-1
\$\begingroup\$

I am a hobbyist and not a native so bear with me. I need a MOSFET that can switch at ~ 100 MHz and withstand up to ~ 40 V and 40 A with a variable duty cycle. If my calculation corrects 1/100 MHz equals 1 Hz per 10 nanoseconds, if the duty cycle is 50%, then 5 ns ON and 5 ns OFF which mean rise and fall time is tight.

So whats specs should i look for? And what are the max values?
Thank you.

\$\endgroup\$
19
  • 2
    \$\begingroup\$ A 1.6 kW project of any kind, let alone one with nanosecond speeds, is a lot for a hobbiest to tackle. At that speed and power level, the gate current alone is going to be several amps. What are you trying to do? \$\endgroup\$
    – AnalogKid
    Commented May 27 at 13:43
  • 2
    \$\begingroup\$ Why 100 MHz and not 100 kHz? \$\endgroup\$
    – Andy aka
    Commented May 27 at 14:09
  • 2
    \$\begingroup\$ The frequency you are attempting to run at is unrealistic for an inverter. Describe what you are attempting to do. \$\endgroup\$ Commented May 27 at 15:16
  • 2
    \$\begingroup\$ In round numbers, you're talking about a 2 kW FM transmitter. That's hard. People are going to keep asking the same question - Why such a high frequency? Separate from that, there is no such thing as a standard transformer at that frequency and power level. That kind of transformer design is at least a month of work for an experienced pro. \$\endgroup\$
    – AnalogKid
    Commented May 27 at 15:47
  • 2
    \$\begingroup\$ That is because they usually are used in linear amplifiers, not switching circuits. Why is your frequency so high? \$\endgroup\$
    – AnalogKid
    Commented May 27 at 17:29

1 Answer 1

1
\$\begingroup\$

To obtain reasonable switching at 100 MHz, you need a transistor that can amplify frequencies at least about 10 times faster. That could be an LDMOS, or a GaN device. The voltage you mention (40V) is no problem, but to find devices for handling 40A of current seems already a bit difficult.

Still you could use what is available, and create a circuit with parallel power stages of course. But that might require complicated power combiners and perhaps a number of expensive power transistors. So the best advise (as already given in the comments) would be to use a lower frequency.

But for completeness: some of the biggest devices, e.g. like the [BLF188], will handle about 1.5kW in linear amplifier mode, which means that in switching operation it can probably handle more! It consists of two FETs with breakdown voltage specified 135V, which can handle about 40A of current (at the specified 50V supply and with 70% efficiency the supply current clearly must be about 40A and is alternately flowing in each of the devices). Thermal resistance seems low enough (since in your your switching operation the efficiency will be higher than the 70% mentioned), but input C of about 600pF might be difficult to drive! (See [datasheet])

\$\endgroup\$

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