Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Battery charger

Status
Not open for further replies.
I am designing a LiPo Charger. I want to charge cells one by one. And put 1000uF capacitor in parallel with every cell so the cell chan charge even if i charge it only 1/3 time.

So i need to put my power supply (0 to 12V) to every cell (3 cells).
But the problem is 4 wires have this voltages (when full): 0V, 4.2V, 8.4V, 12.6V.
And this is what i want to do: Power supply (0-12V) between first cell (0-4.2V).
Power supply (0-12V) between first cell (4.2V-8.4V).
Power supply (0-12V) between first cell (8.4V-12.6V).

I think the best solution here is to do it with 8 N mosfets. Am i right? And would it work this way?
 

Attachments

  • charger.JPG
    charger.JPG
    108.7 KB · Views: 307
And put 1000uF capacitor in parallel with every cell so the cell can charge even if i charge it only 1/3 time.
A cap in parallel will have the same voltage as the cell, so can't charge the cell (if that's what you mean o_O).
And this is what i want to do: Power supply (0-12V) between first cell (0-4.2V). Power supply (0-12V) between first cell (4.2V-8.4V). Power supply (0-12V) between first cell (8.4V-12.6V).
That makes no sense :(.
 
A cap in parallel will have the same voltage as the cell, so can't charge the cell (if that's what you mean o_O).

Why do you think so? I think i am right. Because you put 12V On 4.2V capacitor. Capacitor charges with that 12V and than current flow from capacitor to 4.2V cell. I tried it with 10 000uF capacitor and it worked.

And how can i "put" 12V on each cell if my schematics doesnt make sense?
 
Why do you think so? I think i am right. Because you put 12V On 4.2V capacitor. Capacitor charges with that 12V and than current flow from capacitor to 4.2V cell. I tried it with 10 000uF capacitor and it worked.

And how can i "put" 12V on each cell if my schematics doesnt make sense?

Sorry, I'm with alec_t - what you say makes no sense.

If the capacitor is in parallel with the cell then you can't charge the capacitor to 12V, as the battery will prevent it going higher than 4.2V or so.

Perhaps it's a language problem?, how about posting a full schematic of what you mean (the partial one above doesn't really help at all - it doesn't show any batteries or capacitors for a start).
 
Your understanding of "in parallel" is obviously different from mine :).
If you charge a cap to 12V (which implies it is not in parallel with a 4V cell at that time) and then discharge it into a 4V cell the amount of charge transferred is (12-4) * 10000 * 10^-6 = 0.08 Coulombs. Compare that to the 3600 Coulombs that a charged 1AHr capacity cell holds!
 
Ok, forget about capacitors then...
Is this ok?
Q1, Q2, and Q3 controls passive ballancing for each cell, but what i worry about is my charging "section". With 8 mosfets. I dont know if that would work that way..
 

Attachments

  • charger2.JPG
    charger2.JPG
    212.4 KB · Views: 332
Last edited:
Bit better, but still rather vague as to the battery connections - are they from Cell0 etc. to Gnd?.

If so you appear to be connecting them directly to 12V to charge, and the bottom FET's short them out destroying everything?.
 
Cell 0 = GND
Cell 1 = 4.2V (When full)
Cell 2 = 8.4V
Cell 3 = 12.6V

I have corrected the schematics (balancing part), so here it is again.

So lets say i want to charge cell 0 to cell 1 (0V to 4.2V). I would turn on Q8 and Q5, and turn off Q4 and Q9. Is this correct?
 

Attachments

  • charger3.JPG
    charger3.JPG
    212.7 KB · Views: 317
The entire scheme looks highly dubious, and EXTREMELY prone to expensive and dangerous mistakes.

Are you not able to charge LiPo in series? - assuming not?, I would STRONGLY suggest you used DPDT relays to split them for charging, and not the dangerous method you're proposing.
 
Please remember that the battery and the circuit can not share a 'ground'-connection. Or any other connection than that temporarily made by by your MOSFETs.
Speaking of which, are you providing sufficient gate-drive to the 'upper' devices? You may need PMOS devices there.
 
Are your cells permanently connected in series?

Edit: There is a serious safety issue with your circuit. You show nothing to control the charging current or monitor it, or to terminate the charging correctly. Connecting 12V across a cell via the FETs will DETONATE the cell :eek:.
 
Last edited:
Please post a video of your battery exploding in flames. Then BUY a balanced charger for your multi-cell Li-Po battery.
 
Cell 0 = GND
Cell 1 = 4.2V (When full)
Cell 2 = 8.4V
Cell 3 = 12.6V

I have corrected the schematics (balancing part), so here it is again.

So lets say i want to charge cell 0 to cell 1 (0V to 4.2V). I would turn on Q8 and Q5, and turn off Q4 and Q9. Is this correct?

Hi,

I think you would do well to learn how to charge just ONE single cell all by itself first before you dive into charging a pack of four cells that are wired in series. You first need to understand how a single cell gets charged before you can move on to a more complicated charger.

The single cell has a max voltage spec of 4.250v and a max charge current 330ma to 1000ma typical depending on the actual manufacturer's recommendation.

Second, when you have to charge cells in series and you have access to the tie points you have to monitor each tie point and provide a current bypass for any cell that gets near the max charge voltage before the others. The current bypass is necessary so that the target cell charges normally too even though the others are still getting the full current.
So for example when a single cell charges it starts at say 1 amp and 3.6v, and then gradually the voltage rises to 4.0, then 4.05, then 4.10, then 4.15, and pretty soon the current starts to drop down to 0.9 amps, 0.8 amps, 0.6 amps, 0.4 amps, down to maybe 0.050 amps at which time you want to stop the charge process for that cell. But the other cells may still be at only 4.0 volts, which means they still need the full 1 amp to charge effectively.
The current bypass will therefore bypass the higher voltage cell so that it does not overcharge, while allowing the remaining cells to get their full charge current. As each cell nears completion (around 4.20 volts) each cell gets 'bypassed' a little at a time until they are all charged and then the charger can shut off completely, probably also lighting an indicator light.
So the bypass circuit might look like a voltage regulator, except that it looks more like a SHUNT regulator than a SERIES regulator.

And just to note, with all of the schematics you have provided so far, they all look like you are attempting to pulse width modulate the cells from 12v down to 4.2 volts, and that is not acceptable at all. That is actually a good way to blow up all of the cells.
The only way to do that is either with a linear circuit or with a buck type regulator. For this kind of circuit though the buck regulator would have to be specially designed such that it is able to sink current as well as source it, and that's not the typical buck.

I would suggest that you at the very least study a few reference designs for this kind of charger, and that you work on charging a single cell first before you go to try doing a charger for a series connected pack. You will be much happier in the end :)
 
UPDATE!
I have tested all parts of this circuit again and i schematics has been completely changed! Also, i have made PCB!
Now i charge my 3S battery in series with 15V power supply. I measure all voltages on each cell, and Q3, Q4, Q5 are intended to discharge certain cells whose voltages are above the lowest cell voltage (thats from passive balancing theory). Discharging current is 0.035A (120 ohm resistor).
Charging is controlled with P-CH mosfet Q1.

I am now in the software part, cell voltage measurements are perfect, and balancing works really good (slow but good)
Charging part is still not completed but i have tested it with youst turning on Q1 and it works! I have to control it with PWM so the current would be constant but i know how to do it..

The problem now is that my PCB is ON when only battery is connected (and not the power supply too). That CAN'T happend. Because my battery will be always connected to this charger. How to stop it form working if the power supply is not connected?
 

Attachments

  • charger3.JPG
    charger3.JPG
    185 KB · Views: 289
You may know how your battery is connected to that circuit, but we don't. Show us.
Cell0, for example, seems to connect both to ground and to R17. Is that right?
 
Yes, it is. But R17 is shunt resistor, and it is "ON" only when uC gives LOW input to gate Q3 mosfet.
R17, R18, R19 are all shunt resistors that are controlled by ATmega
 
Hi,

Most chargers can only source current, not sink it, so you dont usually have to do this. The only current they do sink is a small amount due to the voltage divider that is used to sense the voltage for the voltage feedback, and that isnt that much usually so you dont worry about it.
You can not use a diode here because then the voltage feedback circuit can not sense the actually battery voltage.
Perhaps you can show where you think the current might flow and we can see how much that would be, if anything. In most cases it will be very small like a few milliamps.
 
Yep, thats what i thought, that a diode wouldnt be a good idea.
I have tested my PCB and i really dont understand what was the problem. Then i saw something.
I have put P-CH mosfets to balance part of my board, and every other circuit has N-CH mosfet there. Maybe i thought of using N-CH because i put IRLML2502 as designator, but they were FDN352, P-CH mosfets.

So i tested everything on the breadbord, and i dont see any significant current leaking (>0.001A)
I think that this may be a final version of balanced charger, and now i would be very thankful if someone would see my schematics and say if everything is wired correctly :)
 

Attachments

  • charger4.JPG
    charger4.JPG
    192.8 KB · Views: 310
When Q1 switches on....BANG! You have nothing to monitor or limit the charging current.
If your cells are series-connected then the micro won't be able to switch on Q4 or Q5 fully. Vgs should be >> Vgs(threshold).
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top