0
\$\begingroup\$

I'm designing a buck-boost which needs to provide 5A @ 3.6V, and work from an input varying from 2.8 - 5V.

After having a look at the different non-inverting topologies, it looks like the relatively conventional 2/4-Switch Buck-Boost is the way to go. I can't find any parts with integrated FETs at this load current, so it looks like it will be Controller + External Switches. My question is regarding Max Output Current. Is this independent of a controller with external Switches, Inductor etc?

This value is rarely specified in relation to Controllers, and yet things like Gate Drive Strength are specified sometimes, suggesting that there is a Max Output Current limitation related to the controller itself. Is this true?

And if so,
How do I go about determining the Max Output Power a controller can handle, based on it's drive strength or any other parameter?

\$\endgroup\$
1
\$\begingroup\$

It's largely independent.
FET drive needs may be higher with increasing frequency (as you have to charge/discharge the gate capacitance more rapidly in cases where you are pushing the limits of the FET.

FET gate capacitance is device dependant and is typically in the say 1 nF-10nF range with most nearer the lower end (and some lower again). Generally a driver that will provide 100's of mA is OK. Over 1A is less usual. (Note the series resistor usually used in the gate drive circuit to help prevent ringing - this is often about 10 Ohms suggesting that sub 1A drive is expected.

If it matters enough a basic gate driver can be made with 2 x bipolar jellybean transistors (1 x NPN, 1 x PNP) and NO other components. (join emitters, join bases, NPN collector high, PNP collector low, bases to drive in, emitters to drive out. While this potentially has shoot-through issues in some cases, the 2x Vbe center dead band usually helps keep this low enough not to matter.

\$\endgroup\$
  • \$\begingroup\$ Thanks Russell, that's a great help. Would you have a preference for 2-switch/4-switch buck boosts or SEPIC for what I'm looking for? 4-Switch seems to be the most efficient - you lost some real estate but it looks like th ebest performing solution - does that sound right? Many Thanks, Dave \$\endgroup\$ – Dave Mar 23 '17 at 22:30
  • \$\begingroup\$ @Dave I assume by 2 /4 switch you mean using diode rectification / synchronous rectifier out. Synchronous rectification is usually superior loss wise at the cost of the extra FETs. It's what I'd use if efficiency mattered muchly. || SEPIC is "fun" - technically it is extremely hard to control optimally (according to experts) BUT people seem to "just do it" without apparent problem. I'd expect SEPIC efficiency to not match the classic buck-boost with 4 FETS. A possible possibility is an inverting buck boost with transformer output to reinvert the Vout. ... \$\endgroup\$ – Russell McMahon Mar 23 '17 at 23:18
  • \$\begingroup\$ ... Complexity increase is probably too high - and you lose the high efficiencies you get with the classic buck-boost when operating in the area where Vin is just slightly above Vout. How much gain you get depends on what proportion of the time Vin is in this range. eg if the supply was 1 x LiPo or LiIon cell probably slightly more than half the energy would be supplied in the 4.2V to 3.6V battery range if Vmin was set conservatively high to improve battery cycle life. \$\endgroup\$ – Russell McMahon Mar 23 '17 at 23:18

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.