On Mon, 20 Apr 1998 22:46:42 -0400 Dave Celsnak writes: > How about 2(or 3??) voltage regulators to >step >down the voltage with less of a drop? Using linear regulation of any sort, the total amount of heat dissipated by the regulator circuit will always be the same. You can redistribute it among different parts. Two simple methods involve resistors. Power resistors are often better at dissipating heat than ICs. A resistor in series with the regulator input will reduce the input voltage under load, causing less voltage drop across the regulator. There should be a moderately large capacitor at the regulator IC input to stabilize things. This method is good if the input voltage is quite high (e.g. 24V). If the input voltage drops, the voltage at the input to the regulator will be too low, and the regulated output voltage will fall. In an automotive application, the voltage can drop to 7V or less when the starter is operating. If the circuit powered by the regulator has to operate then, no resistor can be used in series with the regulator input. The other method with a resistor is to place it in parallel with the regulator so some of the load current flows through the resistor instead of the regulator. Thus the current flow through the regulator is less and some of the power dissipated by the regulator is transferred to the resistor. But, if the input voltage rises above a certain limit, and/or the load current falls below a certain limit, the regulated output voltage will rise out of control. This can be very bad. In an automotive system, high (50V) transients are common. A parallel resistor circuit would have trouble dealing with them. So neither one of these techniques is really suitable for your application. Another method is to use a "pre-regulator". This can be just an NPN transistor connected as a voltage follower with the base voltage somewhat regulated by a Zener network. The emitter supplies the input terminal of the final regulator with about 8V. A really nice feature of this circuit is that it can withstand high voltage transients at the input (collector of the transistor). The common "current booster" configuration with a PNP transistor is similar. Most of the heat will be dissipated in the transistor. This circuit has poor response to input voltage transients though. Are there any hang-ups with a >switching supply? This is an automotive application, and noise can be >high. I would like to keep parts count to a minumum. I'd stay away from switching regulators in this case, unless there is just no way to use a heat sink. One of the newer 3-terminal regulators designed for automotive use and fitted to a heat sink seems like the best solution. Look at the load again and see if there is any way to reduce the current demand. For example, if it is a large number of LEDs they could be operated from the 12V line distributing the heat lost in reducing 12V to LED voltage among many small resistors instead of all in the 5V regulator. _____________________________________________________________________ You don't need to buy Internet access to use free Internet e-mail. Get completely free e-mail from Juno at http://www.juno.com Or call Juno at (800) 654-JUNO [654-5866]