Mavic Pro Battery Mod
- Thread starter AMars32
- Start date
sprtbkrydr
Well-Known Member
Sorry santa, I mean Wesley.
Are you sure? Everybody has a fear, everyone is scared of something. Just have to figure it out.
This was already covered by me and a few others in another smaller thread that I can't find currently.
Basically, to sum it up it's already known that not all the power from these cells are being used for another reason besides mixing battery technologies. The mavic batteries are the new ones being pushed to the limit, the ones I see more failures of since they charge them to 4.35 per cell and not 4.2 like normal Li-Ion cells.
It's also the restrictions of the battery monitoring that the mavic is currently programmed for. The cutoff voltage is programmed higher then 3V standing, or 2.5V under load.
Theirs a risk of some of the mavics protection not being able to detect over current draw like the error you may see fighting high winds in sports mode with the mavics battery.
DJI also wants this to be easy and doesn't expect people to read the manuals, but also doesn't want a lot of customers claiming a failed battery under warranty even if it's the users fault they made the battery intelligent. The mavics battery is unlike any hobby lithium based battery I have ever seen, it not only has protection for over charging and discharging, but temp monitoring, current detection thought a built in shunt, a microprocessor for firmware with a clock for auto discharging via internal resistors after a few days preventing the battery from damage or swelling for newbies that don't understand these batteries and leave them at a full charge of 4.2V. The battery does the balancing and monitoring of the cells to,ensure it stays balanced, not the charger. This and a little more is built into the battery and all communicates back to the mavic for logging and reporting some errors to the user, or triggering return to home or auto land during a battery failure. But more in depth logs are all kept on an SD card sealed in a shielded can on ge mavics main board that they can use for troubleshooting or to deny a warranty claim. But it's all done from the battery.
Then we have the case of the ESC's not being optimized to run on the 18650 voltage expected range as well. So yes, a rough estimate would be about 30% of each battery is not being used. This isn't optimal by far, but will help the batteries last longer. But now people need to get a hobby charger, and understand they should discharge the cells down to 3.8V per cell when they are not going to fly that day and proper storage of the batteries. So with these battery mods I guess some will have to learn a little more then just open and fly the mavic, but some still won't and wonder why the battery doesn't work next summer after a winters storage.
Thanks for reminding me, I almost forgot to reply to Wesley.Yeah ...I bet you don't.
I missed this one, let's see what I might he able to amswer for you.
Q: 1. Wide variation of tension in the cell versus with what Mavic is expecting to get (NCR has the same issue). 3s? 4s? need to try or open the Mavic and read all the caps and mosfets rating to ensure that it will not burn.
A: Simce the mavics battery pack uses the 3.8 volt nominal cells and has a higher charge of 4.3V per cell. It's nominal voltage is 11.4V unlike the 11.1V of the Titan packs. The 3.8V nominal are not that popular and tend to fail more then the 3.7V nominal cells, I think they are over charging them trying to give people more run time on things and don't care for them myself. I would rather my batteries last longer.
But anyway, since the mavics 3S pack is a higher voltage I domt see why the electronics wouldn't be designed for the higher voltage, running it lower would be ok, but as you know the cutoff will cut off before all the titan pack is used. So I think personally checking the mosfets would be a waist of time if the battery for the mavic is really pitting out 12.9 to 13V at full charge, and I believe it is. This is higher then the full charge of the titan pack at 12.6V
Q 2. What is the battery communication protocol? Has anyone reverse engineered it?
A: This is sort of talked about in this thread.
80% more run time than lipo
I have my input on these pictures being a Arduino Nano, current shunt and voltage boost. Later someone else left comments on the pictures and confirmed this was correct. He is already trying to work out getting the DJIGo4 app to show the proper flight time left, he's is using the Nano to decode current cominication and change it to reflect the difference between the cell voltages. So if you contact him I'm sure he can give you more in the protocol infirmation.
If you can code, and well I might have a inspire battery boards you can use to reprogram as a start to the hardware.if I can't clear the microprocessor then I'll replace it with a blank one and a boot loader, then you can flash your firmware code on it.
Now the biggest difference with this guys project it he want to make it so the mavic runs off 18650 cells only, not mix the mavic battery and a Titan pack with a add on battery mod. I think he shared some of his road blocks later n the thread. It's a much shorter thread then this one.
80% more run time than lipo
Q: 3. How to balance the cells internally cheap? (if possible) - regarding this I don't like the idea of wasting power in resistors. I'm thinking if I can use a capacitor to charge it from once cell and discharge it in another. www.mdpi.com/1996-1073/6/4/2149/pdf
A: Interestimg, I've skimmed over this, never saw this before. Concidering it was something from 1997 my guess is it didn't go that far. It seems to be an active balance ciurcut while the battery is in use by storing poiwer in capacitors and using mosfets to switch between cells and distribute power to cells that are lower then others. I can say this was probably resolved with the betteries being manufactured better having all very close if not identical internal resistance as long as the same cells from the same manufacture are used in the pack (always recommended not to mix different makes and models of cells) therefore allowing an even discharge across the whole pack. And anywhere a resistor is involved is always a low of emergency, it's the basics of a resistor. To resists a certain amount measured in ohms, this is done by displacing the energy into heat. The more power amperage the resistor needs to handle to resist the bigger the resistor needs to be sometimes with heatsincs for cooling. So the heat of the resistor would also add to the heating up the battery pack. And when this balancing idea was written I don't think Li-Ion cells were capable of doing 10 to 20 Amp constant discharge. Even if they did a high discharge spike it would burn up the resistor unless is was massively oversized to handle the possibility of that high current spike.
I hope some of this helps,
Scott
Q: 1. Wide variation of tension in the cell versus with what Mavic is expecting to get (NCR has the same issue). 3s? 4s? need to try or open the Mavic and read all the caps and mosfets rating to ensure that it will not burn.
A: Simce the mavics battery pack uses the 3.8 volt nominal cells and has a higher charge of 4.3V per cell. It's nominal voltage is 11.4V unlike the 11.1V of the Titan packs. The 3.8V nominal are not that popular and tend to fail more then the 3.7V nominal cells, I think they are over charging them trying to give people more run time on things and don't care for them myself. I would rather my batteries last longer.
But anyway, since the mavics 3S pack is a higher voltage I domt see why the electronics wouldn't be designed for the higher voltage, running it lower would be ok, but as you know the cutoff will cut off before all the titan pack is used. So I think personally checking the mosfets would be a waist of time if the battery for the mavic is really pitting out 12.9 to 13V at full charge, and I believe it is. This is higher then the full charge of the titan pack at 12.6V
Q 2. What is the battery communication protocol? Has anyone reverse engineered it?
A: This is sort of talked about in this thread.
80% more run time than lipo
I have my input on these pictures being a Arduino Nano, current shunt and voltage boost. Later someone else left comments on the pictures and confirmed this was correct. He is already trying to work out getting the DJIGo4 app to show the proper flight time left, he's is using the Nano to decode current cominication and change it to reflect the difference between the cell voltages. So if you contact him I'm sure he can give you more in the protocol infirmation.
If you can code, and well I might have a inspire battery boards you can use to reprogram as a start to the hardware.if I can't clear the microprocessor then I'll replace it with a blank one and a boot loader, then you can flash your firmware code on it.
Now the biggest difference with this guys project it he want to make it so the mavic runs off 18650 cells only, not mix the mavic battery and a Titan pack with a add on battery mod. I think he shared some of his road blocks later n the thread. It's a much shorter thread then this one.
80% more run time than lipo
Q: 3. How to balance the cells internally cheap? (if possible) - regarding this I don't like the idea of wasting power in resistors. I'm thinking if I can use a capacitor to charge it from once cell and discharge it in another. www.mdpi.com/1996-1073/6/4/2149/pdf
A: Interestimg, I've skimmed over this, never saw this before. Concidering it was something from 1997 my guess is it didn't go that far. It seems to be an active balance ciurcut while the battery is in use by storing poiwer in capacitors and using mosfets to switch between cells and distribute power to cells that are lower then others. I can say this was probably resolved with the betteries being manufactured better having all very close if not identical internal resistance as long as the same cells from the same manufacture are used in the pack (always recommended not to mix different makes and models of cells) therefore allowing an even discharge across the whole pack. And anywhere a resistor is involved is always a low of emergency, it's the basics of a resistor. To resists a certain amount measured in ohms, this is done by displacing the energy into heat. The more power amperage the resistor needs to handle to resist the bigger the resistor needs to be sometimes with heatsincs for cooling. So the heat of the resistor would also add to the heating up the battery pack. And when this balancing idea was written I don't think Li-Ion cells were capable of doing 10 to 20 Amp constant discharge. Even if they did a high discharge spike it would burn up the resistor unless is was massively oversized to handle the possibility of that high current spike.
I hope some of this helps,
Scott
Oops. It looks that I keep making mistakes
Let's recalculate: 6 batteries x 12.6Watthour = 75.6 Watthour. (note I am being idulgent. The cell cannot store 12.6Watthour, it will store around 11Wh)
How did they get to 120Watts?
Anyway, the battery was removed from their website.
@TiTaN: If you've removesd it because of my posts, please let me know.
Haha, I already questioned this 120W rating. I never figured it out doing the math several ways.Oops. It looks that I keep making mistakes
Let's recalculate: 6 batteries x 12.6Watthour = 75.6 Watthour. (note I am being idulgent. The cell cannot store 12.6Watthour, it will store around 11Wh)
How did they get to 120Watts?
Anyway, the battery was removed from their website.
@TiTaN: If you've removesd it because of my posts, please let me know.
But it's like the tool batteries that claim 20V VS 18V and yet they both run the same number of cells with a 3.7V nominal battery. Since their is no standards arroind how these are tested, reported on a data sheet, calculated with a common load amperage, or any standards at all I think some of it is marketing Mumbo jumbo. Like MegaPixles, or the difference between 720P and 1080i, not that much at all.
After all the world runs on marketing, if they can't sell then they don't stay in buisness.
Screenshot of the Flight Data Viewer, showing the current used.
the end of this flight is captured in this movie:
Not everyone my friend, not Everyone.. And stop calling him Santa.. Cheers.Sorry santa, I mean Wesley.
Are you sure? Everybody has a fear, everyone is scared of something. Just have to figure it out.
Funny you say that as it's the first thing I tried. I had two spare laptop power supplies lying around, the first one just gave power for a second and then died, the second was 19v and the charger detected the extra 1v (it wants 11-18v) and said so on screen so refused to go any further so it was on to a bodge job
and yes it's a 240v lead lol
sprtbkrydr
Well-Known Member
Guys the thread is supposed to be about "battery mods" for the Mavic. I don't know what a 240volt power supply and all this other stuff you're talking about has to do with battery modding a Mavic. If I have missed something please explain. Does anyone else agree?
dirkclod
Well-Known Member
I agree with you so guy's lets stay to the Original Topic and not let this long thread have to be cleaned .
Thanks
sprtbkrydr
Well-Known Member
sprtbkrydr
Well-Known Member
I don't know about the other guys but I had some issues with my bird yesterday and could not do the battery test that I was supposed to do but I'm trying my best to get that sorted out now. I think Maclak is gonna do an antenna test of some sort and report back on that. Right now that's all I know.
djhoodrich
Well-Known Member
- Joined
- Apr 26, 2017
- Messages
- 337
- Reactions
- 134
- Age
- 56
sprtbkrydr
Well-Known Member
Well my bird keeps running away in atty mode so it's hard for me to do anything when it's doing that.
Sorry, just a quick in my defense it does have to do with the new Titan battery mod, well the charger for the titan battery packs. So it is related, just not directly a battery but the charger. Since Titan batteries have entered the mod thread being used for a battery mod they need the proper charger to charge them safely.
But it's not like we are talking about range boosters, signal amplifiers and antennas. It is related to the battery mod.
I hope that clear it up some, or we can make a delegate thread for "Titan" batteries mods with the mavic. Open to recommendations if others agree also. Besides I think the titan battery mod discussion originally started on another thread until someone recommend the titan both merge with this thread.
Thank you,
Scott
sprtbkrydr
Well-Known Member
You can validate yourself by making some more flights with that battery and posting your results. You only posted one pic of one flight. Show us some more of that same setup.
Similar threads
DJI Drone Deals
New Threads
-
-
-
-
-
AIR 3S suddenly begins to ascend during flight- Started by DroneZone Thailand
- Replies: 10
Forum statistics
