Stuck on a power problem (but not with my Arduino)

OK, this is not an Arduino problem, but it IS my Arduino project and I am running out of patience looking for a solution. (I’ve been searching for several weeks) I hope someone here has some power supply design knowledge I can tap…

I’ve got a 3-phase alternator that puts out 12-125 VDC after rectification. I need to control that as efficiently as possible and convert it to around 15-25 VDC @ maybe 3 Amps where it can be better managed for driving high-power LEDs, charging a LiIon battery pack and running my Arduino.

I’ve been looking at a myriad of switching power supplies, but they all have a relatively narrow range of input. I can either find ones that handle 12-36 VDC input or ones that handle 90-260 VAC input. But I need a very wide range of around 12-125VDC input. The top end is the most important because I don’t want to fry the input while I’m riding my bike at night at 40 MPH. That could be disastrous. :o

It’s no problem rolling my own power supply circuits or just buying something that works. But I can’t find enough info to do either. If anyone has information on ‘wide range input voltage’ switching power supplies, I would be utterly grateful! I would love to get advice, schematics, and/or supplier information. :-/


I doubt you will find a DC-DC converter for such a wide range. Most systems do
not require such a wide range. 2 to 1 or 3 to 1 is typical. 10 to 1 is not.

The 90VAC - 260VAC has a voltage doubler on the AC input. If the auto-ranging circuit detects an AC signal in the 120VAC range the doubler is enabled.
The DC-DC converter connected to the output of the rectifier circuit will have a DC-DC
range of 200 - 400 VDC.

I believe there may be an application note on the Vicor website that discsses the
autoranging AC input circuit.

(* jcl *)

Hmmm, OK. I was afraid the wide range was going to bite me like that.

Maybe I’ll devise some sort of multi-staged setup where I have multiple regulators and switch the alternator between them based on its output voltage.

Thanks for the response!

You might want to look at not rectifying it and using a “normal” switch mode supply. While these say they go down to 90V or so in truth they go down even further especially if you are not drawing too much current. So many will be able to cope with 25 - 30V AC.

Ya, I’m not sure how to do that since it’s a 3-phase output. Maybe I can use three regulators in parallel and make the common pole into a ground. I’m going to have a closer look at the little switcher wall warts I’ve disassembled and see how their primary section works.


I’m not sure how to do that since it’s a 3-phase output

Just use on of the phases, that’s what they do with the mains.

That would defeat one of the primary goals of the project - high efficiency. The alternator was purpose-built for the project so I went with 3-phase because it’s much more efficient than single phase. Although I noticed on some of these wall-warts they rectify the input immediately. So I may be able to feed the DC in and still get what I need.

And actually I may be doing something like that for another purpose. I want to shut off one or two phases at a time to help regulate the alternator output at higher speeds. But I’ve run into high ripple problems with that so far, even with very large filter caps.

EDIT: So check this out! I took my cell phone charger, which is rated at 5VDC, 700mA out and 100-240VAC input and I hooked it up to my bench power supply. I was able to pull the full output from the charger with as little as 26VDC input! I’ve got a 12V/1A wall wart at work that I’m going to test this evening. If it works, you have solved my biggest problem so far! AND I won’t have to shut off phases because the wall warts can take up to 240V.

EDIT AGAIN: Well, back to the drawing board. My test this morning was flawed - I hooked up my DVM incorrectly (in series with the load). So the wall wart was not loaded down at all. The tests I ran tonight would not give me any output with the input up to 50VDC. Oh well.