So I got up early to get down to the lakeshore and capture the sunset.  Been to the same spot before.  This is the truncated version.  Lots going on and was engaged in conversation a couple of times during the whole event:
Turned everything on and thought "Doh!  I left the rubber gimbal cover on."  I checked and I hadn't.  Eventually got a picture and took a few quick reference pics and flew a little over the water.  Brought the Mavic back and landed it to change the batter (30% left but try to be super cautious over water).  Turned on the craft again, and no picture again!

IMU took a LONG time to initialize. Even with that it switched into GPS mode and still no image.  Rebooted everything, got the image back, but was too nervous to fly over water.  Callled it a day, landed and tried to pull down the full size images to my phone from Go 4.  Said I had to be connected to the craft to do this, which I was... in GPS mode with IMU calibrated!

Rebooted again and hovered, just to test.  Got a warning about no picture being available and asked it I wanted to RTH.  I was only a few feet away and with the inconsistencies I was experiencing I chose to manually return and land.

Drove home and turned the Mavic on indoors in my office.  Everything worked flawlessly like it usually does.  Downloaded the photos in seconds.

Anyone experienced this?  Don't want it to be a completely wasted trip so I'd like to learn from it.  The two distinct challenges I was seeing seemed to boil down to IMU initialization and streaming the picture, and I didn't think these functions were related?



Basics:
I calibrated the compass at the start

Temps in the upper 50s
Weather was dry
I'd had my first cup of coffee

Why did you calibrate the compass.??

Hadn't been there for a couple of weeks so thought it couldn't hurt.  Calibrated flawlessly.
Here's the connection error:

BuzzCut Posted at 2017-1-22 07:24
Hadn't been there for a couple of weeks so thought it couldn't hurt.  Calibrated flawlessly.

I think that was your first mistake, if you read the Manual p49 clearly says not to calibrate compass unless prompted.
When you put your aircraft on the ground getting ready to take of and you get no warning then just take of your compass is fine.
If you get a warning magnetic etc it doesn't mean you need to calibrate but rather move to a different area to take off, don't worry if your compass needs calibrating you will get that warning.

I take it you have also calibrated IMU at sometime also.

If you have calibrated IMU where and how can have e real effect on IMU initialisation, you should check IMU in your app to see if it needs calibration.
If you calibrate in a warm room and take out to cold air it will take a long time to warm up, most will leave in a cold room for 20/30 minutes and then preform calibration.
Regarding blank screen did you get a warning.

hallmark007 Posted at 2017-1-22 07:50
I think that was your first mistake, if you read the Manual p49 clearly says not to calibrate compass unless prompted.
When you put your aircraft on the ground getting ready to take of and you get no warning then just take of your compass is fine.
If you get a warning magnetic etc it doesn't mean you need to calibrate but rather move to a different area to take off, don't worry if your compass needs calibrating you will get that warning.

No warning for blank screen.  I've never calibrated the IMU as I've never had to.

Given that I'd be stunned if calibrating the compass sent both the IMU and the RC connection into a tizzy.  I'd really like to know why that would be happening.

BuzzCut Posted at 2017-1-22 08:35
No warning for blank screen.  I've never calibrated the IMU as I've never had to.

Given that I'd be stunned if calibrating the compass sent both the IMU and the RC connection into a tizzy.  I'd really like to know why that would be happening.

I don't think calibrating your compass was the cause, I was just pointing out that because you mentioned it.
But if you find your IMU taking longer to warm up, you can check it in the app and as I explained you can do a cold IMU calibration.
I'm sure if you take it outside a couple of times just to check if it's taking longer to warm up.
Whether the blank screen is connected I don't know. Good luck..

What phone are you using?

The screenshot above says GPS mode, but no sats, no bars or anything really, i'd be tempted to pin the blame on the app/phone at the moment.

hallmark007 Posted at 2017-1-22 08:47
I don't think calibrating your compass was the cause, I was just pointing out that because you mentioned it.
But if you find your IMU taking longer to warm up, you can check it in the app and as I explained you can do a cold IMU calibration.
I'm sure if you take it outside a couple of times just to check if it's taking longer to warm up.

I'll look into the cold IMU calibration - thanks for that.

What's weird to me is the combination of what appear to me to be disparate problems.

IMU never errored, but it took longer to initialize.  After it initialized there was still the screen/connection  issue.

Also, when I looked at the battery conditions they were either 'all black' or with red empty bars in the cell visual like the attached.  I never normally look at the battery status, so maybe this status is fine..  The batteries were certainly fine.

So, slow IMU initialization, RC connection issues, and battery status.  All arrived together and seemingly disappeared together too.  I wonder if anyone has experienced something similar

Danny-B- Posted at 2017-1-22 08:55
What phone are you using?

The screenshot above says GPS mode, but no sats, no bars or anything really, i'd be tempted to pin the blame on the app/phone at the moment.

iPhone 6, which has been fine for me before.  I was holding GPS lock, which is interesting.

I assumed the phone GPS wasn't actually used (vs.the Mavic GPS chips)?
RC connection was definitely jacked, as you can see from the photos.  I wonder if this was in general affecting status getting relayed from the craft to the controller/app, whether GPS status, IMU status, battery status, or anything else?  As I type this it seems more and more plausible to me... An RC connection issue which gives the appearance of unrelated non-RC issues...

BuzzCut Posted at 2017-1-22 10:30
I'll look into the cold IMU calibration - thanks for that.

What's weird to me is the combination of what appear to me to be disparate problems.

I don't see where you had issues with RC regarding download it does take a few minutes to download all video and pictures, if you switch off aircraft before pictures are downloaded you may be missing some, but they will be still on the card.

My thinking is your IMU is separate problem to loosing picture, which could have been caused by phone or app as Danny said.

hallmark007 Posted at 2017-1-22 10:53
I don't see where you had issues with RC regarding download it does take a few minutes to download all video and pictures, if you switch off aircraft before pictures are downloaded you may be missing some, but they will be still on the card.

My thinking is your IMU is separate problem to loosing picture, which could have been caused by phone or app as Danny said.

If you think about it the RC connection was completely jacked:

1) The Mavic was not transmitting an image most of the time, hence the black screen
2) I got the warning about no picture being available and was asked if I wanted to RTH that I relayed in my original post
3) I got the on-screen error when trying to download the photos from the craft.  For the avoidance of doubt the craft was ON at this time.

All three of these are RC issues in my mind.  Then, if you take it that there was an RC issue I'm assuming this would interfere with any other information relayed from the craft:
1) IMU status
2) Battery status
3) GPS status
4) Ability to control the camera (e.g. I couldn't rotate the gimbal)

This is why I received plenty of errors related to the RC function but never received errors related to IMU or GPS... because - as far as the Mavic was concerned - there was no issue with these functions

BuzzCut Posted at 2017-1-22 11:20
If you think about it the RC connection was completely jacked:

1) The Mavic was not transmitting an image most of the time, hence the black screen

I think you might be confusing your RC with your app or phone,

hallmark007 Posted at 2017-1-22 11:48
I think you might be confusing your RC with your app or phone,

Maybe I am but I don't believe so.  The RC connection enables the controller to connect to the craft.  There's duplex data transfer:

Out: relays manual stick movements I the craft, camera changes (e.g gimbal movements or camera settings such as focus or white balance adjustment), and initiation of automatic functions that the craft executes (e.g. Tap Fly).

In: Live image, data recorded on the memory card (photos and videos), and craft status (e.g. Motor rpm, craft speed, GPS status, etc.)

So for example, the controller does't know about GPS status or IMU status unless the craft is able to relay its status to the controller, and to the app through the controller.  Neither the app not the controller natively hold status as I understand it... It's all coming from the processors, sensors and other chips on the craft.

So, an RC issue can limit certain functionality (e.g. The ability to see a live view of the craft, or the ability to transfer pics and videos from the craft to the user device) but an RC issue could also limit the ability for any status that's native to the craft (e.g. GPS status) to be relayed to the controller and the app.  The app only knows what it knows because the Mavic tells it.

Hopefully I'm relaying this clearly...

I've lost video display too, using Iphone 7 and Go4 app.
Even though the HD signal bars were showing full strength.
Luckily the Mavic was not to far away so I bought it in closer then let it hover a few metres away in good view.
Whilst hovering I closed the App and also cleared from running in the background expecting to regain video once I restarted the Go4 app. It did't so I landed and packed up for the day.

At home later that day I managed to replicate the fault whilst in the garden and mavic on the ground, I think I found a quick fix to get video back.. All I did was to unplug the USB cable from top left of controller and plug it back in, this was done without leaving the app. After plug was back in the video returned.

On next flight a few days later I lost video again whilst I was about 50m up and 100m away, I did the same thing unplugged and plugged back in, video was back and all good.

I'm guessing this is a Go4 app bug?

BuzzCut Posted at 2017-1-22 12:03
Maybe I am but I don't believe so.  The RC connection enables the controller to connect to the craft.  There's duplex data transfer:

Out: relays manual stick movements I the craft, camera changes (e.g gimbal movements or camera settings such as focus or white balance adjustment), and initiation of automatic functions that the craft executes (e.g. Tap Fly).

Yes but only when connected through the app, so they have to work together. So your not wrong , but your loss of picture looks like it might have been the app.

Have you played flight back to see if you had gps sats at the time of black out.

VAS67 Posted at 2017-1-22 12:23
I've lost video display too, using Iphone 7 and Go4 app.
Even though the HD signal bars were showing full strength.
Luckily the Mavic was not to far away so I bought it in closer then let it hover a few metres away in good view.

I had those video issues to the first time I ever used the Mavic.  Like you I'm guessing there is a bug relating to that but a quick reboot solved the issue.

This differed in that I partially lost data transmission to the craft (e.g. The ability to download pictures from the craft) or the error saying I'd lost picture and should consider RTH.

Love the concept that you know when it's time to pack up rather than pressing your luck.  A very good way to operate!

hallmark007 Posted at 2017-1-22 12:23
Yes but only when connected through the app, so they have to work together. So your not wrong , but your loss of picture looks like it might have been the app.

Have you played flight back to see if you had gps sats at the time of black out.

Yeah, the part that doesn't jive for me is that I didn't just lose picture, but data transmission too.  Data transmission in my humble opinion is exclusively controller to craft, with the app providing the visuals and providing a comprehensive UI to enable the user to send additional commands to the craft (but only ever through the controller in RC mode)

BuzzCut Posted at 2017-1-22 12:41
Yeah, the part that doesn't jive for me is that I didn't just lose picture, but data transmission too.  Data transmission in my humble opinion is exclusively controller to craft, with the app providing the visuals and providing a comprehensive UI to enable the user to send additional commands to the craft (but only ever through the controller in RC mode)

Well now I'm thinking that IMU maybe connected to some of what's going on. I wrote a piece on IMU for my IAA license a year ago I'll have a look and post it,
It may not help but explains connection of IMU  to craft and controller.

hallmark007 Posted at 2017-1-22 13:01
Well now I'm thinking that IMU maybe connected to some of what's going on. I wrote a piece on IMU for my IAA license a year ago I'll have a look and post it,
It may not help but explains connection of IMU  to craft and controller.

Much appreciated.  It'll be of interest regardless!  

BuzzCut Posted at 2017-1-22 13:10
Much appreciated.  It'll be of interest regardless!

IMU "Inertial measurement unit"

As far as I see it, IMU calibration on a level surface updates a table of values the flight controller software uses as a reference for a level stationary hover. From there the craft responds predictably to flight commands.  It also likely measures any sensor noise and thermal drift so that these technical imperfections are accounted for when using IMU sensor data in flight. - this is likely why there must be no vibrations during the calibration process.


Bad IMU calibration could cause drift and attitude issues as the flight controller fights to hold the craft in what it thinks is the correct attitude as opposed to the correct physical attitude.


The IMU usually has 2 types of sensors – angle and acceleration and in turn 3 sensors of each type measuring in the X,Y and Z axis. These sensors can, through vibration, aging, impact etc, drift in their response over time and thus an IMU calibration will establish a new reference for the IMU’s level/stationary state that the flight controller can work with to restore stable flight.


Think of calibration as the bringing back into line the measured craft attitude with the true physical craft attitude.


As to how often an IMU calibration is needed, I wouldn’t like to say; certainly after any impact or if there are suspicions the craft can’t hover in a level attitude in a windless environment. Also after a firmware update is a good opportunity just in case the stored values get corrupted or are incompatible with the later firmware.



Its also good to mention that the IMU is the reason why you are able to shut off the motors using just the throttle down command when on the ground. The IMU reads that the aircraft is unable to move any further, and after three seconds it determines that is has officially landed and can shut off the motors safely.

hallmark007 Posted at 2017-1-22 13:15
IMU "Inertial measurement unit"

As far as I see it, IMU calibration on a level surface updates a table of values the flight controller software uses as a reference for a level stationary hover. From there the craft responds predictably to flight commands.  It also likely measures any sensor noise and thermal drift so that these technical imperfections are accounted for when using IMU sensor data in flight. - this is likely why there must be no vibrations during the calibration process.

That's a really helpful description of the IMU.  That basically confirms that IMU is equivalent to a person's 'equilibrium' and would affect stability etc. but would not affect the comms channel to the drone.

I'll post again one of the screen shots.  Not only is the live stream missing, but all stats that are normally received by the controller from the Mavic are 'zeroed out'.  The next question is why would that happen on a shoreline when the Mavic is 3 feet from the controller.  The only explanations I can think of are either a) a bug; or b) massive interference.