the intended design is going to be 12v and between 1mA to 2mA. I read an article that said you should match the voltage and current of the adapter as closely as possible to the device, but if you have a power supply that can put out 12V at 1A, doesn't that just mean that it can give 12V to the device up to 1A? Afaik, if you're not using 900mA, the power supply should only put out the mA the device is asking for, e.g. 100mA, right?

Also, if the above is true, would buying an unregulated PS of 1A or more actually be better for devices that use only a few mA? From what I read, it seems like the voltage output of an unregulated PS will suffer if you get close to or go over the rated current output-for-voltage of the PS. Wouldn't it be able to provide maximum steady voltage if the current draw of the device is far below the spec of the adapter?

 

If you are asking about Wall-Wart plug-pack transformers, then as long as the current actually being drawn is less than the rated current, then you will not overheat the transformer. These transformers are specially wound to be self-protecting if overloaded, and they are intrinsically over-current limited by design utilizing core saturation..

A downside of using a 1A wall-wart to power a 10mA project is efficiency. Several Watts are wasted just "exciting" the primary winding. If you used a transformer rated for your actual load, it would be more efficient, and produce less wasted heat...

 

I have never encountered a wall wart rated at 1 or 2mA.
The lowest I have seen is 60mA.

 

Afaik, if you're not using 900mA, the power supply should only put out the mA the device is asking for, e.g. 100mA, right?

Also, if the above is true, would buying an unregulated PS of 1A or more actually be better for devices that use only a few mA?

The circuit is only going to draw the current it demands.
I think you will find that the vast majority of wall-wart supplies now are regulated switching type, at least that is what I have found, the days of the old linear type where the load was required to bring it down to the rated voltage is pretty much a thing of the past, and also they are extremely cheap in comparison.
Max.

 

You are designing this, you said. So it will likely be a linear type, with a transformer, rectifier, and capacitor(s).

Your only major topology question is whether or not to regulate it.

I would regulate it, if you think it might ever be used to power a device that could be damaged by over-voltage.

You will want to try to make sure that the bottom of the ripple voltage never dips into the range where it makes the regulator's input voltage minus output voltage less than its dropout voltage spec.

I would design for some reasonably high output current, probably at least half of what the transformer is rated to provide.

Your rated output voltage will be the transformer's peak output voltage (1.4 x its rated RMS output voltage) minus the rectifier forward voltage at your max output current minus the regulator's dropout spec at your max output current minus a dropout safety margin minus the ripple voltage caused by the max output current drawing down the capacitor voltage between charging pulses from the rectifier.

Figure out the max ripple voltage you want/need to allow and then calculate the capacitance needed to provide that, at the max rated output current.

I = C dv/dt so you can use
C = imax dt / dv where
dv is the allowable ripple voltage, dt is the time between charging pulses (1/2fmains if full-wave rectifier or 1/fmains if half-wave rectifier), and C is in Farads (so multiply by a million to get uF).

 

I don't think you can buy a connector for 1ma or 2 ma. You're going to have to use something that is capable of a lot more current than that. The wondrous part? It won't hurt anything.

 

What about a small snychronous motor from a 110v clock which powers a flywheel, connected to a small generator for 2 mA.

These motors for 1.5v quartz clocks can work 2 years from AA battery.

Now the trick is, there is also a 1.5v armature. Every minute it turns on the flywheel motor for 5 seconds.

So the main current is reduced to 1/20.

it needs a seperate 1.5 battery but as said it lasts 2 years.

The flywheel is almost microscopic and cheap to fabricate.

This is only a model for a type of powersupply which is not allowed to discuss here. In the model however there is no electric connection.

 

You can use an energy harvesting chip powered by a small solar cell.