# Thread: Designing New DC-ATX Power Supply

1. Heh... 70 watts is about the amount of heat that an overclocked AMD T-bird puts out.

And you should see all of the water cooling and peltier setups to get rid of that much heat!

2. Originally posted by Raas:
<STRONG>
Jeff
I won't be 'pumping' 10a through just one regulator, it would be 2 of them. (5a each)

When I look at Sproggy's design, He's 'pumping' also 5a through each regulator. And yes he has a heatsink, but those things aren't that big. So I won't think that I would end up with a huge power supply that wouldn't even fit in my trunk.
</STRONG>

What you have to understand is the difference between a linear and a non-linear power supply. On sproggys power supply you see inductors. They are like little batteries. When you push current through them the store up a bit of a charge. If you use the Maxim parts like sproggy used, they won't convert the power that isn't used into heat, they basically store it. If you look at the data sheet for the efficiency of a Buck or Boost DC-DC converter you see that it is around 80 percent.

Examples:

#1.) PWM based inductor DC- DC converter
12 volts to 5 volts @ 5 amps.
80 percent efficient.
@ 5 amps, power converted to heat is 12V*5A*(1-0.80) = 12 watts of heat

#2.) Linear 78XX based regulator
12 volts to 5 volts @ 5 amps.
Variable efficiency.
7v * 5 amps = 35 watts of heat
41 percent efficient.

Some of the PWM inductor based converters reach efficiencies of 90 some percent.

Jeff_

3. My very first PSU was built with linear regulators. I used 3 adjustable ones in parallel. The problem is linear regulators are insanely inefficient. They may work fine inside small electronic devices, but as soon as you try to draw any considerable current from them, look out! Needless to say, the linear PSU I built ended up in the trash and the workbench was cleared to make way for the design and construction of the DC-DC ATX PSU I am using today.

Switchmode PSU's are superior in efficiency to any linear regulator. Most are around 80-90 percent efficient, about DOUBLE that of a linear regulator. With the high efficiency, you get several benefits:

1. Lower heat dissipated. Linear regulators dissipate excess energy as heat. Switchmode regulators are more efficient and thus generate less heat.

2. Higher current capacity. Switchmode PSU's are known for their high current output capabilities. This is why they are used in PC Power Supplies, monitors, and even on motherboards to generate the VCore voltage for the processor.

3. Smaller packaging. Since there is less heat produced, the heatsinks can be smaller. High frequency operation allows for smaller capacitors and other components.

4. Superior voltage regulation. Switchmode power supplies have a built-in mechanism that keeps the output voltage constant even under a changing load. Linear circuits don't do as good a job at this. Computers demand clean power, and switching power supplies are very good at providing it.

IMHO You're better off designing a switching PSU from the start than trying to tweak a linear PSU to work for you. With the proper design, a switchmode PSU can be built using a single transformer at its core. This is exactly what standard PC power supplies use.

Page 3 of 3 First 123

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•