What causes heat generated by the power transistor operating in TV flyback converter?

[tendi]

What accounts for most of the heat generated by the power transistor operating in a tv flyback converter? What is it directly related to? Can a faulty snubber circuit destroy the power transistor??

 

[v_c]

Re: flyback converters

I am assuming that the transistor is a power MOSFET. The power transistor has two components that contribute to loss. First is the conduction loss -- this is due to the current through the device and the on-resistance of the device. The on-resistance is a function of temperature. Second is the switching loss -- this is due to the charging/discharging of parasitic capacitances of the MOSFET and is proportional to \[ C_p V_{ds}^2 f_{sw}\], where \[C_p\] is the parasitic capacitance, \[V_{ds}\] is the drain-source voltage and \[f_{sw}\] is the switching frequency.

Snubbers (usually series RC) are put across the MOSFET (usually from drain to source) in order to absorb the turn off energy of the MOSFET and keep the drain-source voltage from oscillating. If there is no snubber (or if the snubber is damaged) it is possible to overstress the MOSFET and possibly cause it to work outside is SOA (safe operating area). This overstress may eventually result in permanent damage of the MOSFET or a decrese in its operating lifetime.

I hope that this discussion helps you out.

Best regards,
\[v_c\]

 

[tendi]

Re: flyback converters

Thanks a lot for your explanation. Would you care to discuss the bipolar transistor in a self-oscillating, block-oscillator type flyback converter. This type of p.s.u forms the the bulk of all budget tv's in the market. Invariably when the bjt is replaced, the new one fails due to excessive dissipation. This is the crux of my question.

 

[VVV]

Re: flyback converters

If it is excessive dissipation, then I would carefully look at the driver circuit.
The snubber can add to the dissipation, but not too much. There are generally two snubbers: one clamps the voltage from the leakage inductance (it's an RCD circuit , from collector to +Vcc), the other slows down the voltage rise (usually just an RC, though it can be RCD, from collector to emitter). If the first fails, the transistor can be damaged by overvoltage stress, so you would not see it getting hot, it would just fail. If the snubber across the transistor fails, then yes, it is possible to increase the dissipation during transistor turn-off, causing failure due to power dissipation or secondary breakdown.

Again, I will stress the importance of the driver in this whole thing. Check the components used to drive the transistor (resistors, diode, speed-up cap).
One other thing to check is if there is a component across the current sense resistor. If that failed (low resistance), the transistor peak current will be higher, leading to failure.

If you have a scope, look at the transistor collector voltage (PAY CLOSE ATTENTION TO ISOLATION. YOU ARE WORKING ON THE AC MAINS SIDE!). If you see that towards the end of the on-time the transistor tends to go out of saturation, then very likely you have a problem with the driver.

 

[tendi]

flyback converters

Thanks v_c for the pdf paper and thanks vvv for your discussion. God bless you people for sharing your wisdom.

 

[VVV]

Re: flyback converters

You're welcome. If you can post a schematic, perhaps I can offer a few more thoughts.