Flyback with "active" clamp is OK?

[eem2am]

Hello,

I am designing a 330W Flyback. (Vin = 180-373VDC, Vout = 80V, 4A)

The problem is that at light load the RCD clamp dissipates about 2.8W.

So i have come up with this switchable RCD clamp.......

http://i43.tinypic.com/2h71de9.jpg


Can you see any problems with this?

Is it OK?

Here is the LTspice simulation file in .txt formar, so converting to .asc allows it to be run in LTspice


2shared - download ACTIVE CLAMP__TXT.txt

[mag]

Hi,
I think it should work. i have tried something similar to clamp
an output rectifier but using a bjt instead of a mosfet.
Just add a 15v zener diode between gate and source to protect the gate from transient overvoltage.
Anyway you are just trying to move some of the clamp dissipaion on the fet but the total dissipation will be always the same.

From the pitcure I see that your turn ratio is around 3:1 so the reflected voltage is around 120v.
You are using a 800V fet so why not let the clamp swing higher and reduce its dissipation?

On top of that you are using another 800v fet and a bunch of power resistors and zeners so why not go to 2 switch flyback?
You will have rougly the same cost as this solution and you get rid of all your clamp problems.
The fets can be 500v types, less expensives and better perfomances.

Or why not go to a real active clamp flyback with the benefits of zvs switching?

Single switch flyback at that power level is tough to do properly and the performances are poor.

Ciao

-marco

[eem2am]

The transistor M2, will actually be implemented with a cheap (0.04$ @ 5000 pces) KSP44

http://www.fairchildsemi.com/ds/KS/KSP44.pdf

The power resistors will be 4 pieces of MVM5 ($0.119 @ 5000 pces)

......Then thers the cap and the diode...this is definetely cheaper than 2-switch-flyback which has the custom wound gate drive transformer, as well as PNP turn off circuit.

I saw UCC2891 for flyback with active clamp , but you again need a high side drive (or expensive low side PFET.....plus you have to have well toleranced main transformer because the FET delay depends on the exact value of leakage.

I can assure you that the RCDQ Clamp (the one with the transistor in it) gives about 2W less total dissiaption at light load than the RCD clamp...( i am including FET and snubber dissipatiomn there too)........................so there is no doubt that the RCDQ clamp dissipates significantly less in light load.

I dont agree that single switch flyback at 330W is hard to do.....
Loads of 800V TO220 FETs are cheaply available with metal tab and sub 1 ohm rdson..
....with a 2C/W heatsink, and max ambient of 30C, then this FET would be able to dissipate 20W no problem.

...and lats face it, its an audio supply so it just isnt going to dissipate 330W all the time.......remember its a factor of music that it is AC........the sound waves have to compress and retract the air................so this means you are not going to be at peak all the time.........you are constantly going to backing away from the peak power.......realistically , over any period of any time you choose, you are going to be dissipating less than 1/3 maximum.......so thats 110W in this case.

....i just cannot justify an active clamp flyback, or even a two switch flyback for 110W..........and the 110W is an absolute maximum.......really we are talking around 45W with really really heavy guitar playing.

With normal music, it is actually physically impossible to pull maximum power continuously..............remember the 330W corresponds to the compression of the air.......but music needs a train of compressions and rarefactions............thus you see that 330W max power cannot possibly occur all the time....

Active clamp flyback for 45W realistic load?......sorry but no
Two switch flyback for 45W realistic load?........sorry but no