ESP8266 Community Forum

- Fri Feb 27, 2015 4:29 pm #10924

You can expect a drop of up to 700mV across any diode junction, some of the components I am using are from the 70's and 80's so they have degraded slightly, therefore these "consume" the voltage WITHOUT generating heat, when you allow the voltage to be "dropped" by large amounts through a typical voltage regulator the greater the difference between Vin and Vout, the greater the thermal increase.

Sorry, but this is just ridiculous. As long as the input and output current are the same, you cannot "consume" voltage without releasing the energy in some other form. Doing so would violate the law of conservation of energy. So unless your diodes somehow start spinning or glowing, they WILL generate heat. The main difference is, that with your diode arrangement you have a much larger surface area, so each diode's temperature will not rise as quickly as a small LDO package.
What you build there is similar to a linear voltage regulator, but without any kind of regulation. The diode drop is actually not constant, but depends on the current passing through. So in your setup the output voltage will drop as soon as you start drawing any significant amount of current.
If you really want a more efficient power supply, get a DC-DC step-down converter.

- Fri Feb 27, 2015 6:24 pm #10934 As you pointed out this is not about magically making the energy vanish but "AVOID" hot areas concentrated in small spaces, the 9v input to the AM1117-3 is creating a "hot area" when enclosed would be an issue, anything above 6v - the thermal dissipation in the small mass requires, as pointed out earlier a HEAT sink, again increasing size and I have an abundant supply of PTH components, so even to keep the UNO and it's Network board power regulators running cooler I use Diodes.
Thermal management goes hand in hand with stability on devices of this size.