I would like to know what is the fastest SiC FET greater than or equal to 10A in a TO-247-3 package. Currently I am using a C2M0280120D. My switching speed is 7 MHz. I want to know if there is a faster FET. Maybe the C3M0280090D would be better?

So, if I understand correctly, what I should be looking for to determine the speed of switching should be: Lower RDS on. Lower input capacitance. Lower rise time? Turn on delay time?

Is that the idea?

  • 2
    \$\begingroup\$ I have here some C4N2020E's for sale, they're the fastest you can buy. Yep they're very expensive. You want them? Of course I am joking. Instead of asking for "the fastest" you should determine what you need. Yes, that is difficult and requires thinking and calculations. You might want to make a simulation model of how you're using the SiC FET, the load etc. Even if you'd have the world's fastest FET does not mean that that FET is the best solution. Indeed Rdson might be more important, it depends on your usage case. \$\endgroup\$ – Bimpelrekkie Feb 19 at 8:54
  • 1
    \$\begingroup\$ What is your reactive load? Ic=CdV/dt= 1nF* 10V/ns= 10 Amps (hah) What EMI can you handle? dI/dt=4100A/us ? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Feb 19 at 9:19
  • 3
    \$\begingroup\$ I am using it as an RF amplifier Do realize that that means that your're basically "on your own". I mean: these SiC FETs are designed to be switching devices and certainly not designed to be used in an RF amplifier. I am not saying that they cannot be used as such. They can (maybe) but it might be a challenge and you might get very crappy performance. For example: a lot of distortion. Can you explain why you're not using a FET that is designed to be used as an RF amplifier? \$\endgroup\$ – Bimpelrekkie Feb 19 at 9:55
  • 3
    \$\begingroup\$ Yes RF MOSFETs are more expensive. It is unclear to me what you're doing and how you're using this MOSFET. If you think that can be 'fixed" by choosing another MOSFET then you have a lot to learn. You should include your schematic. How you use a MOSFET can make or break performance. \$\endgroup\$ – Bimpelrekkie Feb 19 at 10:22
  • 2
    \$\begingroup\$ 7MHz 10A ??? and you want SiC? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Feb 19 at 14:40

The obvious first disclaimer is that you're using these devices in ways that they are not intended, YMMV. That being said, here is how you might compare devices. I think it will probably work well for non-linear amplifier types (class C, E, F), but linearity will be hard. Harmonics are also tricky as well - I've had hard-switching SiC MOSFET circuits with significant frequency components above 200 MHz.

  1. The internal gate resistance on SiC MOSFETs is relatively high. Check this property on the datasheet. Some manufacturers have lower resistances. Using Ciss and Rgate, you can get a first order estimate of how much your input signal will be attenuated.
  2. The low output capacitance of SiC is only really an advantage if you're using them at significant fraction of their rated voltage.
  3. The TO-247-3 is the worst package for a SiC MOSFET. Look for 4-leaded TO-247 packages, or SMT footprints with a Kelvin source connection.
  4. On the topic of packages, your layout is very important. The Crss of SiC MOSFETs is high, so if you aren't careful the device will turn itself on/off. This will look like oscillations on the switching edge.
  5. Ids max isn't the most useful metric for comparing, look at Coss times Rdson at your expected current and expected bus voltage.
  6. SiC MOSFETs need to be driven all the way to the full specified Vgs to get the advertised Rdson, and they also don't really like negative Vgs. You might be seeing issues getting a large enough Vgs on the device turn it on hard. Since the limits on gate voltage are asymmetric, it can also be difficult to turn the device off.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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