magnitude of the spike in full bridge rectifier with snubber

[yelstor]

magnitude of spike in full brdge rectifier with snubber

hello all.
I designed a full bridge smps, with secondary full bridge rectifier. The circuit work in hard switching mode.
The schematic is as follows

The circuit present a big spikes. I design snubber for solve this.
Secondary voltage without snubbers

and secondary voltage with snubber

This is a RC snubber with values R= 150 ohms and C=680pF
I try many combinations of resistors, and the spike is always present(although it is smaller).
Is it normal that the spike never disappears?
there some other method, for reduce the spike?
thanks and regards

 

[Easy peasy]

Re: magnitude of spike in full brdge rectifier with snubber

Is it normal that the spike never disappears? - yes
there some other method, for reduce the spike? - yes but involves a lot of bits for energy recovery
thanks and regards - welcome

- - - Updated - - -

Put 22 ohm to the gate of EACH fet, this will reduce the spike a little also....

 

[yelstor]

Re: magnitude of spike in full brdge rectifier with snubber

thanks for the answer.
I forgot mention that I place snubbers in diodes rectifiers in the secondary side. I don't place the snubber in FETS because this reduces little the spike, and losses are really big.
You know some aplication note or book which mention the metod for energy recovery in snubbers? or some circuit?

 

[andre_luis]

Re: magnitude of spike in full brdge rectifier with snubber

I don't place the snubber in FETS because this reduces little the spike, and losses are really big.

Is the 40uH inductor in series with both sides of the transformer strictly necessary, or is just part of the manufacturer's model ? Anyway, the the losses should be not so expressive if snubbers are correctly sized.

 

[yelstor]

Re: magnitude of spike in full brdge rectifier with snubber

I place the inductance of 40 uH based in pi model for transformator (see this page: https://softsolder.com/2009/02/16/finding-transformer-pi-model-parameters/ and the book "switch-mode power supply simulation by steven M sandler "), anyway in the simulation I don't see same response that I have in the real circuit.
I designed the snubber whith parasitic capacitance of the diodes STTH30R06W (more or less 58pF) and calculating inductance parasitic (see AN11160 NXP). I measure the oscilation with the oscilloscope. The value for RS calculated is 218 ohms. I place 240 ohms resistor because present better response.
The Spike is product the output choke is keeping the diodes forward biased during the dead time. If I remove the output choke, the circuit do not work.
I dont know the function for this choke, aside from filtering.
why does it need this choke?

 

[BradtheRad]

Re: magnitude of spike in full brdge rectifier with snubber

On the chance this could help...

Suppose you put two DCR snubbers in anti-parallel? One would dissipate energy from the positive waveform, the other from the negative.

The DCR snubber appears to be an effective type of flyback snubber. It has to dissipate DC energy bursts. That is why your AC inverter needs two in anti-parallel.

Where would they do the most good, at the primary or secondary? I'm not sure.

 

[Easy peasy]

Re: magnitude of spike in full brdge rectifier with snubber

The extra chokes act like increased leakage inductance for the Tx, without perfect diodes this will worsen the spikes at full load, yes you need the o/p choke, SiC diodes would help, or a ckt with much lower effective leakage...!

 

[andre_luis]

Re: magnitude of spike in full brdge rectifier with snubber

Are you simulating with an ideal power supply, or it has inner series resistence ? If it is the case, I suspect that the addition of non electrolitic capacitors in paralel with Vs could improve the smoothing of the spike.

 

[RichardHead]

Re: magnitude of spike in full brdge rectifier with snubber

Your leakage inductance values (40uH and 11uH are HUGE) for a decently wound interleaved power transformer. From a practical point of view wind a sample transformer using the actual bobbin and cores and then short circuit the secondary and measure the primary inductance, This is a measure of the leakage inductance. You will quickly discover that interleaving (Winding half primary then all secondary then half primary over secondary) has a huge effect on leakage L.