Voltage spikes control for power converters
A novel inductive overvoltage suppression circuit for power converters is presented. High amplitude voltage spikes are generally occurring in high frequency power converters in presence of small parasitic inductances coupled to the power distribution rails, in correspondence of the switching transitions, particularly when high load currents are required. The presented invention proposes active clamps to limit the amplitude of the overvoltage. Furthermore the excess energy in the parasitic inductances is utilized to provide energy and/or a signal to determine when to turn on the next phase power device with the fastest transition possible without incurring in cross-conduction currents in the power stage of the converter, thus improving its overall performance, and circuit reliability in addition to achieving high conversion efficiency.
1. A circuit to attenuate the voltage spikes caused by a parasitic inductance of a high current interconnection to a power distribution rail, in correspondence of switching transitions of a first power device of a switching apparatus, wherein said first power device is coupled to said high current interconnection to said power distribution rail, comprising:
at least one transistor directly coupled to said high current interconnection of said power distribution rail and to a low current interconnection to said power distribution rail,
wherein at least one of said transistors turns on when said voltage spikes reach the threshold of said transistor;
whereby at least a portion of the excess energy stored in said parasitic inductance is transferred by at least one of said transistors to at least one node of said switching apparatus in correspondence of the occurrence of said voltage spikes;
whereby no portion of said excess energy stored in said parasitic inductance is dissipated in said first power device, and
whereby at least one of said transistors limits the instantaneous voltage across said first power device.
2. The circuit of claim 1 wherein at least one of said nodes of said switching apparatus is the gate of a second power device, and
wherein said second power device is turned on as result of said transferring of at least a portion of said excess energy stored in said parasitic inductance of said high current interconnection to said power distribution rail.
3. The circuit of claim 1 wherein said switching apparatus is a switching power converter.
4. The circuit of claim 1 wherein at least one of said nodes of said switching apparatus is the output terminal of a switching power converter.
5. The circuit of claim 1 wherein at least one of said nodes of said switching apparatus is a node of the driver circuit of said switching apparatus,
wherein said driver circuit of said switching apparatus provides the signals to drive a second power device and,
whereby the voltage of said node is altered by the turning on of at least one of said transistors to indicate that said first power device is turning off.
6. The circuit of claim 1 wherein at least one of said transistors is a MOS transistor.
7. The circuit of claim 1 wherein at least one of said transistors is a bipolar transistor.
8. A method for reducing the voltage spikes caused by a parasitic inductance of high current interconnection to a power distribution rail in correspondence of the switching transitions of a first power device of a switching apparatus, wherein said first power device is coupled to said high current interconnection to said power distribution rail, comprising:
turning on at least one transistor directly coupled to said high current interconnection to said power distribution rail and to a low current interconnection to said power distribution rail;
whereby at least one of said transistors turns on in correspondence to said voltage spikes;
whereby no portion of the excess energy stored in said parasitic inductance is dissipated in said first power device;
transferring at least a portion of said excess energy stored in said parasitic inductance to at least one node of said switching apparatus, and
limiting the amplitude of said voltage spikes by means of conducting current in at least one of said transistors.
9. The method of claim 8 wherein at least one of said transistors transfers at least a portion of said excess energy stored in said parasitic inductance to a gate of a second power device to turn on said second power device in correspondence of the turning off of said first power device,
whereby said transferring of at least a portion of said excess energy guarantees no substantial simultaneous conduction of said first and second power devices.
10. The method of claim 8 wherein said switching apparatus is a switching power converter.
11. The method of claim 8 wherein at least one of said transistors is a MOS transistor.
12. The method of claim 8 wherein at least one of said transistors is a bipolar transistor.
13. The method of claim 8 wherein at least one of said nodes of said switching apparatus is the output terminal of a switching power converter.
14. A method of preventing cross conduction between a first and a second series-coupled power device of a switching apparatus comprising:
turning off said first power device coupled to a high current interconnection to a power distribution rail;
monitoring the voltage spike caused by the parasitic inductance of said high current interconnection to said power distribution rail, and occurring in correspondence of the switching transition of said first power device;
utilizing said voltage spike as a signal to the driving section of said switching apparatus indicating that said first power device is turning off;
transferring at least a portion of the excess energy stored in said parasitic inductance to at least one node of said switching apparatus;
turning on said second power device in response to said transferring of at least a portion of said excess energy;
whereby said transferring of at least a portion of said excess energy guarantees no substantial simultaneous conduction of said first and second power devices and very minimum delay between said turning off of said first power device and said turning on of said second power device, and
whereby said transferring of at least a portion of said excess energy is limiting the amplitude of said voltage spike.
15. The method of claim 14 wherein said switching apparatus is a switching power converter.
16. The method of claim 14 wherein at least one node of said switching apparatus is the output terminal of a switching power converter.