differential Opamp Current measurement for low side & high side sensing - help needed

Irving

Joined Jan 30, 2016
3,843
MOSFET is ON if gate-source voltage is > than spec'd Vgs from datasheet charts. (For PMOS > means more negative). this is NOT Vth, the threshold voltage - thats the Vgs where the MOSFET just starts conducting, something like 100 or 250uA, and isn't a useful figure if you're expecting to pass milliamps or amps. As is happens the FDS6575 isn't a bad device, its useful Vgs is -2v though its drain-source resistance only fully meets spec at Vgs = -4.5v.

Your simulations are flawed though because you are assuming your battery is 8.4v which it won't be if it needs charging, it could be as low as 6v. Futher you're assuming your charger is 8.4v, which is won't be iif its a LiPo charger. LiPo is charged at constant current, not constant voltage, so your charger will match your battery volts until it gets to 8.4-ish then switches to constant voltage until charge current is a few milliamps. Depending on the charger it may not even start charging if its not sure if a battery is connected. So in your charging circuit above, with charger disconnected, if Q2 is on then M1 will be conducting. M2 will be off as its source is floating. When you connect the charger the body diode of M2 will conduct and the charger will 'see' the battery volts less 1 diode drop. The charger may start charging with a constant current, its voltage will rise until current starts to flow into the battery. If Q1 is on M2 will start to conduct as its source will rise above its gate voltage and M1 will start to turn off as its drain rises above its source and the body diode starts conducting.

The problem then arises that the charger will think the battery is charged when the battery volts are at 7.7-7.8v as its not aware of the M1 diode in circuit. So the battery is never fully charged. You need to integrate the charger with your battery control circuit. Also, a 2-cell pack needs balancing, so how are you going to accopmplish that with an external charger?
 

Thread Starter

mishra87

Joined Jan 17, 2016
1,034
MOSFET is ON if gate-source voltage is > than spec'd Vgs from datasheet charts. (For PMOS > means more negative). this is NOT Vth, the threshold voltage - thats the Vgs where the MOSFET just starts conducting, something like 100 or 250uA, and isn't a useful figure if you're expecting to pass milliamps or amps. As is happens the FDS6575 isn't a bad device, its useful Vgs is -2v though its drain-source resistance only fully meets spec at Vgs = -4.5v.

Your simulations are flawed though because you are assuming your battery is 8.4v which it won't be if it needs charging, it could be as low as 6v. Futher you're assuming your charger is 8.4v, which is won't be iif its a LiPo charger. LiPo is charged at constant current, not constant voltage, so your charger will match your battery volts until it gets to 8.4-ish then switches to constant voltage until charge current is a few milliamps. Depending on the charger it may not even start charging if its not sure if a battery is connected. So in your charging circuit above, with charger disconnected, if Q2 is on then M1 will be conducting. M2 will be off as its source is floating. When you connect the charger the body diode of M2 will conduct and the charger will 'see' the battery volts less 1 diode drop. The charger may start charging with a constant current, its voltage will rise until current starts to flow into the battery. If Q1 is on M2 will start to conduct as its source will rise above its gate voltage and M1 will start to turn off as its drain rises above its source and the body diode starts conducting.

The problem then arises that the charger will think the battery is charged when the battery volts are at 7.7-7.8v as its not aware of the M1 diode in circuit. So the battery is never fully charged. You need to integrate the charger with your battery control circuit. Also, a 2-cell pack needs balancing, so how are you going to accopmplish that with an external charger?
Hey,
Thanks for reply and time !

All your assumption are correct and VBAT minimum voltage is 3V for Li Ion and the charger is 9.5V CC charger. Lets forget about cell balancing.
Now coming to current flow of above circuit in post #20.
You assumed this incorrectly, The body diode of PMOS is from Drain to Source not Source to drain.

Charging Mode : 9.5V goes to M2 source. if Q1 is ON then M2 ON. Once M2 is ON, through M1 body diode the source voltage reached battery voltage of M1 (or may be battery voltage is already lets say 6V) so if Q2 is ON then M1 is ON and now battery is charging.

Correct me if i am wrong.

Discharging Mode : If Q2 ON , M1 ON. M2 body diode Turn M2 ON if Q2 is ON > Now battery is being discharge through load.

Correct me if i am wrong.
 

Alec_t

Joined Sep 17, 2013
14,280
Your charging circuit has no reliable charge current control. Lithium cells need a proper charging profile for safety reasons, to avoid fires or explosions. If you haven't already done so, check out Battery University for correct charging procedures.
An 8.4V battery can't fully charge another 8.4V battery.
In the interests of safety, consider buying a ready-made charger.
 

Thread Starter

mishra87

Joined Jan 17, 2016
1,034
Your charging circuit has no reliable charge current control. Lithium cells need a proper charging profile for safety reasons, to avoid fires or explosions. If you haven't already done so, check out Battery University for correct charging procedures.
An 8.4V battery can't fully charge another 8.4V battery.
In the interests of safety, consider buying a ready-made charger.
Thanks for your reply !
I wanna say you i have protection BMS to check the battery profile so no issue on That.
I already lets say Charger is 9.5V and battery is Li Ion 3-4.2V x2 = 6-8.4V Charging

I just wanted to understand MOSFET M1 and M2 working for both charging and discharging path. At a time only charging or discharging will be done.

Hope this will clear the requirement. You should forget everything except MOSFT behavior for now.

Thanks !
 

Irving

Joined Jan 30, 2016
3,843
Charging Mode : 9.5V goes to M2 source. if Q1 is ON then M2 ON. Once M2 is ON, through M1 body diode the source voltage reached battery voltage of M1 (or may be battery voltage is already lets say 6V) so if Q2 is ON then M1 is ON and now battery is charging.
Charger wont start at 9.5v - your assumption is wrong. If CC its voltage will only be as high as needed for CC to flow. If battery at 6v charger output will be at 6v. A good charger won't start until it sees a battery. Usually they trickle a little current in and watch for a voltage rise to determine there's a rechargeable battery there. An off the shelf charger must connect to the battery directly, or you must integrate them. You cant just randomly connect a BMS and a charger and expect this to work reliably and safely And if you do have a BMS it has its own series MOSFETS for charge/discharge protection... so how is that going to interact? As @Alec_t rightly said, you have no control over charging profile.

You assumed this incorrectly, The body diode of PMOS is from Drain to Source not Source to drain.
Unless I'm very much mistaken... as you've drawn it...

If 'charger' starts at 9.5v, which it won't if its a proper LiPo charger, AND Q1 on,then M2 conducts, M1 Drain > M1 source so OFF and body diode conducts. Charger switches to CV at 8.4v but battery only at 7.7-7.8v so not fully charged. Charger switches off after 5 or so hours because charge current hasn't dropped to < 10mA because battery isn't able to absorb more charge as the voltage its seeing is too low...
1597695066417.png
 
Top