uc3852 pfc cct one more try.

Seems to me that what you can do is.....

Ramp valley is 0.2V. Ramp peak is, arbitrarily, chosen as 9V. So ramp valley to ramp peak is 8.8V.

Let's say your minimum line is 90V and your output voltage is 400V. Since this is critical mode conduction the operating duty cycle is

d = (VOUT - VIN)/VOUT

Or 0.775.

Pick your minimum operating frequency, say 40KHz which gives your period as

25uS so the on time is 19uS.

Pick CT and call it 470pF. 8.8V in 19uS on 470pF makes the timing current

217uA so your timing resistor is 22K.

Say your output power is 100W. 100W from 90V is an input current of 1.1A. With critical mode conduction the peak current is twice this or 2.2A. For

2.2A with 90V and 19uS you required an inductor of 750uH.

DNA

Thanks. You make it sound so easy.

Should d not be 0.682 , not 0.775 The 90V is rms which is 127V Pk. I still can't explain the values thay have come up with in the datasheet app. The app note u132 has values that come closer to what you have.The inductor formula comes out at 688uH (90V rms , 100W , 400V , 40Khz) , which is at least in the same ballpark.

getting

inductance

valley

Since

as

current

1.1A.

For

I waved a bit of a wet finger in the air.

If you take your peak voltage at 127V with VOUT = 400V then, as you say the duty cycle becomes 0.683. At 40KHz the on time becomes 17uS. For 100W the 'peak line' current is 1.555A. Double that for 3.11A which, at 127V requires an inductor of 694uH.

At 90V with 17uS and 694uH the peak inductor current is 2.2A with an RMS of

1.1A so the sums work out as 100W.

That means my choosing the RMS value of voltage for duty cycle was wrong and you may curse me, but it wasn't that bad......

If I beat my head about I might begin to see how using that assumption would require continuous operation when the instantaneous line voltage rises above

90V.

Since I was wrong then using the proper sums and deriving the big sum.

d = (VOUT - VP)/VOUT

Ton = d/Fs

L = VP.Ton/ILpk

Ilpk = 2.Ilinepk = 2.SQRT(2).Ilinerms

Ilinerms = Pout/Vlinerms

Vlinerms = VP/SQRT(2)

Ilinerms = SQRT(2).Pout/VP

Ilpk = 2*SQRT(2).SQRT(2).Pout/VP = 4.Pout/VP

L = VP.Ton.VP/4.Pout = VP^2.Ton/4.Pout = VP^2.d/4.Fs.Pout = VP^2(VOUT - VP)/4.Fs.Pout.VOUT

..... 688uH

Well, it's not as if my previous sums agree with the 'proper one' but.

You can tell me where I went wrong.

DNA

Thanks for all the help DNA. I have enough to go on now.I am going to assume that the values that TI have in the datasheet app are incorrect. All the values you come up with tie in with the unitrode app note example.I am looking to build a

400W pfc , but am going to build the 100W version first to get a "Feel" for it. I assume that for the 400W version , most things don't change to much.The inductor will obviously change and some of the power components may have to be beefed up a bit. Chhers , thanks for the help. Rob