160v/5a

[Fob]

So, I must bild 160V/5A from AC. I choose uC3844 on 400/2 kHz, BUP314, L1 = L2 = 30uH, Rs = 0.05 and passive RCD clamp. This is for test purposes, so I put all together with common in/out ground. It works up to 2A. But I have some sine ringing on current ramp over Rs. I make also current transformer - same thing. Also my IGBT burn quietly in lite loads. Any help please.
Thanks

[Fob]

This is the circuit.

[Eva]

Any ringing on Rs voltage waveform is going to be either current ringing from the transformer (resonance of leakage inductance and secondary side diode capacitance) or voltage ringing from the primary side (also due to leakage inductance and diode and IGBT capacitances). Ringing is particularly excited when the converter is operated in continuous-current-mode due to the reverse recovery of secondary side diode. This may be happening above 2A. Did you take that into account?

This IGBT is not intended to be used at 200Khz, 40Khz would be a more reasonable frequency. Excessive losses due to current tail may be causing it to blow, even at low currents. MOSFET may be a better alternative. Also, at 800W and 200Khz, a flyback converter is likely to require active clamping of the primary winding (after IGBT turn-off) because voltage spikes are likely to be quite high, like the amount of energy from transformer leakage inductance that needs to be either recycled or dissipated (.5*L*I^2 on each cycle).

There is nothing inherently wrong with a 800W flyback, you could make it work in the end, but a half bridge converter with the inductor for regulation in the secondary side is likely to give you far less headaches.

[Fob]

Hi Eva,
Actually my question is addressed to you and few other people here so, thanks. I choose this topology because the output voltage is 160V and is more than half of rectified AC 220V. With half bridge I need boost transformer, in push-pull I need hard clamps, and in forward I need transformer bigger as in full bridge. So I decide that the Iprimary p-p is not so dangerous in flayback with pure current mode. I tried with IRFP460 and the heat is much less. Obviously you are right about BUP314, thanks again. And the ringing problem I find is more ground problem than parasitic caps.
What you thing for two transistor flayback? I never did that thing. What will happen to diodes during toff? Every thing what I read for this is said with half mouth.

[mflorin]

Hi Fob,

I would prefer a faster turn-off of mosfets with a local pnp...

http://www.st.com/stonline/products/...re/an/9172.htm

See the application note. It is a 2 transistor forward converter but it could be useful.

[Fob]

ok, I need a litle courage to bild this! Any comment will be in help.
I reduce frequency to 50kHz and the IGBT driver is bipolar. Looks good in simulation but who knows what light and sound will produce. This must be a part from 100V Public Address amplifier.

[mflorin]

I think is something wrong with the .png file inside the archive...

[Fob]

its ok, but Ill post *.pdf later, I am in 4PC in 4 places at time, sorry Flo

[luka]

Nope works ok

[Fob]

Here is in pdf. I found error: there is no connector to emiterq jke ground is from simulation.