Flyback switching power supply

I have enjoyed reading Ken Shirriff's blog on the Apple iPhone charger cube:
http://www.arcfn.com/2012/05/apple-iphone-charger-teardown-quality.html#ref1

I have an iPhone charge cube taken apart as well. I have a 48VDC power supply I want to use as the input voltage (instead of 120VAC). Does anyone have any ideas on how I can make this work?

The answer can you get this charger to work from 48VDC, the answer is yes. It may actually work from 48VDC as is, although it may have UV lockout which may prevent that.

You will need to take 48VDC, and apply the +/- voltage to the DC side of the bridge diode, then turn on the 48VDC. Good idea though to slowly bring it up first to make sure nothing goes wrong, then apply the full voltage to see if it starts. Even if it does start it may not regulate because it would normally want to work from about 130VDC rectified.

Now say it doesn't work, then the primary side of the transformer/coupled-inductor must be re-wound for the same volt-seconds / duty cycle. This is a transformer explanation while in reality it is a coupled inductor. I won't go into those details because it won't help you modify it any better. The resonant clamp circuit may or may not require mods, this serves to protect the MOSFET max voltage exposure.

Basically al this module does is take AC, rectify it to make DC, then all it is factually is an isolated DC/DC converter. A complete schematic would be very helpful.

Also you'll need a primary side schematic to determine if there is a UV lockout, which would require value changes.

Also I will say the design analysis on the charger was well done. Yes Apple charges lots of money but then Apple has deep pockets and if there is a mishap where someone gets hurt/injured/killed, Apple gets sues 10 ways to Sunday. The knock offs are long gone if they catch wind of litigation.

What are you trying to power from, solar, batteries? There may be easier ways than reworking the power cube. And what power does the power cube put out. There may be simple ways to bypass the power cube entirely. Reducing voltage in a dc circuit is a fairly simple matter. Are there any protections incorporated in the power cube you desire to have in your end design? Take some time crafting your question and you will get some good answers.

I have an existing DC converter circuit that successfully converts 48VDC to 5VDC, but I now want to filter out the noise I am seeing. I want to use the iPhone power cube circuitry as a base to build upon, and see if it successfully filters my application.

I applied the 48VDC to the DC side of the bridge diode. But, no luck firing up. I couldn't get 5VDC out of the secondary. The power cube converts 100VAC-240VAC to 5VDC, so I believe 48VDC is to low. Do you think my only solution is to rewind the transformer?

Ok so after some additional calculations, the coupled inductor (transformer) could be rewound. Additional changes would also be required.

1) Would need to change the MOSFET current limit because peak current would be about 5x as much
2) Likely need to change MOSFET because conduction losses would go up about 5X (heat)

Perhaps other changes but these would be minimum and perhaps more than you would want to get involved.

A very simple solution, assumes the module will work from DC only (it should) is to get a small boost converter and take the 48v to something above say 130VDC, then just hook it to the 130VDC. Comment, 240VDC would be a nominal boost input DC given it represents the middle of the AC input which is 85VAC - 265VAC. But 130VDC should work.

I believe you mentioned some concerns with noise, you could easily attenuate this given it's probably in the 200KHz range, both common mode & differential mode.. This is of course is noise input to the module. Noise to the 48v is another issue, if that concerns you.

Yes. I will try a boost converter and see what happens. I will report back.