Re: Need Help with Transients

You absolutely must use filter capacitors, and you really should use a full wave bridge rectifier. Since you have no input filter capacitor, but you DO have an output capacitor, about half the time your regulator chip is powered IN REVERSE!!! This is because the charge on the output capacitor is greater than the incoming voltage most of the time- only during the peaks does the input voltage exceed the voltage on the output capacitor. Rebuild your circuit, use a full wave bridge rectifier and an input filter capacitor. You would also do well to put a large inductor (like a noise filter choke for automotive audio systems) inline with the input, since the raw output of a little alternator like that can easily create transients in the 200 volt range. Incandescent bulbs don't really care, but semiconductors do.

Cheers!

Chip Shults My robotics, space and CGI web page -

formatting link

Reply to
Sir Charles W. Shults III
Loading thread data ...

Thanks for the advice. A couple of questions though: I'm trying to keep the parts count low, the board small, and I was worried that a cap being exposed to large voltage transients might not work for long. What kind of capacity and working voltage should I be looking for on the filter cap: 100uF-1000uF - 200V 400V? If I am using a 1N4001 for half wave rectification before the regulator, how is the regulator being put into reverse? Given it's super choppy DC, but the voltage regulator and LED's shouldn't care. Neither side of the regulator circuit has an electrolytic cap, just some external ceramics for proper triggering as suggested by the datasheet. The 5V output is going directly to the LED's with some current limiting resistors. What would a full wave bridge offer over half wave rectification for such a primitive task? I could see if I wanted to run a CD player or something, but for lights running at just shy of 200mA, I'm not so sure.. All the LED's still work, so I'm guessing the voltage didn't kill it, but a dead short did.. maybe it was just a flaw in how I made the proto-board or the regulator shorted internally?..

-Richard

Sir Charles W. Shults III wrote:

Reply to
yeahright

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.