SEPIC converter is very inefficienct due to "ringing" currents?

[treez]

Hello,

My 5W sepic converter (LED driver) is only 67% efficient

Vin=6V,
Vout = 31V
Iout = 0.13A
F(sw) = 51KHz
SEPIC inductor = MSD1583-223 (22uH)

Schematic:
https://i45.tinypic.com/2qx9cfr.jpg

All the capacitors are ceramic.

I suspect that due to the ceramic capacitors, there is ringing between the SEPIC inductor’s leakage inductance and the input ceramic capacitors.

How can I snub out this ringing?…..is it by putting a series damping resistor in the path of the SEPIC “primary”?

SEPIC inductor = MSD1583-223
**broken link removed**


(The SEPIC inductor feels hottest, of the power components..the FET and Schottky diode feel barely warm)

Peak FET current is 3.2A and the SEPIC is fairly deeply discontinuous

 

[mtwieg]

Looks like you've discovered the reality of coupled inductor SEPICs. Very high coupling coefficient is often problematic because the tiny leakage inductance will form a resonance with the blocking capacitor, and that resonant frequency may be near or above your switching frequency. If that's the cause here, then you have just a few options: increase your leakage inductance, increase the blocking capacitor, or increase switching frequency. The first option is often best.

 

[treez]

increase the blocking capacitor

I've come away from the hardware for a bit and am on the simulator.....your suggestion of increasing the sepic capacitor works great at reducing the resonating current in the coupled sepic inductors coils.
....i think i will try it in the hardware..what are the pitfalls of using a large value sepic capacitor?

For a coupled inductor sepic, with vin = 6V, and vout = 40V, what is the maximum voltage that the sepic capacitor would see?.....on the simulator its just Vin.
......in a non-coupled-inductor-SEPIC, i notice that the sepic capacitor sees a maximum voltage of vin+vout...............why the difference?

 

[mtwieg]

what are the pitfalls of using a large value sepic capacitor?

High cost and bigger size, mostly.

For a coupled inductor sepic, with vin = 6V, and vout = 40V, what is the maximum voltage that the sepic capacitor would see?.....on the simulator its just Vin.
......in a non-coupled-inductor-SEPIC, i notice that the sepic capacitor sees a maximum voltage of vin+vout...............why the difference?

At steady state, the capacitor should see a DC voltage equal to Vin, regardless of inductor coupling. However, during transient conditions the capacitor voltage may swing greatly. You should thoroughly simulate the converter during startup, shutdown, and transient line and load events.

 

[treez]

during transient conditions the capacitor voltage may swing greatly

i find the coupled inductor sepic sees less transient overvoltages/currents than the non coupled sepic, -do you agree?
Also, for a non coupled inductor sepic, what should be the relationship between the power stage L and C resonant frequency, and the switching frequency, and also the feedback loop frequency?

 

[mtwieg]

Transient response should be determined by the control system, including the feedback loop and any soft start mechanisms, not by the converter itself.

With SEPIC converters, your loop response will be limited by the conjugate RHP zeros. I don't know off hand how the coupled inductors affects those RHP zeros (they still exist, but probably change frequency). Any parasitic resonances formed by the leakage inductances and blocking/bypassing capacitors should be will below the switching frequency of the converter.

 

[treez]

My sepic coupled inductor leakage is 0.46uH, the sepic capacitor is 22uF (16V, 1812 size), so that resonates at 74KHz, and the switching frequency is 35KHz with 40V loads.

So have i had it then?........i am going to have to use an enormous sepic capacitor to get the resonance well below 35KHz.......am i doomed?

 

[mtwieg]

Right, such a combination will give poor results. Using a leakier inductor or a higher switching frequency are more realistic options than increasing C further.

You should look over this thread and the article in the OP. Coilcraft apparently makes higher leakage inductors for this specific reason.

 

[treez]

Thanks mtwieg, i forgot to say thta at 35KHz switching frequency it is in DCM, so i take it from the linked thread that all this resonance stuff doesnt matter for >35KHz?

 

[mtwieg]

Well, it still probably matters somewhat since the RMS current will increase due to ringing. To be precise, you'd want your conduction times to still be much shorter than one resonant period. But when you're that deep in discontinuous mode your ripple currents will probably be high no matter what... that alone might be your problem.

 

[Orson Cart]

You could bump your sw freq up to 70kHz, and use uncoupled inductors...

 

[treez]

Thanks but the non coupled sepic is seriously constrained regarding crossover frequency and the L-C-L resonant frequency........


The non coupled sepic also suffers a ringing, and this is with the two inductors and the sepic cap. This is a low frequency ring which is thus dreadfully hard to snub out....(sometimes you see people using huge electrolytic cans as a snubber cap, in series with a huge axial power snubber resistor.)

So i'd say that ringing is far worse in the non coupled sepic. In the non coupled sepic, there is a relationship between the L-C-L resonant frequency and the crossover frequency, the crossover frequency should be at least three times higher than the L-C-L resonant frequency.......and with the SEPIC, this is not at all easy to do because it tends to go unstable. Some people get round it by using 500KHz plus switching frequencies, which makes it easier to get a high crossover frequency.

This is also why you are saying to do the non coupled sepic with big value inductors.....because it reduces the L-C-L resonant frequency , and makes it easier for you to get your crossover frequency three times higher than the resonant frequency.....though its still damned hard.

 

[Orson Cart]

Hello there, usually the load provides enough damping to give good enough performance, we have designed all manner of Cuk converters at 70kHz, which have much the same power stage dynamics as the SEPIC, and although we have a few control tricks to aid stability from no load to full load, essentially the high sw freq and the use of large value inductors, helps enormously in getting to a good starting point, if you can post your ckt so far no doubt some improvements can be suggested.
Regards, Orson