Phantom 3 RC LiPo battery repair

Joined
Jul 14, 2015
Messages
536
Reaction score
183
Location
West Yorkshire, UK
Hi All

Thought I might share this snippet for anyone interested, My P3P RC battery failed, at first I thought it was discharged due to leaving it on.. ? anyway it would not accept charge, so I opened up the RC to investigate.

I measured the battey voltage with an AVO, nothing 0 Volts, so I went further, stripped away the outer protective PVC film to expose the small circuit board within the battery casing, measured the voltage at the PCB terminals, Cell 1 = 3.74 volts, Cell 2 = 0V, on closer inspection I found the problem.

Without trying to go into too much detail, the outer shell of the LiPo cell is formed from a foil coated with a conductive resin ( Look on line you will find a better description ). The connection to the circuit board is achieved by bonding a thin strip of this coated foil to a nickel tab which is then solderd to the PCB terminal.
This thin strip is the week point, it had slowly corroded away so that it was no longer attached to the nickel tab.

Having carefully cut away the blue outer PVC sheath and peeling back a strip of the Grey inner protective sheath I exposed the conductive surface of the LiPo cell, now when I measured the voltage between the Common 0v connection and the surface of the cell I got 7.6 volts.

So the only problem with the battery was the foil strip had corroded away, now obviously you cannot solder a new connection to the foil so what I did was to make a small copper foil pad to which I soldered a thin wire, then I applied some Silver Conductive Epoxy glue to the pad ( link to glue info https://www.techsil.co.uk/media/pdf/TDS/MGEP00005-tds.pdf) and stuck the pad to the conductive surface of the Cell.
When it had cured I covered the repair with some gorilla epoxy glue and stuck back the Gray outer sheath that I had cut back to expose the conductive surface, and waited for that to cure.

Job done ( nearly ) I soldered the wire to the PCB connection pad, checked the voltage at the connector, plugged it into the RC, and it works perfectly.

Finally I covered the repair with Duct tape, now the jobs done :)

If you want more info post your question.

PS I would not do this repair to a Flight battery, way to much current drawn, but for an RC it's OK

Here are some photo's


Lipo Repair.jpg
Img_5870.jpg
Img_5871.jpg
 
Nice job, well done, and thanks for sharing. I guess most of us would have just bought a new battery.
I concur, do not muck about with flight battery.
 
I'm thinking about building some batteries for my p4. You can get 3.7v lipo's for less than $2 a piece. Now the p4 battery is 15.2v and wiring four 3.7v cells in series gets you 14.8v. I imagine that the p4 should run on 14.8v. The other option is to wire 5 cells in series to get 18.4. I need to do some research and find out which way to go. Once I decide which voltage to go with I can build multiple cells and parallel wire as many as I can fit together and fill ever bit of that space with battery. A new high capacity p4 battery costs around $170 usd. I can build a battery for $30-$40 and probably double the capacity. I'm seriously considering it.
 
I'm thinking about building some batteries for my p4. You can get 3.7v lipo's for less than $2 a piece. Now the p4 battery is 15.2v and wiring four 3.7v cells in series gets you 14.8v. I imagine that the p4 should run on 14.8v. The other option is to wire 5 cells in series to get 18.4. I need to do some research and find out which way to go. Once I decide which voltage to go with I can build multiple cells and parallel wire as many as I can fit together and fill ever bit of that space with battery. A new high capacity p4 battery costs around $170 usd. I can build a battery for $30-$40 and probably double the capacity. I'm seriously considering it.
Wow, very brave. I think the the P4 may detect a false battery and refuse to play, but keep us posted.
 
  • Like
Reactions: ROD PAINTER
I'm thinking about building some batteries for my p4. You can get 3.7v lipo's for less than $2 a piece. Now the p4 battery is 15.2v and wiring four 3.7v cells in series gets you 14.8v. I imagine that the p4 should run on 14.8v. The other option is to wire 5 cells in series to get 18.4. I need to do some research and find out which way to go. Once I decide which voltage to go with I can build multiple cells and parallel wire as many as I can fit together and fill ever bit of that space with battery. A new high capacity p4 battery costs around $170 usd. I can build a battery for $30-$40 and probably double the capacity. I'm seriously considering it.
I don't know man but you should buy the battery that is supposed to fit in it and not trying to create one yourself. I am to scare to even trying myself to do it. What if the battery died unexpectedly? good bye Phantom 4.
 
Correct Jayfdee, you must include the 'cap' which contains the intelligent portion as well. Without it you have a LiPo paperweight as far as the P4 is concerned.
 
I don't know man but you should buy the battery that is supposed to fit in it and not trying to create one yourself. I am to scare to even trying myself to do it. What if the battery died unexpectedly? good bye Phantom 4.

What if the p4 battery dies unexpectedly? The same thing that happens when a custom made battery dies, you crash. You will still see the voltage and percentage left on the battery. If you wire it properly you should still see the individual cell voltages (depends on how the telemetry is communicated) You won't see the number of cycles but who cares. The p4's OEM battery is more likely to fail than a dumb battery. If the circuitry in the smart battery fails it kills power to the AC even though the battery itself is perfectly fine. The p4 will fly with other than OEM batteries. There are several battery mod videos on YouTube. The only thing I want to do different is build my own battery banks so that I can utilize every square inch of space in that compartment. I've been using these 3.7v lipo's in my toy quads for quite some time and I have never had one fail in anyway and they stay quite cool. Trust me, if I build my own batteries you can bet your *** I'm going to excessively test and abuse them in every conceivable way under controlled conditions. If I wasn't reasonably certain that it wouldn't fail under normal operating conditions then I wouldn't do it. I'm not interested in loosing my quad any more than the rest of you guys. There is no reason myself or anyone else could build a quality custom battery for your quad. You just gotta know what you are doing.
 
Correct Jayfdee, you must include the 'cap' which contains the intelligent portion as well. Without it you have a LiPo paperweight as far as the P4 is concerned.

Not so my friend. There are videos all over YouTube with p4's flying with standard run of the mill dumb batteries. That smart battery is nothing more than a on/off switch, a volt meter, and cycle counter. You can fly without it no problem.
 
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
Maybe it's a semantic issue but sure you can add batts in parallel or replace individual cells but the interface must remain for batt to aircraft communication.
 
Maybe it's a semantic issue but sure you can add batts in parallel or replace individual cells but the interface must remain for batt to aircraft communication.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
Go to the 4:30 mark of this video and you will see a phantom 4 running on an aftermarket battery alone. You clearly do not need the oem battery to power up a p4. That's just not the case.
 
Last edited:
Well I see the OEM battery pack ( white) installed which like I said has the interface in it. I know you can 'sister' other cells in parallel but the interface is a must. Not sure what you're seeing that contradicts this.
 
Well the oem battery is not powered on. I will concede that perhaps the firmware will not let it fly with a battery warning which would be pretty crappy of DJI to force you to buy there battery especially when you won't be able to get these battery's forever. The p4 will power up without having that oem in place if you parallel the aftermarket battery leeds to the oem battery leeds, as soon as you plug in the battery it's coming on, period.
 
And if it won't fly with a battery warning that's ok to. I would still build my own cells and just add Mwh to supliment the oem like you described earlier.
 
Hi All

Thought I might share this snippet for anyone interested, My P3P RC battery failed, at first I thought it was discharged due to leaving it on.. ? anyway it would not accept charge, so I opened up the RC to investigate.

I measured the battey voltage with an AVO, nothing 0 Volts, so I went further, stripped away the outer protective PVC film to expose the small circuit board within the battery casing, measured the voltage at the PCB terminals, Cell 1 = 3.74 volts, Cell 2 = 0V, on closer inspection I found the problem.

Without trying to go into too much detail, the outer shell of the LiPo cell is formed from a foil coated with a conductive resin ( Look on line you will find a better description ). The connection to the circuit board is achieved by bonding a thin strip of this coated foil to a nickel tab which is then solderd to the PCB terminal.
This thin strip is the week point, it had slowly corroded away so that it was no longer attached to the nickel tab.

Having carefully cut away the blue outer PVC sheath and peeling back a strip of the Grey inner protective sheath I exposed the conductive surface of the LiPo cell, now when I measured the voltage between the Common 0v connection and the surface of the cell I got 7.6 volts.

So the only problem with the battery was the foil strip had corroded away, now obviously you cannot solder a new connection to the foil so what I did was to make a small copper foil pad to which I soldered a thin wire, then I applied some Silver Conductive Epoxy glue to the pad ( link to glue info https://www.techsil.co.uk/media/pdf/TDS/MGEP00005-tds.pdf) and stuck the pad to the conductive surface of the Cell.
When it had cured I covered the repair with some gorilla epoxy glue and stuck back the Gray outer sheath that I had cut back to expose the conductive surface, and waited for that to cure.

Job done ( nearly ) I soldered the wire to the PCB connection pad, checked the voltage at the connector, plugged it into the RC, and it works perfectly.

Finally I covered the repair with Duct tape, now the jobs done :)

If you want more info post your question.

PS I would not do this repair to a Flight battery, way to much current drawn, but for an RC it's OK

Here are some photo's


View attachment 81672 View attachment 81673 View attachment 81674
No good repair job is ever completed without duct tape!:)
 
Hi All
Thanks for the comments, I must reiterate that I would not do such a repair on an inteligent flight battery, not even on a dumb one, the current ( 10's of Amps ) drawn would seriously challenge that repair.

Also a replacment battery for the P3P RC here in the UK is non existant, my choice was to buy a second hand RC or send the RC to DJI in the far East for repair, DJI Europe said they could not repair the RC.

OK that aside, what I did discover was this, if you connect a 2 cell LiPo ( 8.2 V ) or a 7 - 8 VDc supply to the RC connector it will not switch the RC on, also using a variable dual output DC power supply, even connecting two separate 3.7 Vdc supplies to the batteries own on board circuit board will, also not switch on the RC.

At the time I did not have 2 single cell 3.7 V batteries so I could not see what would happen If I connected them to the circuit board, my guess is that it would work, where the power supplies did not, and for this reason ( this is only supposition ) the original battery is constructed from 2 x 3.7 v cells connect via the onboard PCB in series with each other, so I think what happens is this:-

The circuit board is always powered by the voltage from the batteries and when the RC is switched on I think that the batteries on board PCB sense's the current drain pulled by the RC and momentarily it switches off a power transistor which forms the junction allowing the series connection of the 2 3.7 v cells, the RC in turn sense's the momentary voltage drop, and recognises the battery as an original device, by this time a timing circuit on the batteries PCB re-establishes the series connection and restores full voltage, I think than the RC looks for a timed momentory voltage dropout from the battery and provided this happens the RC shitches ON.

If someone knows how the RC's battery PCB works I would love to know.

Regards
Waylander
 
Hi..

Did you find out how the PCB works?

I have a similar problem with my P3P radio battery.. contacts corroded. I "stitched" it back with a fine wire. That worked to put the batteries in series again.

However, the protection circuit on the PCB seems to have turned on and stops power from going to the connector. I have correct voltage readings on the battery terminals so the batteries are ok and charged. But no readings on the connector.

i have attached a couple of pictures..

PS.. i used a thermal camera on the PCB and there seems to be hot spot (about 45-48degC) on the PCB near the corroded area. I believe that the protection circuit is activated and it seems to be doing a slow discharge of the cells.

Anyone has a picture or circuit diagram of the PCB on the other side?

corroded contact
IMG_3890a.jpg


IMG_3891a.jpg

edited to add the battery model..
 
LiPo Circuit.jpg
Hi JayW
Find attached a drawing which I will use to give you my take on this problem.

Firstly as my repair has resulted in a stable and usable battery, I am reluctant to now pull it apart for further investigation, so I can't shed any more light on the circuit schematics of on board PCB.

Further to my original post, here are my thoughts:- The battery is constructed of two separate 1 cell LiPo batteries, which in my sketch I name LiPo 1 and Lipo 2, these two batteries are connected to the circuit board as I indicate.
If the battery is healthy, with it disconnected from the RC you should be able to get a Voltage reading of around 7.6 - 8.0 volts at the solder connection tabs of the outgoing connector cable.

Now as I said in my first post, the on board Battery PCB is continuously powered from one or both of the cells ( LiPo 1 and LiPo 2) when the power connector is plug into the RC it presents to the RC's circuit the full voltage of 7.6 Vdc.

As you have probably discovered connecting a 2 cell LiPo to the RC will not allow the RC to switch on, so therefore I believe that some control function occurs on board of the Battery PCB.
I found that the charging circuit from the RC gives a charging voltage to the battery of around 8.0 Vdc so one can assume from this that the two cells LiPo 1 & LiPo 2 are permanently connected in series with each other as the "Normal" or quiescent state, by the solid state device I have termed as Circuit "A" in my drawing.

Now I postulate that the following occurs when the RC is switched on, The batteries on board Monitoring circuit detects the current drawn by the RC, then at some time interval after the initial current draw is detected the device I termed as Circuit "A" switches off for a brief period of time, effectively disconnecting the supply voltage from the RC, before switching back on again and thus restoring the supply.

I think that a circuit in the RC looks for this momentary disconnection, (it may only be a few milliseconds) , and if it detects this quick drop out, accepts the battery as the correct and legitimate type.

If what I think is true, then an alternate design of PCB which does as I say could allow any 2 cell Lipo to be used, but for now if your battery is defunct, then you will loose nothing by connecting two 1 cell LiPo's to the original battery PCB, in the configuration I have shown, if it works, YeHa post back and we will all have learned a bit more.

Good Luck Waylander
 

Attachments

  • LiPo Circuit.pdf
    6.5 KB · Views: 331
I disassembled the battery to take a look at the protection circuit. These are some of the pictures. I notice a couple of blackened spots on the PCB around some via but they could be the exposed copper test points.

IMG_4041small.jpg IMG_4048small.jpgIMG_4050small.jpg


I am starting to think that if the cells are disconnected for any reason, the circuit will turn itself off permanently and stay that way. This is to avoid any potential DIY change of cells.

Just to add.. I connected the transmitter directly to my lab DC supply set at 8.4v. The transmitter turned on and off without any issues.IMG_3931low.jpg

Any thoughts?
 
Hi JayW

At first I also connected the RC to my variable voltage DC supply, I tried at different voltages up to 8.6 volts and the RC illuminated all the LED's on the First press of the On/Off button, but it would not remain On after the second press, I can't remember what version of software was in the RC at the time, or whether that could make a difference or not, but mine would not switch On and stay On when connected to the power supply.

As for the photo's of the PCB, I do see the area's of discoloration you mentioned, but I would de-solder the cables and clean up the solder pads and then have a good look, it may be that some corrosion has occurred but it may not be significant.

I don't quite follow what you said about the cell disconnection, my thoughts are:- one or both of the 1 cell LiPo's permanently power up the batteries on the on-board PCB, the two LiPo's have to be connected in series to give the full voltage, if Circuit "A" in my drawing momentarily opens the series connection of the two cells, the RC's own circuitry could detect that momentary drop-out, if the RC's circuit is looking for this "drop out" at a specific point in time after switch On to verify a "legitimate" battery then I don't see how the circuit would turn itself off permanently, the timed opening and closing of the series connection would be "hard wired" into the solid state devices of the PCB's IC's.

However what I have said is all crap if you have been able to power on the RC from a DC power supply, but not withstanding that, the battery in question is made up from two 3.7 volt 1 cell LiPo's , the RC presents around 8.4 volts to the battery connector lead during charging, therefore the batteries own on-board PCB must handle the "balance charge" requirements to each individual cell, and it must also detect when the cells are fully charged, so to construct your own battery and have it charge up whilst installed in the RC, you will need the batteries own on-board PCB or something that will do the same job.

let me know how you get on if you decide to connect two 1 cell LiPo's to the battery PCB, providing that the PCB is OK I can't see why it would not work...

Good Luck
Waylander
 

Members online

No members online now.

Forum statistics

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