I was concerned about damaging my PC (or worse) using the PC power cable described in the previous post. To make things safe I built a current limit using some scrounged transistors. I also added a switch and two LEDs to show the power status. Now its safe to use the +5 from my PC, if I manage to short something, current is limited to less than 200 mA.
R1 sets the max current at about 0.7/R1. R1 is 36 Ohms so that makes the max current about 195 mA. D2 (LED) shows that PC power is active and D1 (LED) shows when power is OK. When power is shorted D1 will be off. The voltage drop is about 100mV and since PC +5 outputs about 5.2 Volts there is plenty of room to spare.
T1 is the pass element, meaning the current passes though it. T1 is turned on by R2 pulling down its base so current flows through R1, T1, and into VCC or our powered circuit. T2 is connected so that if the voltage drop on R1 gets high enough to turn it on (0.7V), it pulls up on the base of T1 turning it off. T2 has a lot of gain so the current limit will be sharp.
A HFE (beta) tester built into a DVM identified two PNP transistors from my junk bin. The one with the best HFE was used for T2. T1 should be saturated under normal operation and would be fine with a transistor with a beta of about 100.
HFE is a measure of current gain and indicates how much collector-emitter current you get with a given base-emitter current. The transistor selected for T2 showed a HFE of over 200 therefore it should supply 200mA for every 1mA of base current.
The two pictures below show my prototype current limiter in action. The DVM shows the normal input current at about 50mA. In the second picture the output is shorted to ground and the DVM shows about 190mA of input current.
8 Responses
Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.
Isn’t the correct value for R1 3.6 ohms, not 36 ohms?
Jeff,
You are correct. 3.6 ohms is the value I used. I had an error on the schematic, good catch.
David
Thanks a lot, exactly what i was looking for to not damage my usb port while playing with my atmegas
Is there any way to make this a Variable power supply?
Juan,
Probably the best way to make it variable is to add an adjustable voltage regulator. Maybe a LM317 which also has built in overload protection but I would still use a current limiter on the PC side. If you were to short the regulated supply output it would likely surge the current from the PC and cause it to reset.
Kevin,
Remember that there is some voltage drop on this current regulator. If you start with 5V, like the USB port, you will end up with less at your hardware. The 5V from the main PC supply tends to be higher then 5V. The USB maybe as well as it probably comes from the same point.
DFowler,
I would like to be able to provide about 1000 mA for my proto system. In addition to the change in R1, what should I change in the rest of the components (especially in the transistors)?
Thanks
1000ma or 1A would require that you change the pass components to high power parts. A power transistor for example. R1 will need to be set to drop about .7V at 1A and would be disapating .7W I would go for a 1W resistor and I would assume the transistor would need to disapate about 1W as well. A T0-220 package for the transistor would likely be OK. You should mount it to something to disapate the heat.