MOSFETs’ Kelvin connection

[@mehdirezai5347]

Thank you professor. All respect from Iran

[@sambenyaakov]

👍🙏

[@hetjoshi2892]

@ 4:55, what if we use a Ferrite bead (instead of resistor) to attenuate high-frequency noise (current)? It would also help by maintaining the same Rg_on and Rg_off.
Thanks in advance

[@esijal]

Insightful presentation. Thanks, dear Professor.

[@sambenyaakov]

So nice of you. Thanks.

[@doronlola1763]

As always, thank-you professor

[@sambenyaakov]

👍🙏

[@hgp8858]

Thank you professor, accurate presentation as always.

[@sambenyaakov]

Thanks. Much appreciated

[@tseckwr3783]

very informative. thanks.

[@sambenyaakov]

Thanks for watching!

[@Chris_Grossman]

Why not have unique floating drivers for each FET and avoid the ground loops?

[@JacquesMartini]

Possible, but more expensive.

[@sambenyaakov]

Cost

[@cls4694]

perfect

[@sambenyaakov]

Thanks

[@ЁбрагимИпатенкоибнАдхарма]

As basic (based) as important. A lot. Thanks!

[@sambenyaakov]

Thanks

[@sanjikaneki6226]

i only saw them in high speed devices like GANfet from what switching frequency does this become a necessary ?
Also wouldnt be easier to simply use a higher drive voltage?

[@JacquesMartini]

The kelvin connection is used in new SiC an GaN FET devices, mostly due to ther insane switching speeds, which generates substantial voltage drops across the source inductance. A higher drive voltage would NOT solve the problem! It would increase driver power dissipation, also modern FETs have lower limits for gate drive voltage.

[@sambenyaakov]

Indeed. Thanks.

[@sXAPOLLls]

But why it is only from source and not from drain?

[@JacquesMartini]

Because the gate voltage is referenced to source, not drain.

[@sambenyaakov]

Indeed. Thanks.

[@sXAPOLLls]

@@JacquesMartini I would agree this works for eliminating PGND issues, but usually I see Rds sensing for current control, where you may want to have full kelvin output.

[@JacquesMartini]

@@sXAPOLLls How many MOSFETs have a kelvin connection on drain? 1% or less?