Thread: `NEW' Small DC-DC ATX Power Supply

i'm interested if the price is reasonable.

Originally posted by Fosgate
What I want to know is the operational input range for the supply and what technology has been used eg linier, flyback, forward
And lastly the efficiency of it.

Yeah, I'm quite interested in this too. Afaik, it can't be linear because that would dissipate too much heat to not require fans or heatsinks at high load. It can't be a forward or flyback converter because that requires coils/inductors. Perhaps a charge pump/switched capacitor? Don't know how that could get enough line regulation at high currents though.

Having built a small high power supply, I can appreciate that if the claims are true it is quite an impressive acheivement. Although mine was small and not as high powered as this one, it definately needs it's heatsink...!


Rob

I'm interested too, dependant on the price!

how much????

Hmm interesting ..

NO inductors
NO coils
NO heat sink
NO fan

...... measures only 4" x 1.5 x 1.5 .... if all said above is ture .. then i think Keypower is soon going out off bussines...

Good work ...Kaboom

Keep us posted.

Mastero

only prob is its low power.. :P does drawing for current than you have specified generate some heat and then require a heatsink?? or what??

Mastero - I was able to find some awesome DC converters from TI (texas instruments) and was working on them with one of my advisors and we were able to come up with something that uses this devices at nearly 90% efficiency. I threw together Sproggy real quick just for kicks and found out without a heatsink those Maxim chips go into thermal shutdown in a matter of seconds. This sucked, because their efficiency is somewhere in the area of 75-85% so if you're converting 12V to 5V and pulling say... 3 Amps, you're dissapating (7V x 3A = 21Watts) over that little chip. Having a huge block of aluminum to do this is a waste of space/size/weight, etc... not acceptable for me... so i checked out, consulted a few people and 'think' i've come up with a better solution that will be test early next week. These devices are close to an all-in-one solution where you rid the inductors and only keep some resistors and caps in the circuit. I may have jumped the gun a little early, only tested one line (12V to 3.3/5.0) and found that it worked and assumed the whole circuit is a go and got everyone too excited but it was an effort...

oh... forgot to mention i'm doing this through ISR's
integrated switching regulators
that seems to be what everyone was asking...

KaBoom just another question.
When you done with testing and everything are you planning on sharing the schematic of the PSU with the rest of the comunnity?

ISRs are cool! They are a really easy way for a design engineer go get a power supply. The Part numbers below get you started with +3.3@6 amps, +5 @6 amps and +12 at 1.5 amps. But look at the cost, your already up to 60 bucks and you haven't done the board, glue logic, and filter caps.
He probably used parts like the ones listed below. They do have heatsinks and they do have coils, but our friends in China have already fab'd them for us since they are predesigned and put together, all you have to do it through them on a board.. Being 85% efficient, it is about on the same level as sproggy..

Go to TI's website and search for parts like these:
Part#..Vinlow.. Vinhigh.. Iout.. Vout.. adustable range out.. cost

PT6625...9...16..6.....5......2.9...6.5 $18.99
PT5071...7...16..1.5..12....10....15 $21.17
PT6621...9...16..6.....3.3...2.3...4.5 $18.99

Kaboom, what is the specified input range of your PSU? The ones I found would only go down to 9 volts min..

-Jeff