PAGE 1 of 2 - PICKIT 3 External Dual Power Supply (3v3-5V) Circuit
First of all, why... Why use such a circuit since the PICkit 3 has a built-in power supply software controlled. There are basically two reasons to build this circuit: First of all, it may happen - The internal PSU may be destroyed. Actually, i fried the PICkit3 PSU myself and i know for sure that it may happen...
There is also another very important reason: To increase the current supply capability. According to the PICkit™ 3 Programmer/Debugger User’s Guide:
|When using the USB connection, PICkit 3 can be powered from the PC but it can|
only provide a limited amount of current, up to 30 mA, at VDD from 1.8-5V to a
small target board.
Consider this: A single LED may consume as much as 20mA...
So simple: Based on the LM317, its just a constant voltage power supply. To select between 3.3V and 5V, a miniature slide switch is used (click to enlarge):
First, the input capacitor. I've left the voltage range open on purpose. Typically, a 60V or higher voltage rating is suitable for any input voltage rating. But I'd rather suggest you select the proper capacitor as per your input supply voltage. If you plan to use a 12Vdc PSU for input, then a 16V input capacitor is enough, you don't need to select higher rating to keep the capacitor size down to minimum.
The input voltage must be DC from a power supply. Make sure the polarity is correct. The LM317 can regulate anything between 1.5 and 40Vdc. But since we need to have 5V output, the input voltage must be at least 6.5V (there is typical 1.25 to 1.5 voltage drop across the adjustment resistors).
The power dissipation on the LM317 depends on two factors: The voltage difference (Vout - Vin) and the current:
Here is an example. Say we set the output voltage to 5V. The input voltage is 24V. The circuit draws 100mA (0.1A). The power dissipation is:
|P = (24 - 5) x 0.1 => P = 1.9 Watt|
It may not sound much, but if the circuit runs for a long time, the temperature on the LM will increase significantly. I strongly suggest to use proper heatsink for power above 1.5 Watts. To keep the power dissipation low, select a lower input voltage. If for example the input voltage is 9V instead of 24V:
|P = (9 - 5) x 0.1 => P = 0.5 Watt|
As you can see, there is absolutely no reason to us higher supply voltage than 9 Volts. After all, the output voltage will never exceed the 5V.
The switch SW1 is used to select between 3.3 and 5V output. Here is how it works. Say that the switch is OPEN (as in the schematic). The two 150 Ohms resistors (R3,R4) are connected in series with the 360 Ohms resistor. The total series resistor is then 150+150+360=660 Ohms. The output voltage is then:
|Vout = 1.25 x (1 + (R2/R1)) = 1.25 x (1 + (660/220)) => Vout = 5 Volts|
Now say that the switch is CLOSED. In that case, the two 150 Ohms resistors are bypassed by the switch and are not connected in series with the 360 Ohms resistor. The output voltage is:
|Vout = 1.25 x (1 + (R2/R1)) = 1.25 x (1 + (360/220)) => Vout = 3.3V Volts|
One may ask "Why use two 150 Ohms resistors and not one 300 Ohms?". Well, i just found it easier to connect them on the pre-perforated PCB. You can definitely use one 300 Ohms resistor instead...
Keep on reading to see how I soldered the parts together on the pre perforated PCB (click here)
|At 18 May 2014, 4:29:23 user Cristi wrote: [reply @ Cristi]|
Hello. Idea is OK, but not necessarily quite discontinuing of Vcc from pickit3?
At 24 September 2013, 19:11:30 user jdraughn wrote: [reply @ jdraughn]
You might want to insure that the switch you use is a make before break, otherwise the voltage will spike as you switch from one voltage to another.
At 23 September 2013, 21:20:51 user solipso wrote: [reply @ solipso]
Nevertheless, the idea is great as we, lazy dogs, always like to power our PIC circuits directly from PICKIT - regardless the current needed. Your hack justifies this bad habit and makes us look a bit less silly, though.
I also like these switches very much, especially with multiple poles and with good contacts (surely not those eBay cheapies with 0.3 to 0.7 Ohm ON resistasnce).
At 23 September 2013, 17:09:18 user Giorgos Lazaridis wrote: [reply @ Giorgos Lazaridis]
@solipso My original design involved a HDD jumper. Jumper ON means 5V, otherwise 3.3. It was safe enough, but then i saw my stock bin with these sweet slide switches... I just could not resist.
At 23 September 2013, 13:58:58 user solipso wrote: [reply @ solipso]
Surely it does! :-) Nevertheless, this is based on my very own experience. I have had something like this on my breadboard PSU and I killed quite a lot of devices stupid enough to get killed by humble 5V before I realized to shorten the lever to about 3mm, which proved to be enough to switch easily, but intentionaly only.
At 23 September 2013, 13:50:38 user Giorgos Lazaridis wrote: [reply @ Giorgos Lazaridis]
@solipso And this makes things more interesting, doesn't it???
At 23 September 2013, 12:16:23 user solipso wrote: [reply @ solipso]
The long protruding lever of the voltage switch literally asks to be switched to 5V when the rig is connected to 3V3 micro and let the magic smoke escape free.
HOT in heaven!