use.

The question is, does the A/C report the number of satellites it locks onto or the number of satellites it sees but not necessarily of any use due to the low SNR? I would ask DJI directly but from what I've read the technical staff don't seem to eager to give out information of a technical (drone) nature.

Wessex · Feb 18, 2017

dwallersv said:
This is really a matter of degree, if you think about it.

If a car were constructed of 2" thick metal, it would be a most excellent Faraday Cage, and you'd never get any RF inside except through the windows. Let's make them lead crystal glass 2" thick too, and you'll be safe in that cabin from any RF intrusion!

If you have 2 P4 batteries, you can perform an experiment in your living room to demonstrate this. Set your cell phone down on a table, check on your location, note that you are getting GPS.

Now place the two batteries side-by-side over the phone and note that you lose GPS pretty quick, and it doesn't come back as long as the batteries are there.

2" of lithium ion electrolyte looks to RF like a 2" thick chunk of conductive metal.

I have tried what you said placing two P3S batteries on top of my Blackberry STV100-4 (Android Phone) and the readings remained the same. There was no loss of GPS signal.

There is a large gap between the battery and the A/C GPS antenna. R/F radiates from a transmitting antenna and is received by an antenna attached to the GPS unit. The SNR may be weaker but it will not be eliminated. Satellites transmitting from either side of the upside down A/C will still be received.

pomonabill220 · Feb 18, 2017

Meta4 said:
The Phantom GPS antenna has a shield under it to protect the GPS from interference from the internal circuitry of the Phantom.

This was also done on the P2. It is a sheet of foam with two layers of copper film.
This is done so the GPS receiver is shielded from the EMI radiated from the flight controller, and was well documented on the fix for the P2.
I have opened my P3A and the shield is there as well, so wether or not you are an electrical engineer or not (I am an elecTRONICS engineer), until you actually LOOK at what is being discussed, you should refrain from making assumptions.
And I did NOT say the battery is shielding the GPS.

Meta4 · Feb 18, 2017

Wessex said:
Why Mike's A/C lost 66% of the satellite count I can't answer for sure. However what I am certain about is in my case my A/C only lost 10% which is a big discrepancy to Mike's A/C.

I'm going to give up on trying to explain this.
Just watch this : https://photos.smugmug.com/photos/i-Qh67vcf/0/640/i-Qh67vcf-640.mp4

Wessex said:
There is an App for the Android phone called GPS Test. It shows a few things of importance to this discussion and the main one that concerns us is the Signal to Noise Ratio (SNR). This shows that there is a big difference between the Signal strength of each satellite received. Although the app shows 22 satellites in view from my location, only 9 are in use.

You must be using a Standard which does not receive signals from the Glonass sats.

Wessex said:
The question is, does the A/C report the number of satellites it locks onto or the number of satellites it sees but not necessarily of any use due to the low SNR?

Your GPS receiver will track all of the available satellites simultaneously, but only a selection of them will be used to calculate your position.

jwmcgrath · Feb 18, 2017

I never realized how much satellite/GPS/radio/comms schooling I was going to get here!

Sent from my iPhone using PhantomPilots

tevek · Feb 18, 2017

jwmcgrath said:
I never realized how much satellite/GPS/radio/comms schooling I was going to get here!

Sent from my iPhone using PhantomPilots

Yep, same here, FWIW, I took my drone outside and performed the same test as Meta4 and came up with similar results.

microchips · Feb 20, 2017

Just wondering what effect radar waves would have on the gps and associated electronics?

Sent from my iPad using PhantomPilots

Wessex · Feb 20, 2017

Good question Microchips. It would depend on the frequency of the radar pulses. Band L has a similar frequency to GPS but the main question is what caused the drone to flip period I think we've done the GPS Theory to death.

Sent from my STV100-4 using PhantomPilots mobile app

MacPhantom · Feb 20, 2017

My bet would be failure or it collided with something, most likely a bird. Loss of GPS most likely the result not the cause IMHO.

Meta4 · Feb 20, 2017

microchips said:
Just wondering what effect radar waves would have on the gps and associated electronics?

I fly close to and around ships every week without ill effect.
You would have to position your Phantom close in front of a radar antenna to have any effect.

dwallersv · Feb 20, 2017

Wessex said:
But rf doesn't just come directly from above but from 24-32 satellites. They are not all stacked above one another are they? The receiver antennas are omnidirectional.

Yes, but not spherically.

The best way to visualize this is to think of a sphere around the aircraft. Because of the physical layout of the battery, battery cage, and GPS module and antenna, the antenna can only see the top-half sphere. It can not see anything below 0 deg azimuth, due to the battery cage and battery.

This is why when it's upside down, it can't see the sky horizon-to-horizon, and it loses contact with satellites.

Wessex · Feb 20, 2017

No, your argument is flawed. Antenna receives signals from satellites. R/F signals. If what you were saying was true then there would be no R/F getting to the a/c inside a building. The antenna does not have to "see" the satellite to receive it's R/F. The R/F is not directional. It is not transmitted in a narrow beam like satellite TV signals.

Sent from my STV100-4 using PhantomPilots mobile app

rstekeur · Feb 20, 2017

Wessex said:
Lots of replies are guesses. And the antenna is not directional.

If I put my hand over the top of my gps the sats drop in half, so I would guess an upside down phantom with the gps pointing through the system board and lipo that I would expect to see a drop in sats.

Meta4 · Feb 20, 2017

Wessex said:
No, your argument is flawed. Antenna receives signals from satellites. R/F signals. If what you were saying was true then there would be no R/F getting to the a/c inside a building. The antenna does not have to "see" the satellite to receive it's R/F. The R/F is not directional. It is not transmitted in a narrow beam like satellite TV signals.

Any comment on this:
https://photos.smugmug.com/photos/i-Qh67vcf/0/640/i-Qh67vcf-640.mp4