In Reply to: Shunt reg TL431question. posted by -Steve on January 10, 2001 at 08:16:43:
My guess is - at this I/U values you'r overloading the regulator. 5*50=250mW at 156 C/W rated thermal resistance causes a 40C increase in chip temperature vs. case, plus allow for inadequate ventilation, that's 60C over ambient. The chip can live with it although quite hot inside. But if you disconnect load - TL431 power increases 50%, and that's too close to it's end.
Start with minimum/maximum U(input) and maximum I(load). Then TL431 must conduct at least 2 mA (minimum Uinput - maximum Iload). Then maximum Rseries =(Uinput min - Uload)/(Iload max + 2mA). Then, for maximum Uinput and zero Iload recalculate current and power through TL431. It shouldn't exceed 250 mW to be safe. If you have enough power margin, you can decrease Rseries of course, increasing TL431's minimum current, but why? The whole regulator's output impedance is in TL431 and output caps, not the resistor.
Also you probably know - power ratings on power resistors are misleading. A 1W resistor running half of rated power can warm up to above 200C and fry nearby components, - I usually at least double worst-case continuous wattage and place them as far as possible from the finer parts of a circuit.
That was exactly what I need to know.
I replaced the Jung style series regulator circuits in my Audio Note DAC 1.1 with some kludged TL431 that were running pretty hot.
Boy howdy, do those '431s sound good!
This post is made possible by the generous support of people like you and our sponsors: