Powering a microcontroller with a variable AC source

Hi,

Take a look at the PBZ Programmable Bipolar Power Supply from Kikusui, frequency response to 100kHz but limited to 40Volts.

See below Kikusui link for further details.

Regards

Doug http://telonic.co.uk  or follow us on twitter http://twitter.com/telonic

http://www.kikusui.co.jp/en/product/detail.php?IdFamily=0105

Doug,

Thanks for the reply.

This device looks close to what I want, but I need it to be portable. Nothing bigger than an average "smartphone" in size.

Let me know.

Cabe

Hi Cabe,

In your spec you have a varying frequency and varying voltage amplitude. This situation is similar to wind turbines, where the frequency and the magnitude of the generated voltage is variable and unpredictable within certain limits.

What they do is convert the varying freq, varying amplitude signal and convert it to a varying DC level through rectifiers (Thus removing the varying frequency element of the input power) from there they use an inverter to get 120VAC, perhaps a similar idea is applicable to your project, you haven't given to many details so I'm speculating.

Since your final output is DC then you don't need the inverter. If the resulting DC voltage will be higher and lower then the desired output voltage (5V) then you'll have to use a buck-boost,SEPIC, Cuk, or Zeta Converter. If you can somehow constrain the variability of the input power source then maybe you can avoid the higher-order topologies.

hth,

Jorge Garcia

Jorge,

Thank you for the reply. I did not know that is how they handled wind turbines.

My device needs to maintain a consistent DC voltage on the output. And I want to take as much of the excess voltage, from the AC source, and store it. So when the AC source drops low, I can supplement the deficiency with the stored energy. And I would like to not dissipate energy as heat from my device. A buck-boost topology could definitely handle the output.

The difficulty here is handling the source. Perhaps I should make a two stage device? One for energy storage, and one for the output.

Cabe

Hi Cabe,

That definitely sounds like the way to go. You could rectify the source and then use a buck-boost converter for the output to your main device. You could then use a secondary converter to charge batteries of whatever chemistry you plan to use. You would do well to have the micro monitor the amount of power available and assign priorities.

For example the uC circuit will have the highest priority, and the battery charger will have the lowest. The battery charger will only run if sufficient power is generated to run both the uC and the charger, otherwise only the microcontroller will be active.

hth,

Jorge Garcia

Hi,

I'd like to propose a different approach, based on a preregulator followed by an LDO. I sketched it out in the schematic below. Q1 is a controlled diode (like the SCR bridge you can find in 50Hz transformer based AC/DC supplies), switched on only for rectified voltages below a threshold.

Ipetacchi,

Interesting idea, amplify if it drops below a threshold. I think I would like to try this setup with some sort of charging circuit for an onboard battery.

Thanks again.

Cabe

I am definitely thinking of moving in this direction:

AC source -> conditioner circuit -> charging circuit -> battery -> microcontroller or other components

Thoughts?

Cabe