I'm searching for the recommended application circuits for series fet configuration . They are noted in the datasheet.
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
I'm searching for the recommended application circuits for series fet configuration . They are noted in the datasheet.
Hi Thorsten,
As you know from the datasheet the bq77908A has a 'floating' push-pull CHG driver which lets it follow the charger voltage. You can see in the electrical characteristics that when the CPCKN reference is left to float above GND that the shutdown supply current can increase. If you are working on a series FET design you are likely aware of the impact of system transients on the driver from the app note www.ti.com/lit/slua612. While the internal driver works well within its capabilities, you may need something more for your application. You will need to select a circuit which meets the needs of your specific application, and test thoroughly. There are many circuits possible.
If you want to keep the floating driver, you could use a pair of diodes to connect the CPCKN reference pin to the lower of GND or PACK-. This way if the charger voltage is perhaps 8V more than the battery for example, the CHG pin goes down with charge FET source and the FET stays off. In the event of a discharge fault where PACK- goes to or above PACK+, you need to either block current back into CHG which could damage the bq77908A and possibly your FET with a diode, or disconnect/switch out the CHG signal in this case. A blocking diode provides an on-only drive with the gate-source resistor turning off the FET, perhaps slowly, but in many cases that may be fine.
Another possible circuit to consider is grounding CPCKN and using CHG as a control signal for an external driver. You might use a single P-ch FET with a diode for an on-only driver, or use additional FETs to also switch off the power FET. See the circuit concept attached. 8422.CHG-drive-on-off.pdf The on-only approach will provide the lowest current. Select resistors to minimize quiescent and shoot-through current while providing adequate switching voltage and speed for your application. Instead of this circuit you can use the now ground-referenced CHG signal to control your favorite push-pull FET driver or level shifting buffer.
Thank you for answering in detail.
WM5295 said:If you want to keep the floating driver, you could use a pair of diodes to connect the CPCKN reference pin to the lower of GND or PACK-. This way if the charger voltage is perhaps 8V more than the battery for example, the CHG pin goes down with charge FET source and the FET stays off.
You mentioned a pair of diodes. Are these ones the diodes D20 and D21 in the app note www.ti.com/lit/slua612 (page 12, figure 16)?
Regards
The diode pair mentioned would be D3 and D4 of the attached concept circuit. These keep CPCKN the lower of GND or PACK-. 5482.CHG-drive-only-diode.pdf The CHG driver can follow PACK- down so that the FET stays off although it is R7 which is turning and keeping the FET off.
D21 in the app note Fig 16 would not be needed here since CPCKN can not go above BAT in this circui. D8 may still be needed depending on the maximum terminal voltage. For example if the system has a max voltage of 32V and the charger transient or system specification indicates it must withstand 200% of normal, 200% of 32 is 64V which exceeds the abs max of the bq77908A. If D8 is used, D20 may be needed to block current into the BAT node as discussed in the app note. In the same 32V pack, if the transient or specification is only 130% this would be only 41.6V which is below the abs max so D8 and D20 are not needed.
D3 and D4 in the above CHG-drive-only-diode diagram would be Schottky diodes with a low forward voltage to keep CPCKN near ground for low shutdown current or near PACK- for low Vgs off for the charge FET. The reverse voltage must be sufficient for the pack transient, perhaps a minimum of 2x the nominal pack voltage.
That should work if it satisfies your system requirement. Since CHG won't be driving a FET, CCAP can be made smaller. Remember that CHG was intended to drive a FET gate, so it should drive a relatively high impedance circuit. Current from CHG will drop its voltage, comes from the battery and will add heat to the IC.
Also realize that CHG is derived from BAT with respect to CPCKN, so there will need to be sufficient voltage between your PACK terminals and the part will have to be awake to get a CHG output.