Hello everyone!
I need to build a CMOS-controlled MOSFET switch with short circuit protection. The attachment shows what I have came up to so far.
The schematics of current-limited MOSFET switch is taken from "The Art of Electronics" by Horowitz/Hill. When control input is high, Q1 opens and pulls down the gate of Q4, thus opening it. If the current through Q4 reaches approx. 1A (as a result of short circuit, for example), the voltage drop across I-SENSE reaches 0.6V and opens Q3, which pulls Q4's gate up and prevents current from further increasing.
The problem, however, is this: when Q4 output is shorted, the current through it is limited to 1A, but there is 16V across it, so it dissipates 16W and overheats very quickly. During normal operation Q4 stays cool, so I think using a radiator just to dissipate short circuit heat is ineffective.
So I added Q2, T-SENSE and R4, to shut down MOSFET completely when it becomes hot. T-SENSE is a posistor (temperature-dependent resistor), which is physically attached to MOSFET. It has normal resistance of <500 ohm at 25 C. At about 90 C its resistance jumps goes up to 10k, and voltage drop across it becomes enough to open Q2, which pulls Q4's gate up to 16V, and thus shuts down Q4 completely until it cools down. Q2 will not open until voltage drop across T-SENSE reaches approx 0.6V, which, according to posistor datasheet, will not happen until 70-80C of temperature.
I have simulated this in Proteus, and it seemed to work ok, but I'd like to hear some qualified opinions. Am I trying to invent a bicycle? Maybe there's some better/simpler solution? I would appreciate any thoughts on this. Thank you!
I need to build a CMOS-controlled MOSFET switch with short circuit protection. The attachment shows what I have came up to so far.
The schematics of current-limited MOSFET switch is taken from "The Art of Electronics" by Horowitz/Hill. When control input is high, Q1 opens and pulls down the gate of Q4, thus opening it. If the current through Q4 reaches approx. 1A (as a result of short circuit, for example), the voltage drop across I-SENSE reaches 0.6V and opens Q3, which pulls Q4's gate up and prevents current from further increasing.
The problem, however, is this: when Q4 output is shorted, the current through it is limited to 1A, but there is 16V across it, so it dissipates 16W and overheats very quickly. During normal operation Q4 stays cool, so I think using a radiator just to dissipate short circuit heat is ineffective.
So I added Q2, T-SENSE and R4, to shut down MOSFET completely when it becomes hot. T-SENSE is a posistor (temperature-dependent resistor), which is physically attached to MOSFET. It has normal resistance of <500 ohm at 25 C. At about 90 C its resistance jumps goes up to 10k, and voltage drop across it becomes enough to open Q2, which pulls Q4's gate up to 16V, and thus shuts down Q4 completely until it cools down. Q2 will not open until voltage drop across T-SENSE reaches approx 0.6V, which, according to posistor datasheet, will not happen until 70-80C of temperature.
I have simulated this in Proteus, and it seemed to work ok, but I'd like to hear some qualified opinions. Am I trying to invent a bicycle? Maybe there's some better/simpler solution? I would appreciate any thoughts on this. Thank you!
