Charge pump power supply and its application on BMS

I've been too busy recently. When I got home , I just wanted to sleep. I didn't want to open the computer, so I didn't have time to check the messages from many classmates. However, the weather was indeed much cooler, so I slept more comfortably.

This time let’s learn one type of switching power supply: charge pump power supply and its application in BMS .

What is a charge pump?

An explanation from Baidu is: A charge pump, also known as a switched capacitor voltage converter, is a DC-DC converter that uses so-called "flying" or "pumping" capacitors (rather than inductors or transformers) to store energy. DC (converter). The electrical efficiency of the charge pump circuit is very high, about 90-95%, and the circuit is also quite simple. (Pictures from ON official website)

working principle

The principle of the charge pump power supply can be understood by referring to the following figure (picture from TI official website): During the charging phase, switches S1 and S4 are open, switches S2 and S3 are closed; current flows through S2 and S3, and the capacitor CFLY is charged to the voltage VI .

During the discharge phase, switches S1 and S4 are closed, and switches S2 and S3 are open. The negative terminal of CFLY is connected to the power supply VI, so the level at the positive terminal is equal to twice VI; current flows from VI through the capacitor CFLY and switches S1 and S4; the charge is transferred from CFLY to the output capacitor CO to produce an output voltage approximately equal to 2VI .

The above is the form of double voltage output. Another common form of charge pump is negative voltage output (picture from TI official website). First close S1 and S4, and the power supply VI charges the capacitor CFLY; then open S1\S4 and close S2 and S3, at this time, the positive electrode of the capacitor CFLY is connected to GND, and the negative electrode is connected to the load capacitor CO, thus realizing the positive and negative voltage flip, and the external output voltage is -VI.

The difference between the above two forms is the connection of switches S3 and S4. After a little research, you will understand that this is where the charge pump circuit changes, and some charge pump chips have fixed connections of several internal switches. In addition, these two forms are open-loop methods, that is, there is no feedback adjustment of the output. For example, MAX16945, a charge pump chip of MAXIM, has an unregulated output.

There are also many charge pump chips that have feedback loops added inside them. For example, TI's LM2775 has an internal integrated error amplifier, which is used to adjust the internal current source to achieve voltage regulation.

When the charge pump solution is used in situations where the load is not large and a boost or negative voltage is required, the cost and solution complexity are more advantageous.

Application of charge pump in BMS

1. Negative voltage power supply

Some BMS will use operational amplifiers to build computing circuits. At this time, the operational amplifier may require positive and negative power supplies to be powered at the same time. Here, using a charge pump to achieve negative voltage is a better solution. (Pictures come from the Internet)

2. Voltage doubler circuit

Some places on the BMS may require a slightly higher power supply voltage, but the load is very small, such as driving MOS, etc. In this case, the charge pump is a better choice.

One solution is to directly achieve several times of voltage boost through a charge pump chip. For example, the triple boost circuit in the figure below (from ON's NCP1729) can be achieved externally through a simple series and parallel connection of diodes and capacitors; the internal switch of NCP1729 The connection method is to achieve negative voltage output, so if you want to implement a voltage doubling circuit, you need to add an external diode to match. The connection method is to connect the anode of the diode to the power supply, and the capacitor to the switch. The switch can switch between ground and power supply.

There is another solution, which is to use the switch node of the switching power supply instead of using a charge pump chip, as shown in the figure below (picture from TI official website): Based on the BOOST circuit, a diode is connected to the output end, and the switch node SW Connect a capacitor (RFLY in the figure below is to increase the output resistance of the charge pump to limit current).

3. AFE internal power supply

Finally, there is a charge pump circuit inside the AFE that samples the cell cells. The typical representative is MAXIM's MAX178XX series; for example, such a charge pump is integrated in the MAX17854. The output terminal is the HV pin, which is 6.9V higher than DCIN. Left and right, used to power the internal multiplexing strobe switch; the charge pump circuit integrates two switches, one can be connected to ground, the other can be connected to DCIN, the capacitor CCP is CFLY.

Note that the RC outside the HV pin is connected to the DCIN pin and not to ground, because there is no need to double the voltage here, it only needs to be higher than the DCIN level; in the actual application circuit, this resistor is selected to be about 10Ω, which should be The current is limited when the capacitor is charging, otherwise the inside of the AFE may be damaged.

Reference " Pump it up with charge pumps – Part 1 "

Summarize:

I have learned a lot in practice recently, which feels good; all the above are for reference only.