0
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

Could you please tell me if it is possible to drive a high-side power MOSFET IRL40B209 with a high-speed logic gate optocoupler SFH67XX by tying the optocoupler GND to the MOSFET source.

Thanks

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Add a schematic of your idea. There's a button on the editor toolbar. \$\endgroup\$
    – Transistor
    Commented Mar 23, 2018 at 23:54
  • \$\begingroup\$ If you don't have to drive the MOSFER super fast, you could consider using a photovoltaic gate driver like this: vishay.com/ppg?83469 . It doesn't need an output side power supply. The input side is just a 10ma 1.4Vf LED and the output is ~8.4V of gate-source voltage from the photovoltaic cell \$\endgroup\$
    – crj11
    Commented Mar 24, 2018 at 4:15
  • \$\begingroup\$ There are other ways to do it. There are a lot of bootstrap-based high side drivers for NMOS. In general, they can't operate at 100% duty cycle, though. \$\endgroup\$
    – user57037
    Commented Mar 24, 2018 at 4:16
  • \$\begingroup\$ Why have you chosen to use an NMOS instead of a PMOS on the high-side? \$\endgroup\$
    – Harry Svensson
    Commented Mar 24, 2018 at 8:33
  • 1
    \$\begingroup\$ Simply because in general the N MOSFETs have a best performance compared to the P ones. \$\endgroup\$
    – settimed
    Commented Mar 24, 2018 at 9:29

2 Answers 2

Reset to default
2
\$\begingroup\$

In a nutshell, Yes. Some questions to ask:
1. How is BAT1 generated? Is it an isolated power supply or literally a battery?
2. How fast do you need the switching to be and at what frequency? The output driver of the optocoupler may not be up to the task. If it is just powering a load then it should be fine.
3. If the duty cycle doesn't have to be 100% and galvanic isolation is not a requirement then high-side floating gate drivers are a better bet with a simple bootstrapped power supply - https://www.infineon.com/dgdl/an-978.pdf?fileId=5546d462533600a40153559f7cf21200

\$\endgroup\$
6
  • \$\begingroup\$ The BAT1 will be a battery. The schematic previously shown was just an example, the whole picture is to drive 6 power MOSFETs in a multilevel power inverter in order to build up a 230Vrms at 50Hz from two isolated power supplies. The duty cycle in most cases doesn't exceed 50% (I am using a fundamental frequency modulation not PWM). \$\endgroup\$
    – settimed
    Commented Mar 24, 2018 at 9:39
  • \$\begingroup\$ If that is the case then you'll need a separate battery for each high side MOSFET. Bootstrapped supplies will be constrained at such low switching frequencies but still possible with proper sizing of the capacitor. Based on what you told us, the optomos type drivers like what @crj11 suggested are a good bet. \$\endgroup\$
    – user183368
    Commented Mar 24, 2018 at 15:44
  • \$\begingroup\$ In the subject of optomos gate drivers, I checked 3 different manufacturers' datasheets, for instance, TOSHIBA TLP3906, PANASONIC APV1122 and VISHAY VOM1271, but all of them did not mention the rise/fall time, there are only on/off delay time. Why? What would be the rise/fall time and do they relate to those of the power MOSFET in use? \$\endgroup\$
    – settimed
    Commented Mar 25, 2018 at 22:50
  • \$\begingroup\$ @MohammedSETTI Didn't have time to check all 3 parts but the VOM1271 is given two parameters - on-time and off-time. Unlike a regular floating-mos gate driver, the source/sink currents of a optomos are limited and different. The source/sink current for the VOM1271 depends on the LED forward current and load impedance. When turning on the MOSFET use the Isc(short-circuit) value and FET gate capacitance to compute rise time. This is in addition to the on-time which can be seen as propagation delay from the chart on page 1. \$\endgroup\$
    – user183368
    Commented Mar 26, 2018 at 4:41
  • \$\begingroup\$ Do you think adding a current buffer (e.g SN74LS245N) at the gate driver output pin OR putting it at the end of the MCU I/O port pin (to amplify the LED forward current) will make the turn on fast? Is it a good idea to do that? I would say that because I am planned to use an MSP430 MCU and the current sourcing capability of its GPIO is so limited. \$\endgroup\$
    – settimed
    Commented Mar 27, 2018 at 11:35
0
\$\begingroup\$

you can use isolated dc-dc converter for driving the high-side mosfet. Here is the link for isolated dc-dc converter: https://www.mouser.in/ProductDetail/Murata-Power-Solutions/CRE1S1212SC?qs=pFP7BvE1Pmz5XALT59wZHw%3D%3D

isolated dc-dc converter will give you isolated VCC and GND . To turn on the mosfet a gate voltage above threshold voltage is required with respect to source terminal.

FUNCTION OF DC-DC CONVERTER: consider that the required voltage to turn on the mosfet with respect to source is 12V. If the source already has 10v(floating),then we have to supply 10+12 =22V to the gate in-order to turn on the mosfet. So DC-DC converter will lift up/adjust the "vcc" according to the GND in the output side.And this supply is given to the driver ic/optocoupler . So output pwm will be swithing with this vcc and ground. enter image description here

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.