12V to 3V High current Buck converter

[cowboy303]

Hay Guys
I was working on a project that needs a lot of amps (30 amps) and lower volts say 3 but I need to be able to run it from a 12  volt car battery.
I was wondering if any of you have circuit suggestion for me efficiency is kind of important I don't care about ripple that much.
thank you all very much and have a great day.

[Kremmen]

OK, an easy one for a change : Multiphase buck with synchronous rectification.
The idea is simple: connect a number of regular buck circuits in parallel and organize a phase difference between them. Say you operate 3 parallel ones; then their switches will be operated with a 120 degree phase shift.
For efficiency, the freewheel diode is replaced with a FET switched in synchrony with the main switch to minimize diode conduction and recovery losses.
Probably easy to locate a suitable controller from the numerous devices offered by DK, Mouser, Farnell &c.

[mikeselectricstuff]

There are plenty of off-the-shelf DC-DC modules for this - used for processors etc. look for "POL Converter" (POL = Point Of Load)

[Psi]

Obsolete computer motherboards may have most of the switchmode parts needed to take 12v and step it down to 3v.
Fets, inductors, diodes and a driver IC

Some work pretty close to what you need already. They generate the ~1.5v 50A cpu power from the 12v rail.
Others use the 5v rail for input but 12v is becoming more common because its more efficient at high currents

[sacherjj]

Some work pretty close to what you need already. They generate the ~1.5v 50A cpu power from the 12v rail.
Others use the 5v rail for input but 12v is becoming more common because its more efficient at high currents

I never really thought about the current numbers on a processor before.  50A, that much?  Then I think about them being under 2V and still close to 100W power consumption.  That is a bunch of electrons flowing.

[SeanB]

Think about how do you get 125A plus into a CPU. You use half of the pins on the package for power and ground, just to keep the voltage drop down low enough. Then think about the newer units needing both an IO power rail,and a variable, depending on mode and speed, core voltage and having strict requirements as to how fast they must go to the target voltage. Some of the top peak currents needed are 145A at 0.8V for some Intel units ( from reading a data sheet some months ago) and designing the CPU to dissipate 120W without going nuclear. All that current is just used in clock drivers to switch the various parts of the CPU. There are even on CPU board decoupling capacitors as 1cm of trace is just too much impedance.

[T4P]

OFFTopic
That's quite old already . Imagine the graphic processors . 300W .

The crossfires stand for dual graphic cards if you don't know .
Man , just look at the number of caps and heatsinks for the PWM modules for the AMD 990FX series ASUS Crosshair V motherboard .

That thing ( GTX590 ) eats up probably around 500W . Just think of it .

Man , how did they dissipate all the heat ?
(There are actually only 300+ pins left out of 942 for everything else for a FX-8150 i think)

Oh god , 12 layers.

[cowboy303]

There are plenty of off-the-shelf DC-DC modules for this - used for processors etc. look for "POL Converter" (POL = Point Of Load)

I would really like to build one for my self so I can customize it to my needs.
I should have been clearer on my original post I would like a circuit diagram and a parts list.
And the smaller the better.
I'm not trying to be lazy but if I had a circuit diagram I would be able to put it in on LTSpice and find out how it works and why it works and see what every thing does.
Sorry it's the best way I know of to learn.
Thank you very much.

[Jeff1946]

Go to http://www.daycounter.com/Calculators/Switching-Converter-Calculator.phtml and see what components you need.  Looks like you need an inductor that will have max of 90A going thru it.   This is why as others have said you need multiple buck converters in parallel and at different phases.

You start small.   Get some MC34063s and experiment at low currents.   Then use the MC34063 to drive a P MOSFET for higher currents.  It is important to drive the MOSFET correctly.  A good way to do this is get a MOSFET driver like the IXD series from Claire (Mouser has these).  You also need the appropriate inductors.

Good luck and watch out for smoke!

[bfritz]

I suggest going to some websites and searching.

I went to Maxim's website, and found this using the parametric search tool.

http://www.maxim-ic.com/datasheet/index.mvp/id/4602

Did you even try?