EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Nimsch on May 07, 2017, 01:37:34 am
-
Hi everyone!
I am designing a buck converter that would charge a battery and planning to use an N type MOSFET on high side and FAN7085_GF085 driver to go with it.
Datasheet:
https://www.fairchildsemi.com/datasheets/FA/FAN7085_GF085.pdf (https://www.fairchildsemi.com/datasheets/FA/FAN7085_GF085.pdf)
Application notes:
https://www.fairchildsemi.com/application-notes/AN/AN-9052.pdf (https://www.fairchildsemi.com/application-notes/AN/AN-9052.pdf)
https://www.fairchildsemi.com/application-notes/AN/AN-4171.pdf (https://www.fairchildsemi.com/application-notes/AN/AN-4171.pdf)
If the converter happens to go to discontinuous mode however, the diode stops conducting (current through it drops to 0) -> voltage at MOSFET source becomes something >0, and the internal recharge path of FAN7085 activates (or at least that's how I understand it according to the application note 9052, page 2).
My question is what causes the internal switch to activate? A detection of voltage > 0 at VS or something else? (as AN-9052 only states this: "In the event that both of the switch (S1 and S2) are turned off, the internal recharge switch of the FAN7085 will provide the path to charge the bootstrap capacitor.")
Doesn't this connect the battery through an inductor to VS pin of the driver circuit (which is connected to the ground at that moment)? What if I put a diode in series with an inductor (like I did in this capture of the schematic)?
Thank you in advance.
-
Did you check the speed on that TLP185, by the way?
I suppose 100uH might be big enough to handle that, but geez, to get any useful PWM range out of that, you're going to be down in the audible range, or very nearly. I suppose that gives you plenty of time to control this in software, but, geez...
On the upside, you don't need the gate driver IC, or any worries about parasitic reactances in the layout. You've got whole microseconds to burn, while things turn on and off. And the Hall effect sensor will be pretty much real-time, not guesswork.
Tim