Electronics > Beginners
LLC vs LCC Converters for High Output Voltage
<< < (3/6) > >>
state_of_flux:

--- Quote --- They're saying you can double the power with full vs. half-bridge (or vs. push-pull).  Obviously the transformer needs to double as well. 
--- End quote ---

Oops, clearly lost in translation there. My bad for misunderstanding.  |O


--- Quote --- Note that PP has the duty cycle penalty (each half of the primary is used at D = 50%, so sqrt(2)-1 ~= 40% more primary copper is needed), but not the flux penalty.  Half bridge has neither, but requires coupling caps or split supplies.

Nice thing about LLC is the coupling caps come for free (you need the resonant cap(s) regardless), so half bridge is popular there.
--- End quote ---

With the half bridge converter however, does it not only present half the input voltage to the transformer primary? Meaning an even larger number of turns on the secondary and thus more stray capacitance. As you say I very often see the LLC used in the half-bridge topology in lower power, step-down applications - but rarely see them in step-up, high output voltage applications!
T3sl4co1l:

--- Quote from: state_of_flux on May 13, 2019, 01:41:12 pm ---Exactly. This is the most important trade-off for me, finding the right switching frequency that reduces the transformer/passive component sizes while ensuring proper operation and utilising the sec. capacitance. The winding configuration of the EHT by Multisync CRT is really interesting - I will look into this. I also looked into the use of planar transformers for high voltage transformer since they allow tighter control of transformer parasitics -  but it's unlikely they can be used in voltage outputs at this level without causing significant issues of proximity effect between the layers.

--- End quote ---

Yeah, planar transformers have close coupling -- wide, broadside facing traces.  No good at HV.  Fantastic at low impedances (typically low voltages and high currents, and modest ratios).

That's the other thing, high ratios suck, because of trace width and space limits.  Besides which, a high ratio winding necessarily has a high impedance, but it can't, because planar.



--- Quote ---That's interesting - i've never actually heard of transformer external tuning before. I'm unsure what you mean by the last point in regards to the LCC being acceptable - is this because it can provide tuning in some way?

--- End quote ---

Yeah, because tuning is either setting F_res just so (F_sw stays fixed), or varying F_sw to suit (F_res stays fixed).  Or both I suppose, but who would do that.

Typical resonant controllers vary output power by controlling F_sw (which, I think you already knew, so this is just going to be a "duh"?  It's alright. :) )



--- Quote from: state_of_flux on May 13, 2019, 01:45:33 pm ---With the half bridge converter however, does it not only present half the input voltage to the transformer primary? Meaning an even larger number of turns on the secondary and thus more stray capacitance. As you say I very often see the LLC used in the half-bridge topology in lower power, step-down applications - but rarely see them in step-up, high output voltage applications!

--- End quote ---

Ya.  Half primary turns. :-DD :-+

Tim
state_of_flux:

--- Quote --- Typical resonant controllers vary output power by controlling F_sw (which, I think you already knew, so this is just going to be a "duh"?  It's alright. :) )
--- End quote ---

Of course - it's fine, I appreciate your help. It was to my understanding though that if a pre-regulator is used you wouldn't actually need to vary the switching frequency at all - optimizing it for a certain operating point. I'm not sure if you have missed my earlier reply to you? Apologies if you did not.


--- Quote --- Ya.  Half primary turns. 
--- End quote ---

Phew, glad I didn't get this one wrong as well! Otherwise I think even the Beginners forum wouldn't be suitable for me!  :-DD I'll stick to full bridge then.

Thanks Tim.
MagicSmoker:

--- Quote from: state_of_flux on May 13, 2019, 01:25:06 pm ---...however as you mention yourself the best way to mitigate the issues with HV designs is forming a sinusoid or lowering the switching frequency. The fly-back is still going to present a square waveform to the transformer...
--- End quote ---

The flyback primary has a square voltage impressed upon it, but the secondary acts as a current source. Remember, the primary and secondary never conduct at the same time in a flyback so the transformer is really a coupled inductor. Consequently, the main design criterion is that Amp*Turns are conserved between primary and secondary. E.g. - if applying 300V to a 10t primary causes current to ramp up from 0A to 10A then upon the primary side switch turning off the secondary diode will attempt to deliver a peak current of 1A (ramping down to 0) for a 100t winding, or 0.1A for 1000t, etc. This is what gives the flyback more flexibility with respect to turns ratio than forward mode converters, though keep in mind that when the switch is on the primary voltage reflects to the secondary according to the turns ratio and the inverse occurs when the diode is on, and this sets the voltages that each device must withstand (plus the input and output voltages, for switch and diode, respectively).

Which is sort of a long winded way of saying the flyback - particularly the two-switch variant - is still very much in contention here, especially since it will be far easier (though not easy!) to design compared to any kind of (quasi)resonant topology.


--- Quote from: state_of_flux on May 13, 2019, 01:25:06 pm ---In literature, similar projects to mine do stack multiple secondaries and connect them together in series at the output, which reduces the stray capacitance and allows use of lower voltage rated components - but to what extent can this value be mitigated without affecting circuit performance is the issue.
--- End quote ---

This is commonly known as "pi winding" for historical reasons (though I don't have the foggiest clue why) and while it does reduce interwinding capacitance, it tends to greatly increase leakage inductance. Remember the first law of engineering: There Ain't No Such Thing As A Free Lunch!


--- Quote from: state_of_flux on May 13, 2019, 01:25:06 pm ---
--- Quote --- Also, resonant converters tend to be intolerant of either open-circuit loads (series resonant) or short-circuits/overloads (parallel resonant), further restricting their usefulness.
--- End quote ---

Exactly - I did look into auxiliary circuits to improve the low-load performance of series-resonant circuits, however my converter should operate at full load for a large amount of time, so poorer performance at low-load conditions might not effect overall efficiency too much?
--- End quote ---

It's not so much a matter of lower efficiency if you open-circuit a series resonant converter or short-circuit a parallel resonant one, it's more a case of instant switch destruction, though it is typically easier to protect the former from abuse than the latter.


--- Quote from: state_of_flux on May 13, 2019, 01:25:06 pm ---
--- Quote --- Which reminds me of another topology that would be a good fit - the compound buck current-fed full bridge. This uses a width-modulated buck converter feeding a full bridge run at a fixed 50% duty cycle 
--- End quote ---

This sounds like a good possibility. I did look into two-stage conversion topologies with resonant converters at fixed frequency. In this case, could it be possible to modify the full bridge to achieve some soft switching? Possibly the use of snubber circuits across the MOSFET switches?
--- End quote ---

The bridge switches operate under very benign conditions and it would be best to use IGBTs because they have much lower parasitic capacitances than equivalently powerful MOSFETs and their lazy turn-off is an advantage here as you need overlapping conduction of each bridge leg. It is the buck switch(es) that have to endure exceptionally unfavorable turn-on conditions, and which benefit the most from a lossless snubber (turn-off for the buck switches is relatively benign, however).


--- Quote from: state_of_flux on May 13, 2019, 01:25:06 pm ---
--- Quote --- One might think the full bridge could handle twice the power with the same switches, and it can, but that would require doubling the conductor area of the transformer windings so you end up needing a bigger transformer (in window and/or core area)
--- End quote ---

This is actually something I hadn't thought about. The half-bridge is definitely not applicable here but I will definitely reconsider other uni-polar topologies if they can reduce transformer size.
--- End quote ---

Not the best example I could have used - two bipolar-driven transformer topologies - as the real point I wanted to emphasize is that you can't necessarily double the flux swing in a bipolar topology - and therefore double the power from a given transformer - because iron losses go up exponentially with flux swing, and that giving the forward converter a switch that is just as powerful as two switches in a bridge narrows the gap in maximum power output between them quite a bit.

EDIT- clarifications
state_of_flux:

--- Quote --- This is what gives the flyback more flexibility with respect to turns ratio than forward mode converters, though keep in mind that when the switch is on the primary voltage reflects to the secondary according to the turns ratio and the inverse occurs when the diode is on, and this sets the voltages that each device must withstand (plus the input and output voltages, for switch and diode, respectively).
--- End quote ---

I suppose then a design trade-off would be selecting the turns ratio as to reduce the parasitic capacitance while also ensuring the voltage blocking capabilities of the MOSFET switches in the primary and the diodes in the secondary don't become unreasonably large? Maybe again the use of voltage multipliers could be adopted? What I like about the fly-back is how suitable it is for multiple outputs and regulation of them - my application can have anywhere from 2+ number of outputs.


--- Quote --- The bridge switches operate under very benign conditions and it would be best to use IGBTs because they have much lower parasitic capacitances than equivalently powerful MOSFETs and their lazy turn-off is an advantage here as you need overlapping conduction of each bridge leg. It is the buck switch(es) that have to endure exceptionally unfavorable turn-on conditions, and which benefit the most from a lossless snubber (turn-off for the buck switches is relatively benign, however).
--- End quote ---

But aren't IGBT's only capable of low switching voltages, say, 30kHz? Would that not mean a significantly larger transformer size, or is there a point in which an increase in frequency no longer has a significant impact on size?
The primary design criteria for me is to ensure a small, lightweight solution.

  So, essentially, incorporating some kind of lossless snubber to the buck-converter for turn-on will allow one to reach a high efficiency solution despite using a hard-switched full-bridge as the main regulator to the HV transformer?


--- Quote --- Not the best example I could have used - two bipolar-driven transformer topologies - as the real point I wanted to emphasize is that you can't necessarily double the flux swing in a bipolar topology - and therefore double the power from a given transformer - because iron losses go up exponentially with flux swing, and that giving the forward converter a switch that is just as powerful as two switches in a bridge narrows the gap in maximum power output between them quite a bit.
--- End quote ---

Just so that I understand - what you are saying is that uni-polar topologies can reach similar power levels to bi-polar counterparts just by incorporating a second switch - while reducing the iron losses due to reduced flux swing (since they only occupy one quadrant of the hysteresis loop)?




Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod