View Single Post
Old 05-24-2019, 12:08 AM   #38 (permalink)
serialk11r
Master EcoModder
 
Join Date: Jan 2012
Location: United States
Posts: 1,756

spyder2 - '00 Toyota MR2 Spyder
Thanks: 104
Thanked 407 Times in 312 Posts
Old thread but I was thinking about this again today, even though I have no car. I was thinking how I'd hypothetically go about increasing alternator output for minimal effort if I were to change a bunch of accessories on an older car to electric.

I noticed it's pretty easy nowadays to find pretty low RDS_on MOSFETs with synchronous rectifier controller on a board like this: https://www.ebay.com/itm/15A-50A-Ide...e/153280765383

At 1.5mohm, you're looking at about 0.34V drop across two in series, and 40W/6=6.666W heat dissipation at 120A output, which seems manageable with some small heat sinks.

I'm not sure if those switch fast enough though. An alternator at 16000rpm with what, 6, 8 poles? would would switching at about 1000Hz. I kind of ran out of energy looking at MOSFET specs trying to figure out if the gate capacitance would be an issue.

There's also this: https://www.mouser.com/ProductDetail...020DyZ8w%3D%3D

It seems like it can handle 80A at 50% duty cycle square wave, equivalent to 80*sqrt2=113A continuous in terms of heat dissipation. A 150A rated alternator would see a 106A amplitude 50% duty cycle square wave which is below that, and the voltage drop across two of these things would be about 0.6-0.8V, which is not too bad. Since they're less bulky, wiring them up might be easier, but they need more heatsinking.

In the second case, the voltage drop is being cut by a bit over 1V compared to a normal alternator rectifier. That gets probably something like 10A for free, which is just about enough for an electric power steering pump at idle + electric water pump

Last edited by serialk11r; 05-24-2019 at 12:15 AM..
  Reply With Quote