simple power supply question

[BobM3]

I'm trying to figure out Microchip's application note TB008. It describes a simple nonisolated power supply that can be used for PICs. The author starts out with a simple circuit: The line voltage across a series connected resistor and capacitor. He states that as long as the reactance of the capacitor is significantly larger than the resistance of the resistor you will have a nearly constant current. I don't get that. Wouldn't the current still be a sine wave (not constant)?

I'd like to use this circuit but I'm uncomfortable not understanding how it works.

 

[Skogsgurra]

I suppose that you are happy with your constant mains voltage? Even if it is a sine wave.

Same thing here. The AC current RMS value is more or less constant because the voltage drop in the RC combination is many times higher than the voltage drop across the following rectifier/capacitor/zeber combination.

So, I think that it is all a question about how you look at it. The AC current is, of course, time-varying. But its RMS value is constant.



Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[MikeHalloran]

The current _in_ is clearly sinusoidal. I think he's trying to say that the circuit in figure 2 is capable of sourcing at least 20mA of uninterrupted DC current for each uF of capacitance. IOW, it might be better to say 'minimum current' than 'constant current'. The point of the discussion is just giving you a simple way to figure out how big a capacitor you'll need.

Unless you're intending to produce the powered circuit in quantity, so cost really is an issue, I'd strongly suggest you just buy a DC wall wart instead of using a line- powered nonisolated supply.





Mike Halloran
Pembroke Pines, FL, USA

 

[logbook]

I have used this sort of circuit in the past. The "preferred" version uses a large wirewound resistor rather than a capacitor. The point is any over-voltage events go straight through the capacitor whereas the resistor suppresses them to a large degree. Also it is easy to over-specify the resistor in terms of power rating. Over specifying the capacitor to withstand mains transients is all but impossible. The capacitor nominal dissipates no power, which is the only attraction.

For reliable operation the capacitor would need to be rated above 1000V AC (on 240V rails).

 

[BobM3]

If the purpose of the capacitor is solely to get an energy efficient voltage drop then I'd rather use the resistor per your suggestion logbook. My worry running directly off the line are the noise spikes.

Mike - I want to use the PIC to measure line zero crossing and RMS value and also to measure the current (via a series sense resistor) in my circuit. I don't have much room on the board to do all this. Using the nuetral as a common would work nicely. I too am worried about the line transients.

 

[MikeHalloran]

You'll be amazed how long a PIC can run off a 9V battery.



Mike Halloran
Pembroke Pines, FL, USA

 

[jimkirk]

PIC devices are quite robust regarding line transients (IEC61000-4-4 level 4) and EMC. Check out

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=2147&param=en023557&pageId=64