Auto takeoff height

Any who - the point I'm making is we don't have really any direction of what the issues or true cause is. But people tend to bloat about "do this" to the full and most ideal settings and setup so they can feel like they actually know something. It's always just the same run around and I see it as not really helping but more creating fear and covering up the problem.

But there is and i just gave it to you. That's why i said if you want to shut that group up once and for all just duplicate the flying conditions that i have given you. It is that simple. Until then battery people are going to insist that there was a issue.

Personally i think you would lose the argument and end up losing your drone if you try, considering DJI have also confirmed this particular issue.

All these "ideal" setup you mentioned was just attempt to not match that specific flight condition. And prior FW 1.6/1.06 and during winter season, it was just easier to simply top up battery to full. Then after you fly hard for a bit, the battery temperature would rise and avoided the cold battery flying condition. That's all.

It is often that first initial flight when battery is still cold (20C or lower) that is triggering the faulty logic. Once you have few flights and tge same battery warms up, it is not so much of an issue as you would have avoided that particular combination i gave you.
 
But there is and i just gave it to you. That's why i said if you want to shut that group up once and for all just duplicate the flying conditions that i have given you. It is that simple. Until then battery people are going to insist that there was a issue.

Personally i think you would lose the argument and end up losing your drone if you try, considering DJI have also confirmed this particular issue.

All these "ideal" setup you mentioned was just attempt to not match that specific flight condition. And prior FW 1.6/1.06 and during winter season, it was just easier to simply top up battery to full. Then after you fly hard for a bit, the battery temperature would rise and avoided the cold battery flying condition. That's all.

It is often that first initial flight when battery is still cold (20C or lower) that is triggering the faulty logic. Once you have few flights and tge same battery warms up, it is not so much of an issue as you would have avoided that particular combination i gave you.

Interesting. I posted a thread on here with exact opposite as you described. This was while I was at Lake Tahoe a few weeks ago in 10 degree weather :)

I'm pretty sure I'm prior to 1.6. How can I find out my FW without turning my drone on?

Here's my flight log from the other day.. I'm still mind boggled by it.

HealthyDrones.com - Innovative flight data analysis that matters
 
Awesome. Me too. I just never do that on firmware 1.5 with a 50% battery :)
Am I doing it wrong then? I always start the props... and let it warm up for a few seconds on the ground before I take off. Am I the only one doing this? I usually get it moving after it is about 98%.... Too conservative?
 
let it warm up for a few seconds on the ground before I take off
There is no need to do that. In cold weather, it's a good idea to hover for about a minute after you take off to let the battery warm up.
 
Interesting. I posted a thread on here with exact opposite as you described. This was while I was at Lake Tahoe a few weeks ago in 10 degree weather :)

I'm pretty sure I'm prior to 1.6. How can I find out my FW without turning my drone on?

Here's my flight log from the other day.. I'm still mind boggled by it.

HealthyDrones.com - Innovative flight data analysis that matters
The temperature won't be an issue if you are not on FW 1.5/1.05.

You can check your FW version in Go app. Go to About.
 
Here is the exact issue as far as i can gather:

DJI battery shows a voltage drop that is a function of battery temperature. In other words, the degree/the amount of voltage drop under heavy load on these battery depends on the battery temperature. (There was an YouTube video earlier demonstrating this). On a cold battery, the drop is over 1V.

That by itself is not enough to cause issue because you still have power to power the drone.

The version 1.5 / 1.05 FW has logic build in to protect the battery. Basically when battery reach critical voltage (less than 3V on any one cell) it would power off the battery.

Again, this on its own isn't an issue.

However when you fly with a less than full battery say 3.99V in one of the cell. And your battery temperature is under 20C, and you place it under heavy load, this will cause the battery reading for a cell to dip below the 3V critical level. And the FW battery preservation routine would kick in shutting down the battery in mid air.
 
The version 1.5 / 1.05 FW has logic build in to protect the battery. Basically when battery reach critical voltage (less than 3V on any one cell) it would power off the battery.
DJI has never explained this, so we cannot be certain. But, I think the firmware installed on the Phantom is not reading the battery voltage correctly. It seems to think it's higher than it really is -- which allows the pilot to continue flying until the battery reaches 3.0V. I don't believe the Phantom firmware is shutting down the battery. The battery just naturally shuts itself off when the first cell reaches 3.0V.

That's my interpretation of the situation :)
 
DJI has never explained this, so we cannot be certain. But, I think the firmware installed on the Phantom is not reading the battery voltage correctly. It seems to think it's higher than it really is -- which allows the pilot to continue flying until the battery reaches 3.0V. I don't believe the Phantom firmware is shutting down the battery. The battery just naturally shuts itself off when the first cell reaches 3.0V.

That's my interpretation of the situation :)

I am under the impression that this condition was discovered by a user community and was able to consistently reproduce the behavior to DJI thus allowing DJI to narrow in on the problem and release a fix in 1.06/1.6 FW. I read that on one of the thread on here that then reference to another thread on DJI forum. Obviously, I have been trying to re-locate that thread since as reference but am unable to.

So you are right that DJI never explained this.
 
DJI decreased the acceleration (when needed) to prevent the rapid drop in the battery voltage. I'm assuming that's why they did it anyhow. Again, they did not really explain. People no doubt performed the same type of flights on a low battery prior to firmware 1.5, so I'm thinking they fixed something else too -- and, just did not mention it.
 
DJI has never explained this, so we cannot be certain. But, I think the firmware installed on the Phantom is not reading the battery voltage correctly. It seems to think it's higher than it really is -- which allows the pilot to continue flying until the battery reaches 3.0V. I don't believe the Phantom firmware is shutting down the battery. The battery just naturally shuts itself off when the first cell reaches 3.0V.
That's my interpretation of the situation :)
The explanation I got from dji at ces was the 1.5 phantom 3A/P firmware reads the voltage lower than reality. The display shows the right voltage but the firmware reads it lower, hence battery protection kicks in way before it needs to, way before a cell reaches the critical 3.0 voltage level in reality. Since the firmware falsely thinks it's @ 3.0 it starts auto land, or in worst cases below 3.0 it shuts off all power.

Dji claims (in my conversation at ces) this only exists in 1.5, however several here claim this happened to them with 1.4 also. I used 1.4 for a couple months, but I always charged up right before flying, never start a new flight on a partial charged battery, and never had an issue.

I'm so concerned about battery shutdown that I'm running 1.6 and just accept the fact my drone is slower at acceleration than before.
 
Last edited:
hence battery protection kicks in way before it needs to, way before a cell reaches the critical 3.0 voltage level in reality
In 1.5, it seems the battery protection is kicking in way later than it should. Otherwise, it seems the battery would not be getting to the point where it shuts itself off. No?
 
In 1.5, it seems the battery protection is kicking in way later than it should. Otherwise, it seems the battery would not be getting to the point where it shuts itself off. No?
The way DJI explained it is the voltage reading on our display is correct in 1.5. However the firmware thinks the voltage is lower than reality for some reason, I presume a coding error. So when it appears the battery has an acceptable level of voltage in the Go app, and we see at 3.7 or 3.6V per cell (~ 60% to 50% on the slider), and the battery really does have an acceptable level to fly, the firmware is seeing something much different, falsely reading the level at 3.0 or 2.9V, therefore it kicks in the battery protection too early because it thinks the battery is lower than it is in reality. High throttle accelerations at this mid level charge is what I think triggers auto land or shut down. Others that don't aggressively accelerate after 50% battery level don't see the problem, however some have reported higher battery levels in the 60 and 70% range with sudden shut down.

From this discussion at CES, I surmise the the power shutoff is not controlled by the battery firmware, it's controlled by the Phantom motherboard firmware. This makes sense because the motherboard has to trigger auto-land when it reads a unacceptable voltage level.

It's my belief that when aggressive power applied (full throttle vertical ascent) at less than 50%, this will bring the voltage down to a point that the 1.5 firmware "THINKS" the voltage is below 3.0V, but in reality it's at 3.5V. Cold ambient temps might contribute to voltage level drops at full throttle.

As for the battery firmware update in 1.6, I believe this was an implementation of a governor, prohibiting full power to be pulled from the battery once a cell touches 3.6V. I've seen this voltage level trigger my full throttle vertical ascents to go from 10-11mph down to 7-8mph once any cell reaches 3.6V. I have more testing to do on this. As far as I'm concerned it's unwise to run 1.5, and from what others are saying, 1.4 also. I'm staying with 1.6 regardless of the downside of the NFZ geofencing. It sucks, but that's life.

I don't think we'll ever know the exact details of this because DJI isn't going to explain this to the world. That would put them in a high liability situation depending on what incident happens. These DJI guys probably told me too much as it is. They said this was only an issue with 1.5, but others in this forum have reported this behavior in 1.4 also.

Both guys I spoke to individually seemed to be in the know on this subject. Although one was not an engineer, he was in a management position in the customer service dept and spoke very confidently about the issue, and it was something that seemed like he had been dealing with a lot lately.
 
Last edited:

Members online

Forum statistics

Threads
143,066
Messages
1,467,359
Members
104,936
Latest member
hirehackers