Mosfet specs

Kelso

1 µW
Joined
Jan 19, 2022
Messages
3
Location
Cochrane, AB
Hey all,
New member here, learning about a hundred new things at the moment and only have so much time to dedicate to reading up on each so forgive me if this seems like a total noob question or repetitive but I've done as much searching and reading as I can at the moment.
I'm working on my first real E-bike repair and think I'm on the right track but it's led me to some questions such as this for future.

What are desirable specs for MOSFETs to determine a good quality vs poor quality one?

I've had absolutely no help from the "bike manufacturer" of course even just asking if they could supply a new controller (although they did spoil me with this little tidbit of advice "if it still doesn't work, try to turn it on and see if it works." haha! I've determined I had a failed MOSFET in the controller and its a Toshiba TK150E09NE. My local electronics supplier doesn't carry any similar spec MOSFETs (I looked up the data sheet before shopping) so I went to Mouser and was completely overwhelmed lol. My understanding is that when replacing a single MOSFET you need to closely match certain specs such as Rds, Vgs, Vds and Id and so I ordered some Vishay SUP70030E-GE3 which seem close but slightly improved. Anyways this got me wondering if these are even decent quality and if it's worth carrying a bigger supply of a single MOSFET that I could use to upgrade cheap controllers when they fail or if it's a better approach to order them based on specs for each bike needing this repair.

Thanks for any help, I've got a ton of questions and learning to do but I'll hopefully do more learning before posting more questions!
 
Main issue seems to be the sheer quantity of fakes out there- only buy from digikey etc seems to be the way.

Also, having large numbers of parallel mosfets instead of high spec ones seems to be the way to get high current handling with high resilience: even the cheapest controllers offer a dozen mosfets. Finding one that has a quick response and useful throttle curve and then upgrading the mosfets could be a good way to do it.
 
Key specs I would look at (simplified, obviously.... doctoral thesis can and have been written on any one / all of these points):

Vds - Voltage, drain to source: This is how much voltage the mosfet can handle. Obviously you'd want to give yourself some headroom....Vds of 100V for a 72V battery would not be uncommon, etc. Really, the higher the better so long as all other specs are in reason.

Id - Drain current: How much current the mosfet can handle. Again, more headroom = more better and the higher the better all else being equal.

Rds - Resistance, drain to source: The resistance when the mosfet is on. This generally governs how much heat the mosfet produces. Lower is better here. The typical see-saw battle you fight is that higher volt / higher current mosfets typically have higher Rds...so again, balancing all three

Then you get into drive voltage, gate capacitance, temperature rating, etc. Drive voltage obviously has to be compatible with what ever driver you plan to use and the driver has to be able to support the gate capacitance at the frequency you intend to drive. Though this should be fairly closely matched. If you try to drive a 15V mosfet with a 5V low current driver, the mosfet may not fully switch on, or may not be able to switch at the frequency you need which can lead to excessive heating. ...and you probably want the temp rating to be fairly robust. If the thing is going to crap out at 85°C, you could hit that pretty easily on a hot day + a cramped controller layout. You can pretty easily find mofets with 150 - 200°C ratings, though again balancing all other factors above.

Then there are minor factors... the case size/type, cost, availability, etc. Overall, you'll have quite a few points to decide how much is enough and where you're willing to compromise on the quest for the best mosfet.
 
Back
Top