[SOLVED] Understanding oscillations at the end of an inductance energy discharge

[rgb83]

Hello Edaboard members,

I'm trying to understand the reason of the oscillations that appear at the end of the fast switch off phase of an inductive load(see the zoom of the picture). See attached Ltspice circuit for reference (rename it in *.asc).

I understand the usage of the Zener diode used as freewheeling diode to provide a path to suppress te voltage spike generate by an inductive load whenever its supply is removed.

Anyway where those oscillations come from? Is there something that enter in resonance?

What are the mathematical equations to explain this behavior?

 

[FvM]

Resonance of L1 with Cds + Czener. The oscillation has relative low frequency and low energy, it doesn't hurt much.

 

[rgb83]

Can you help me in writing down the formula of the equivalent circuit in this condition?

Thanks for your answer

 

[rgb83]

Guys someone can help me?

 

[D.A.(Tony)Stewart]

- BSC0904NSI spec has a 30V Zener internal to Vds
- yet you have a 60V zener (D1) added as shown in the simulation with 60V pulse.

- regarding resonance, it appears to be 125kHz when switch turns off which with 3425uH is approx 473pF similar to the values of Coss of MOSFET and diode capacitance at 20V

- the decay rate is related to the Q or bandwidth of the bandpass resonance , a snubber could reduce the resonance if EMI is an issue.

 

[rgb83]

Hi SunnySkyguy,

your answer solved a lot of doubs I had.

I've just a couple of other questions:

1) Why the voltage is not clamped to 30V (internal MOS zener) instead of 60V of the external Zener?
2) How should I connect the snubber to smooth the oscillations?
3) Do you think those oscillations should be there also if will connect a real inductive load (like DC motors, valves and so on)?

Again thanks upfront!

 

[FvM]

1) Why the voltage is not clamped to 30V (internal MOS zener) instead of 60V of the external Zener?

If you look at the MOSFET model parameters, you see that the breakdown voltage isn't modelled at all.

 

[D.A.(Tony)Stewart]

Hi SunnySkyguy,

your answer solved a lot of doubs I had.

I've just a couple of other questions:

1) Why the voltage is not clamped to 30V (internal MOS zener) instead of 60V of the external Zener?
2) How should I connect the snubber to smooth the oscillations?
3) Do you think those oscillations should be there also if will connect a real inductive load (like DC motors, valves and so on)?

Again thanks upfront!

1) SImulation error
2) No need, wait for a real application with high power,
3) this is very low power and often load will dampen turn off ringing.

 

[rgb83]

If you look at the MOSFET model parameters, you see that the breakdown voltage isn't modelled at all.

You're absolutely right! Thanks for pointing it out!