P4P Front and Rear Sensor

Joined
Nov 29, 2015
Messages
77
Reaction score
26
Age
33
I recently had a minor crash because I did not lock a prop down entirely. In any case, I had not flown the drone at all at night until the night of the crash.

The sensors on the front and rear are saying ambient light too weak and are not working, however, the drone seems to make a weird hesitation when flying backwards or forwards and then will go. It also makes a weird noise like the drone is stuttering and then flys normally.

This originally led me to believe there was a motor issue after the crash, but I spoke with DJI and they said it was because the sensors are trying to work so the drone hesitates, then flies due to low light.

Any one else had this issue? I turned the sensors entirely off to test, and the drone did not hesitate or make the same noise, so I am curious if I have a bad sensor or they were designed this way.

Video shows the drone hesitating and the weird noise it makes.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.


Sent from my iPhone using PhantomPilots
 
I did try that as well. Reset the IMU, did a compass recalibration. Basically everything they recommended. I think it seems to not be a good design element, but DJI says it's normal. Curious if anyone else has seen any behavior like this.


Sent from my iPhone using PhantomPilots
 
The forward and rear sensors/cameras do not work properly in low light conditions and is sending conflicting information to the drones ICU causing hesitation. That is normal. Turn them off, sport mode and you will fly just fine.
I agree that it is a bad design, perhaps they should of put a light sensor in line with the sensors of the drone which would turn off forward and rear sensors below a certain lux level and a audible and visual warning on the display screen via the dji go app at the same time.
 
I agree that is what is going on. I spoke with DJI (for the 4th time) and they said that it is a sensor flaw and to request a refund from the dealer as it is malfunctioning. So now I don't know what to think


Sent from my iPhone using PhantomPilots
 
No need for a light sensor -- there are two facing forward, and two facing back :)

Average illumination can easily be determined from an image, and in fact, it will be more accurate for the FOV the cameras are looking at than a separate fixed-point sensor.

So... DJI? Get on the ball! Fix this in the firmware!
 
I guess its like all things in life. Any new product gets put out on the market and its the consumer who pays the price as I feel a lot of things are not fully thought through
 
  • Like
Reactions: zrherda
I guess its like all things in life. Any new product gets put out on the market and its the consumer who pays the price as I feel a lot of things are not fully thought through
I think this a bit harsh, and simplistic.

As an engineer by background but at this point in my career having broad experience across all aspects of product development and release, it just ain't that simple.

Engineering is evolutionary. Thousands of minds (ours) using something (after release) think of things a handful of minds simply didn't, with a myriad other features, issues, requirements, deadlines, marketing window, competition, etc. competing for mental space.

So to say it wasn't "fully thought through" is really a rather meaningless statement, as there is no standard for "fully thought through" against which to evaluate the product. It's a mythical target.

In my opinion, the P4 is a pretty solid, well-engineered product. I'm not surprised there are things that can be improved through some simple software changes. That's the nature of the beast.
 
I wasn't just referring to the p4 but to things in general.
By this I mean new products are brought out and not tested for long enough to see if there is going to be possible issues in the near future.
As an electrician for the past 25 years and family in the motor trade ie mechanic's we have had first hand experience of this very thing I have mentioned. Every body is entitled to there opinion and this is mine.
Obviously,
engineering is evolutionary but as I said before, in the end it is the consumer that pays the price
 
I think it's true that engineering is evolutionary and a few minds aren't going to be able to find every problem that hundreds of thousands will eventually find.

However, this issue and many I see around the forums seem to sometimes be from a complete lack of quality control and testing at DJI.


Sent from my iPhone using PhantomPilots
 
Just to change the subject a little and to prove my point I once had to pay £500 and a further £500 to have a part changed on one of my land rovers because a tiny little value had failed inside a unit that there was no serviceable parts on. The vehicle was just out of warranty with only 7000 miles on it. Why build a unit that costs an arm and a leg when it could of been made to be serviceable and the cost to replace that little valve would of been a few pounds only.
An overwhelming desire to rely on such a week part. Forward thinking out of the window along with a narrow vision.
I don't know what testing they do at dji but I do think like many companies out there its that not enough tests are carried out and its all about profit and less care.
Lets take a look at the forward and rear sensors on the p4p. Did they not test it at night to see what would happen? It states in the manuals that sensors and cameras don't work in low light conditions so if they had tested what would of happened in flight at night they could of altered the product to switch off the sensors and trigger warnings on the system to warn the pilot.
This didn't happen, why? lack of testing or they new about it and sent it out anyway as it would of been an additional cost they didn't want to spend, who knows for sure but dji.
 
To be fair, Land Rovers are at the top of the list as the most expensive to repair based on cost and reliability of any engine manufacturer in the world. I saw this listing on the Ram Eco Diesel forum where the subject of engine reliability was being discussed. Best in class was Honda followed by Toyota and the Honda was better than the Toyota by a factor of 100% :) I know this does nothing to dispel or differ from your theory. Sometimes I wonder if design engineers ever used their own products :) Personally, I think the DJI products are a marvel from a tech standpoint. The biggest shortcomings in the past have been issues with early releases of new models (rushed to market possibly) and poor customer service. I flew a P3 for a year and didn't have a single issue with it, other than some issues with the DJI GO App initially. Each update of the App seemed to make the P3 fly better. I think DJI is a leader in the hardware field but still learning when it comes to some of the software. In an imperfect world, I think that's the way I would prefer it :)
 
Every body is entitled to there opinion and this is mine.
Of course, no need to get defensive! I'll add that in matters like this, there is no right or wrong, only opinion. Yours is as valid as mine.

You see this as an oversight justifying a conclusion that DJI short-changed proper consideration in the design process (unless I've misunderstood what you meant by, "not fully thought-out"). In this particular discussion, your, reason for this is the lack of a light-level sensing feature, as it could improve the functioning of the Visual Positioning System, and make the aircraft safer.

Let me know if I've mistaken any aspect of your thinking on this.

I simply disagree that that is reason for such a harsh judgement. Having been directly involved in product development like this (not drones, but that's irrelevant), I have a perspective on it that I thought might temper your judgement. Tried to explain why.

I see this no more as evidence of sloppy engineering practices as I do a long list of other things the first-generation drone with these systems could, but doesn't, do. As already discussed, you are absolutely free to continue to assert the engineering of the P4 to be not fully thought-out, with no harm or foul.

And I'll continue to hold the opinion that that's a harsh judgement. ;)
 
Hay I'm not getting defensive, well may be a little. This is, after all a discussion site to which everyone is entitled to an opinion so that every one/anyone can come to their own conclusion's on what is being discussed so on that bombshell I will have to agree to disagree with you and leave it at that :D
 
To be fair, Land Rovers are at the top of the list as the most expensive to repair based on cost and reliability of any engine manufacturer in the world. I saw this listing on the Ram Eco Diesel forum where the subject of engine reliability was being discussed. Best in class was Honda followed by Toyota and the Honda was better than the Toyota by a factor of 100% :) I know this does nothing to dispel or differ from your theory. Sometimes I wonder if design engineers ever used their own products :) Personally, I think the DJI products are a marvel from a tech standpoint. The biggest shortcomings in the past have been issues with early releases of new models (rushed to market possibly) and poor customer service. I flew a P3 for a year and didn't have a single issue with it, other than some issues with the DJI GO App initially. Each update of the App seemed to make the P3 fly better. I think DJI is a leader in the hardware field but still learning when it comes to some of the software. In an imperfect world, I think that's the way I would prefer it :)
I agree with you, sometimes things can be rushed to market and this was what I was referring to in my earlier post. I spent a lot of time on research to which drones appeared to be the best on the market and dji "in my opinion" came out on top. As with all things in life there always going to be teething problems and to a certain degree that would be acceptable
 

Members online

No members online now.

Forum statistics

Threads
143,054
Messages
1,467,297
Members
104,919
Latest member
BobDan