The values depend on the leakage inductance (Le) of the transformer. Off the top of my head select C such that the resonant frequency of the LC circuit is 10 times that of the switching frequency. Fres=(1/2*pi)*1/(sqrt(L*C)). Select R=sqrt(L/C); it might be 2*sqrt(l/c)
Great link indeed 🙂
A more elaborate method includes finding out the characteristic RLC values of the circuit (no snubber), and then using simulation software to find out a RLC network capable of providing critical damping (single cycle with no oscillation) with minimum losses. Series inductance in snubbers is actually useful as long as it's not too far from the optimum value.
A more elaborate method includes finding out the characteristic RLC values of the circuit (no snubber), and then using simulation software to find out a RLC network capable of providing critical damping (single cycle with no oscillation) with minimum losses. Series inductance in snubbers is actually useful as long as it's not too far from the optimum value.
what ridley doesnt bother to explain is whats going on and how to figure it out when you DONT know either L or C.
this can also be found in an old unitrode app note, but this is how I snub ringing circuits:
- measure ringing period T1, using nice low capacitance scope probe.
- T1 = 2*pi*sqrt(Lo*Co) where Lo, Co are the elements responsible for the ringing.
- add some capacitor C1 to the circuit - eg across the primary winding if we are snubbing that, or across a diode etc.
- this capacitor will increase the resonant period,
T2 = 2*pi*sqrt(Lo*(Co+C1))
(if T2 doesnt really change this is because either C1 << Co, or you stuck the cap in the wrong place)
the ratio of these two periods T2/T1 = sqrt((Co+C1)/Co)
so (T2/T1)^2 - 1 = C1/Co
thus you now know Co, and from T1 can calculate Lo
then calculate Zo = sqrt(Lo/Co). this is the value to use for the snubber resistor Rs
The snubber capacitor Cs needs to be 2x or 3x Co. Using Cs < 2Co decreases the amount of damping; using Cs > 3Co wont increase the damping, but does increase Fsmps*Cs*Vpk^2 losses
- it is clear from the above maths that if you keep adding C1 until T2 = 2*T1 then C1 = 3*Co
measure the peak snubber cap voltage Vpk. average losses are Fsmps*Cs*Vpk^2
measure the peak snubber resistor voltage, and calculate Ppeak = Vpeak^2/Rs. make sure your snubber resistor can handle this
note also that if you know your scope probe capacitance, you can correct for it when calculating snubber components
this can also be found in an old unitrode app note, but this is how I snub ringing circuits:
- measure ringing period T1, using nice low capacitance scope probe.
- T1 = 2*pi*sqrt(Lo*Co) where Lo, Co are the elements responsible for the ringing.
- add some capacitor C1 to the circuit - eg across the primary winding if we are snubbing that, or across a diode etc.
- this capacitor will increase the resonant period,
T2 = 2*pi*sqrt(Lo*(Co+C1))
(if T2 doesnt really change this is because either C1 << Co, or you stuck the cap in the wrong place)
the ratio of these two periods T2/T1 = sqrt((Co+C1)/Co)
so (T2/T1)^2 - 1 = C1/Co
thus you now know Co, and from T1 can calculate Lo
then calculate Zo = sqrt(Lo/Co). this is the value to use for the snubber resistor Rs
The snubber capacitor Cs needs to be 2x or 3x Co. Using Cs < 2Co decreases the amount of damping; using Cs > 3Co wont increase the damping, but does increase Fsmps*Cs*Vpk^2 losses
- it is clear from the above maths that if you keep adding C1 until T2 = 2*T1 then C1 = 3*Co
measure the peak snubber cap voltage Vpk. average losses are Fsmps*Cs*Vpk^2
measure the peak snubber resistor voltage, and calculate Ppeak = Vpeak^2/Rs. make sure your snubber resistor can handle this
note also that if you know your scope probe capacitance, you can correct for it when calculating snubber components
That looks like good stuff, Terry.
Check out this Maxim appnote, which also has a similar procedure to determine R and C for a snubber, using scope measurements:
http://www.maxim-ic.com/appnotes.cfm/appnote_number/3835
And here is the rest of my snubber-design appnote/article collection:
http://www.cornell-dubilier.com/design.pdf
http://www.hagtech.com/pdf/snubber.pdf
http://www.ridleyengineering.com/snubber.htm
http://archive.chipcenter.com/circuitcellar/november00/c1100rp58.htm
Check out this Maxim appnote, which also has a similar procedure to determine R and C for a snubber, using scope measurements:
http://www.maxim-ic.com/appnotes.cfm/appnote_number/3835
And here is the rest of my snubber-design appnote/article collection:
http://www.cornell-dubilier.com/design.pdf
http://www.hagtech.com/pdf/snubber.pdf
http://www.ridleyengineering.com/snubber.htm
http://archive.chipcenter.com/circuitcellar/november00/c1100rp58.htm
- Status
- Not open for further replies.
- Home
- Amplifiers
- Power Supplies
- Calculation for Snubber Value