using pic to reduce wattage by changing duty cycle

[3v0]

To reduce heating pick a relay with a coil that can be energized using your unregulated DC power. Using the unregulated power also prevents the microprocessors prower supply from getting messed up when the relay kicks in. But I am guessing you knew that.

If you are into making your own PCBs you should also checkout the board stock under "Copper Clad" in the index. They have the .021 and .016 thicknesses. I have not tried it but have read elsewhere that it is good stuff. The price is right.

 

[eblc1388]

OK, that works. .47uF cap gave overall current draw of 20ma and voltage at the relay of 50V. The only non-polarized cap I had of high enough voltage was bigger than the relay, so I'll have to shop around for something smaller! What would be the voltage rating of the capacitor required? 50V or 120V or 250V ??
Metal Polyester the smallest?

If physical size matters, then capacitor dropping method is not quite suitable because one should use a 125V "X-rated" capacitor which is designed to be connected across the AC main and their size is usually big.

We're dropping 2 watts somewhere, bigger than my original loss with the resistor, so I have to try the bridge method.

How do you know it is real power and not capacitor power of which will not registered on an energy meter.

 

[jimg]

I just measured the total amperage in, and subtracted what the relay was using. With the bridge, it's back down from 20ma to the 8ma the relay uses. The size is important so I probably can't use this method, but it did work very well. Thanks.

 

[jimg]

For all you skeptics out there, you were correct. I finally got PWM going with a 12F683. Using a one in 24 duty cycle, it created a beautiful sawtooth waveform varying from 46-48V across the relay/cap, everything worked just fine. The only problem was the mosfet got just as hot as the old resistor did, so the net result (other than learning about PWM) was zero. I'm using a VN2410L, a nice little mosfet with what looked like good spec to my uneducated eye . So much to learn, so little time (sigh).

So I want to look at the capacitor thing again briefly. What is an "X-rated" capacitor when looking in the DigiKey or Newark catalog? The only X I could find is like X7R which can't be it.

I'm also going to look into 120V AC relays and a triac as a last resort. So far I haven't found any triac's with gate trigger current of less than 5ma according to the specs, which is about what I need for my original circuit so no gain there. Perhaps they don't actually need this much.

 

[3v0]

I am confused.

You must be getting the Vcc for the pic from somewhere. Can you use the unregulated DC to energize the relay coil?
The heating would be limited to that generated by the resistance in the 2N222 or whatever you choose to switch the coil voltage.

Mouser has 10A @ 120V relays with coils at 5, 12, and 24 volts for $1.50 or less. One 24V is $.98. Is there a reason you can not use one of these ? What am I missing?

3v0

 

[eblc1388]

What is an "X-rated" capacitor when looking in the DigiKey or Newark catalog? The only X I could find is like X7R which can't be it.

X rated capacitor are those designed for connection across the AC power terminals.

You can select from the following DigiKey link. I think the MKP type is also suitable.

**broken link removed**

 

[jimg]

3v0 - See this thread - https://www.electro-tech-online.com/threads/power-supply-for-a-touch-switch.23176/
Since I have to throw away any voltage/current not used by the relay as heat, I used the highest voltage/lowest current dc coil relay I could get at a reasonable price. Using a 24V relay would double my problem. Of course the optimum solution is a transformer, but they are too big and too expensive especially when you only need 6 to 8 ma to run the relay and under 100ua for everything else combined.

Thank you Mr. Chung.

 

[3v0]

Thanks for clearing that up. Now that I know what you are doing it makes perfect sense.

Please post or start another thread when you get it up and working.

3v0