Fast efficient (low power) opamp suggestions?

[madshaman]

Apologies for the relatively open ended question.

I'm just trolling for any favourites/recommendation for an opamp (preferably dual or quad package) that's decently priced, fastish (1-10Ghz GBW) that doesn't require/waste much power.

I'll be using it for some PWM based DC-DC conversion circuitry I'm designing for a solar power project I've been putting together.  I'd like the PWM frequency to be near the low Mhz range to reduce supporting component size.

I *am* drudging through IC data sheets to see if I can find a pre-packaged solution but am thinking building from the ground up is going to be the way to go, or the way I want to go during prototyping.

Thanks!

edit: doh! (1-10Mhz GBW) -> (1-10Ghz GBW)

[madshaman]

P.S. rail to rail is nice but not a requirement, ultimately output will be driving mosfets directly so only need 10V or so at worst at the output, 5-6V I could get by with.

[CarlG]

Hi, I'm not sure that I understand what you're up to, but I think you're heading in the wrong direction. If I get it right, you want to buffer the PWM signal(s) with RF opamps, driving the heavy capacitive loads of power MOSFET gates for DCDC conversion? Why involve opamps here instead of using MOSFET drivers?

Maybe a simple block diagram could clear up some things?

[madshaman]

Hi, I'm not sure that I understand what you're up to, but I think you're heading in the wrong direction. If I get it right, you want to buffer the PWM signal(s) with RF opamps, driving the heavy capacitive loads of power MOSFET gates for DCDC conversion? Why involve opamps here instead of using MOSFET drivers?

Maybe a simple block diagram could clear up some things?

Apologies, thank-you for the reply; I've been busy so haven't been on the blog consistently for a bit.  I'll work up a (real) block diagram tonight.  My goal for this part of the system is fairly simple: to collect energy from the solar cells at the maximum power point.  I'll be using a micro and software to scan the cells periodically to determine/re-evaluate the maximum power point and, using feedback (probably via a power monitoring IC which will do the current sense / voltage sense multiplication for me), present the correct load to the cells by chopping the output using MOSFETs/PWM following a low-pass LC filter.  Following this would be some DC-DC conversion to charge a bank of super-capacitors at their low voltage. these banks will be switched in such that they always provide less impedance than that I want to present to the solar cells).

Diagram tonight, but that's the basic idea.  The reason I want to use such high frequency PWM is to reduce the size required of filter and other DC-DC conversion related components.

Because I'll be doing a bunch of real-world testing and not really sure what the final circuit will evolve into as I discover problem after problem, I figured while experimenting, it would be easier for me if I did all the PWM stuff myself using opamps.

Does this make any sense?

I'm somewhat green when it comes to power conversion so I'm up for any alternate routes someone with experience could suggest.

My primary design goals are:

1) efficiency
2) cost/size
3) longevity/durability

And I'll go with whatever gives the best trade-off.