First a few basics:

• A USB connection is used to transmit data and/or power. Since the Pocket is battery-powered, a connection between it and a phone/tablet should only carry data.
• Power consumption is measured in Watts [W].
• Battery capacity is measured in Watt hours [Wh]. A 10Wh battery can supply 10W of power for one hour, or 1W of power for ten hours, you get the gist.
  
  
  

When using the Pocket with my OnePlus 6T and Samsung Tab S5e, I noticed that they always showed a charging symbol in the top right corner.
First I thought that they must be interpreting the USB connection wrongly, since there's no way that DJI would make it so that the Pocket with its tiny battery would try filling up the huge battery inside my tablet, right?

Well, after I bought a USB Type-C Power Meter (https://www.aliexpress.com/item/32968303350.html) which showed excellent accuracy in tests (https://lygte-info.dk/review/USBmeter%20RD%20Tech%20USB%20Meter%20TC66C%20UK.html), I got to further inverstigate this issue.

As feared, the Pocket was actually charging my devices at between 2.30 and 2.51W. Now, for my devices that was next to nothing, as they normally get charged at 18W, but for the Pocket, that small additional dischage may be catastrophic.
This discharge by the way is constant. It doesn't matter if im recording, if Mimo is even open or what resolution I'm at, or which of my two devices I'm using.

Next I had to put those 2.30-2.51W into context. If the Pocket itself consumes something like 10W of power while recording, then an additional 2W wouldn't be that bad (though still idiotic), only cutting down the overall recording time by ~17% [loss=additional dischage/(self discharge+additional dischage)], but should the power consumption of the Pocket be considerably lower than 10W, then the loss may be devastating.

To test this, I ran my pocket without any connection/accessories from 100% to 0% twice, each with different settings, and then looked at how long the Pocket was able to record.
To calculate the power consumption, I then had to divide the battery capacity of the Pocket, which is 6.738Wh (as seen here (https://forum44.djicdn.com/data/attachment/forum/201901/10/200812k5105nj43y0yyfrr.jpg)), by the recording time in hours.

My testing results can be seen here:


Obviously this was yet another oversight by the programmers. While other attachments like the Controller Wheel probably need to be powered by the Pocket, our smartphones and tablets certainly don't.

dji, please fix this. A reduction in recording time of 55% is just ridiculous.

Sincerely, staticV3

the solution i have chosen to fix this problem is to get ROOT permissions on your device and install this program to disable loading:

DISABLE USB CHARGING (ROOT)

That, imo, is a workaround, not a solution.
Still, thanks a lot for the tip. I appreciate it.

From the beginning, this has been the case with Android devices.

Unfortunately you wasted a whole lot of your time testing, this has zero to do with DJI as google has always implemented this so an accessory cannot  drain the phone battery without the users knowledge.

Rooting your mobile is not advisable regardless of what you are trying to accomplish.   

DJI-Osmo-User-1 Posted at 6-27 05:33
Unfortunately you wasted a whole lot of your time testing, this has zero to do with DJI as google has always implemented this so an accessory cannot  drain the phone battery without the users knowledge.

Rooting your mobile is not advisable regardless of what you are trying to accomplish.

Not necessarily. I have some experience with USB devices on Android and I know that almost all of them will work even when there's no +5V connection between them. That means that USB communication is possible even when you use an external power supply for the USB device.
Knowing that, I tested if I could prevent the discharge by just taping over the +5V contact on the back of the Type-C adapter, which interfaces with the pin on the Pocket. Sadly though, the Pocket will no longer be recognized by my phone with +5V interrupted.
That being said, I do not see a reason for why this bug couldn't be resolved in software/firmware.

Oh and btw, many Android phones will happily supply 1W or more to USB devices without the user's knowledge.

staticV3 Posted at 6-27 05:53
Not necessarily. I have some experience with USB devices on Android and I know that almost all of them will work even when there's no +5V connection between them. That means that USB communication is possible even when you use an external power supply for the USB device.
Knowing that, I tested if I could prevent the discharge by just taping over the +5V contact on the back of the Type-C adapter, which interfaces with the pin on the Pocket. Sadly though, the Pocket will no longer be recognized by my phone with +5V interrupted.
That being said, I do not see a reason for why this bug couldn't be resolved in software/firmware.

Because it’s not a bug, the usb devices that have no battery of course use mobile power, where a battery in a usb peripheral is present then the zener diode allows power transfer.

Cutting or taping the power rail on a cable will not work.

Hm, I still don't understand. If my other USB devices can communicate with my phone with a severed +5V cable, so with only GND, D+, and D-, then why shouldn't the Pocket be able to do the same?
My other peripherals get their power externally through a y-splitter, and the Pocket should get it's power from its own internal battery.

Because it requires a degree of current to complete the more intensive transfer of data at speed. Which is why you must use a data+power usb cable to communicate with the osmo for that mode.
Yes a data only cable will show up as a USB device but not as the correct accessory.

DJI-Osmo-User-1 Posted at 6-28 05:00
Because it requires a degree of current to complete the more intensive transfer of data at speed. Which is why you must use a data+power usb cable to communicate with the osmo for that mode.
Yes a data only cable will show up as a USB device but not as the correct accessory.

Here's my testing regarding this matter:

I used a USB 2.0 card reader (to mimic the Pocket's 2.0 connection) and a microSD card in all tests.

With the android-capable 2.0 card reader plugged straight into my Tab S5e, I got this result:


With the same card reader plugged into my tablet using a y-splitter, where I cut the red wire between the host end and the device end, and then use the power end to supply power to the device, I get this result:


No difference whatsoever...

Next I tested the same thing on Windows 10.

Direct power:


External power:


Again, no difference between the two.

Do you have any official specs, trustworthy sources, or tests which show that an active +5V connection must exist between the host and the device for full data throughput?

Your talking about an entirely different mode now, you were talking about the osmo being connected to your mobile, for that you must use a power+data cable.

DJI-Osmo-User-1 Posted at 6-29 09:07
Your talking about an entirely different mode now, you were talking about the osmo being connected to your mobile, for that you must use a power+data cable.

You still haven't given a solid reason for why Power is necessary between the mobile and the Pocket.
You said that power is necessary to reach the full USB 2.0 speed, which I have disproven with my testing. Having Power between host and device or not having it makes no difference for the data throughput.

DJI-Osmo-User-1 Posted at 6-27 05:33
Unfortunately you wasted a whole lot of your time testing, this has zero to do with DJI as google has always implemented this so an accessory cannot  drain the phone battery without the users knowledge.

Rooting your mobile is not advisable regardless of what you are trying to accomplish.

I had two options:
-Continue with Osmo pocket with low battery due to unwanted charging of android device
-Root a mobile that I only use to record with Osmo Pocket (I don't use it as the main phone)

I decided to choose the second option, this way I can use Osmo pocket with its 100% battery. I use a Xiaomi Redmi Note 5 and it works very well

Hi. I am sorry for the trouble this has caused and thank you for reaching out to DJI Forum. Regarding the DJI Osmo Pocket charging the Android devices. In relation to this matter the Android mobile device needs on-going 500ma to maintain the USB connection, so there is no option for the Android device to disable the charging mode, it is related to the android system and the communication protocol. This is regulated by the Android Open Accessory ( AOA ). Thank you for understanding and support.

DJI Paladin Posted at 6-30 03:04
Hi. I am sorry for the trouble this has caused and thank you for reaching out to DJI Forum. Regarding the DJI Osmo Pocket charging the Android devices. In relation to this matter the Android mobile device needs on-going 500ma to maintain the USB connection, so there is no option for the Android device to disable the charging mode, it is related to the android system and the communication protocol. This is regulated by the Android Open Accessory ( AOA ). Thank you for understanding and support.

Thank you for the response.
I'm curious. Why does the Pocket need that 500mA when all my other USB devices (Card readers, Ethernet adapters, USB hubs, external DAC/Amps) can work without power flowing between them and the phone?

They do NOT, they NEED the current to power the device any device. Unless you have a magic card reader. Your jumping all over the place attempting to prove your correct. But sorry you are wrong. As a side note you already rooted your mobile to disable charging. Are you now saying you factory wiped and reinstalled the stock rom unrooted to continue your efforts?

Didn’t think so.

staticV3 Posted at 6-30 05:20
Thank you for the response.
I'm curious. Why does the Pocket need that 500mA when all my other USB devices (Card readers, Ethernet adapters, USB hubs, external DAC/Amps) can work without power flowing between them and the phone?

As DJI Paladin explained, Google demand it with AOA guide line.

DJI is in a position that is required to follow the guide line. Unless Osmo Pocket would not be certified by google as Android compatible.

If there was an alternative, DJI would certainly have choose to preserve Osmo Pocket’s battery as they could with iOS devices.

It is not DJI but Google who can give you an answer to you curiosity.

Also please share us if you got the answer from Google. Many forum members have been curios about why Google set the Pocket an active role providing power not passively draw power from the phone since the beginning (although we have the speculation) and could have been a better way for Osmo Pocket’s battery sake.

DJI-Osmo-User-1 Posted at 6-30 09:09
They do NOT, they NEED the current to power the device any device. Unless you have a magic card reader. Your jumping all over the place attempting to prove your correct. But sorry you are wrong. As a side note you already rooted your mobile to disable charging. Are you now saying you factory wiped and reinstalled the stock rom unrooted to continue your efforts?

Didn’t think so.

It seems like you are misunderstanding something here as a result of not reading what I'm saying carefully.

Of course my other USB devices need power to function. What I said was "they can work without power flowing between them and the phone". That doesn't mean that no power is flowing at all. I simply interrupted the +5V power line from my phone and am instead using an external 5V power supply to power them.They all work exactly the same no matter where the power is coming from.
That's why I was curious why the Pocket isn't recognized by my phone without the +5V. I assumed that it was unnecessary since, just like my other USB devices function with only an external power supply, the Pocket should, to my logic, also function with only its internal power supply.

"As a side note you already rooted your mobile to disable charging. Are you now saying you factory wiped and reinstalled the stock rom unrooted to continue your efforts?"
One more proof that you need to read more carefully. The person your're referring to is Shescor, not me.. I have not rooted my phone or my tablet. Both are still 100% stock.

Tide Posted at 6-30 18:51
As DJI Paladin explained, Google demand it with AOA guide line.

DJI is in a position that is required to follow the guide line. Unless Osmo Pocket would not be certified by google as Android compatible.

Thanks for your patience in explaining this. Now I understand why the Pocket is charging my phone.
However, I still don't know why a +5V connection specifically between it and the phone is required for the Pocket to be recognized (all my other stuff just needs 5V from any supply. It doesn't need to be powered by the phone directly to be recognized by and usable from the phone). I doubt that Google has the answer to that question ;)

It’s because the osmo pocket exceed the power consumption that most small devices normally use, so it in turn must provide the overhead to the mobile power consumption. It a weird one I am not saying the mobile nor the osmo MUST have the power to function, it’s that google have made it mandatory for such a device. If for example a higher consumption device was connected to your mobile, but idle then in theory when you actually want to use it, your mobile could have a depleted battery in an emergency situation.

If for example a company made a product without any thought to power management (unlikely I know)  the the Mobile’s battery could be depleted without user knowledge, this is why google has implemented the rule.

While it’s a bit of a pain you can still power the osmo via a power bank while being connected to your mobile with minimal fuss provided the power bank is discrete.

I 3D Printed a small clamp with a 1/4 thread that screws into a small wooden handle and balances the rig perfectly, if interested I can provide a photo.

This provideds all day power if required.