4
\$\begingroup\$

I'm using 2EDL8024G to build a high-side switch, only HO part is used. While this circuit works OK when I PWM it, but the HO seems keep stays on when HI is off (pull to GND).

when VIN=13.8V, VCC=12V, the HO stays at around 8V with fluchuation. Hence it is turning the NMOS on as well but the output is the same as HO.

when a load is connected (while HI is still GND), the voltage is dropped close to 0V and have very small current draw (1mA). I guess it is not fully off either.

How to make it fully off?

enter image description here

\$\endgroup\$
2
  • \$\begingroup\$ How are you measuring it? Are you measuring Vg in respect to ground, or Vgs, i.e. Vg in respect to Vs? How much is Vout at J7? \$\endgroup\$ Commented Feb 18 at 6:21
  • \$\begingroup\$ @Justme Vg to Gnd indeed. J7 Vout = HO when measure it in respect to ground \$\endgroup\$ Commented Feb 18 at 8:56

1 Answer 1

5
\$\begingroup\$

This chip is a bootstrap half bridge driver. From the datasheet:

enter image description here

The low side driver for the bottom MOSFET is powered directly from the chip's power supply.

To provide enough Vgs to turn on the top MOSFET, the high side driver needs a power supply at a higher voltage than the top MOSFET's source. In your case this is done with a bootstrap capacitor (C3 in your schematic).

When the bottom MOSFET turns on (or when the load pulls the output to 0V) the bootstrap cap will charge through the internal diode. Then, to turn on the top MOSFET, first the low side MOSFET is turned off, then the high side driver uses the bootstrap cap as power supply to turn on the top MOSFET. The top MOSFET source then rises to the main supply voltage. Since this cap is bootstrapped to the top MOSFET source, which is its local "ground" for the high side driver, the positive side of this bootstrap supply follows the source and the capacitor provides a supply above VCC to keep the MOSFET on.

However, it's a capacitor, and the chip will draw some current from it, so it will eventually discharge.

When the bootstrap cap is discharged, the top driver simply loses its power supply and stops operating.

Thus, these bootstrap high and low side drivers are not capable of keeping the top MOSFET on continuously. They need to periodically recharge the cap by turning off the top MOSFET, and either turning on the bottom MOSFET or allowing the load to pull SW node to 0V.

Since you don't have a bottom MOSFET, the circuit won't work without a load to pull SW to 0V. With a load present, you can't use 100% duty cycle either, it has to turn off periodically to recharge the cap.

For a high side switch that can remain on continuously, some common choices are :

  • P channel MOSFET with adequate driver: does not require an extra supply above the main supply voltage.

  • N channel MOSFET: in this case the extra supply can be generated with a charge pump or an isolated DC DC converter.

\$\endgroup\$
3
  • \$\begingroup\$ so if a charge pump circuit is implemented the cap can be charged continuously even when HI is off so that the HO Mosfet can be fully turned off? \$\endgroup\$ Commented Feb 18 at 9:02
  • 1
    \$\begingroup\$ Yes, that's correct, if the cap is recharged continuously then the top driver will work without a time limit. There are high-side switch chips with internal charge pump intended for this use case. \$\endgroup\$ Commented Feb 18 at 9:06
  • 1
    \$\begingroup\$ Note your top MOSFET is probably off, and the voltage you observe at the output without a load most likely comes from the high side driver drawing a little bit of supply current, which flows through the bootstrap diode, through the high side driver, then into the SW node. This current will act as a weak pullup for the SW node, so I'm not surprised about your 8 volts. When connecting a load it will pull SW down as you observed. \$\endgroup\$ Commented Feb 18 at 9:09

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.