Series-resonant LLC Half-Bridge Topology and features
Multi-resonant LLC tank circuit Variable frequency control Fixed 50% duty cycle for Q1 & Q2 Dead-time between LG and HG to allow MOSFET’s ZVS @ turn-on fsw » fr, sinusoidal waveforms: low turn-off losses, low EMI Equal voltage & current stress for secondary rectifiers; ZCS, then no
recovery losses No output choke; cost saving Integrated magnetics: both L’s can be realized with the transformer.
High efficiency: >96% achievable
you are welcome put forward problems about LLC. i will reply you .
thanks
Leo Liang
Multi-resonant LLC tank circuit Variable frequency control Fixed 50% duty cycle for Q1 & Q2 Dead-time between LG and HG to allow MOSFET’s ZVS @ turn-on fsw » fr, sinusoidal waveforms: low turn-off losses, low EMI Equal voltage & current stress for secondary rectifiers; ZCS, then no
recovery losses No output choke; cost saving Integrated magnetics: both L’s can be realized with the transformer.
High efficiency: >96% achievable
you are welcome put forward problems about LLC. i will reply you .
thanks
Leo Liang
I use these quite often. Good choice. 96% is pushing it. The venerable L6599 is a good choice thought there are others that have a built in PFC function.
Discreet L is easier than than integrated magnetics though that is what I most commonly use.
If you look through the threads you will find some good data on it.
Good luck and let us know how things go.
Tony
Discreet L is easier than than integrated magnetics though that is what I most commonly use.
If you look through the threads you will find some good data on it.
Good luck and let us know how things go.
Tony
LLC = no good for Class D audio amplifier supply.
Because LLC is voltage mode and this gives slower transient response then current mode.
-you cannot do LLC in current mode.
-LLC with integrated magnetics = expensive custom built bobbin......audio amp's are low volume and its not worth the expense of custom built bobbin.
-LLC is TERRIBLE at line regulation......the dynamic gain is just too low.
-Sorry but class d audio amplifier supply needs current mode topology with low duty cycle at max load to give high RHPZ frequency.
Because LLC is voltage mode and this gives slower transient response then current mode.
-you cannot do LLC in current mode.
-LLC with integrated magnetics = expensive custom built bobbin......audio amp's are low volume and its not worth the expense of custom built bobbin.
-LLC is TERRIBLE at line regulation......the dynamic gain is just too low.
-Sorry but class d audio amplifier supply needs current mode topology with low duty cycle at max load to give high RHPZ frequency.
Have to disagree with all your points.
Most ATX (desktop) power supplies that meet the 80+ efficiency guidlines run LLC's, You highest efficiency server supplies are going LLC. You newer Adapters from 130W-240W will all be LLC.
Voltage Mode DOES NOT=Poor response. Power Integrations has been making voltage mode controllers for years and making a killing off them.
As to LLC's having terrible line regulation... See above. We have had this discussion before so no need to rehash it here. Dynamic Gain too low? What is that exactly? Do you mean the gain at DC is too low on the Bode plots? I average about 12dB gain at DC, 60 Degree phase margin and 10dB gain margin which is comparable to all the other topologies I have used because that is the Engineering Spec that I have to meet, regardless of topology choices I make.
Loop response handles the average power on a PSU. That means any transient greater than the loop crossover frequency has to be handled by the output capacitance until the loop can catch up. Look at the old tube amps as a case in point. They ran at 60Hz and used big L-C's to handle the ripple. there wasn't any "Line regulation" or transient response, just output C to hold things up.
For Audio power Amps, choose a amp topology with an inherently good PSRR and do a reasonably good job on the power supply and you shouldn't have an issue.
EEM2AM have you actually built any LLC's or are you only doing simulations? This isn't a slam and I am not trying to be offensive. If I seem abrasive, I must apologize. I am just curious as to why you are so dead set against the topology. I have used this topology from 90W to 2KW and I do know what I am talking about when it comes to the topology.
All that being said, there are other topology choice one can make and achieve similar results. I do not knock any of the different topologies as each has their place in my toolbox.
Tony
Tony
Most ATX (desktop) power supplies that meet the 80+ efficiency guidlines run LLC's, You highest efficiency server supplies are going LLC. You newer Adapters from 130W-240W will all be LLC.
Voltage Mode DOES NOT=Poor response. Power Integrations has been making voltage mode controllers for years and making a killing off them.
As to LLC's having terrible line regulation... See above. We have had this discussion before so no need to rehash it here. Dynamic Gain too low? What is that exactly? Do you mean the gain at DC is too low on the Bode plots? I average about 12dB gain at DC, 60 Degree phase margin and 10dB gain margin which is comparable to all the other topologies I have used because that is the Engineering Spec that I have to meet, regardless of topology choices I make.
Loop response handles the average power on a PSU. That means any transient greater than the loop crossover frequency has to be handled by the output capacitance until the loop can catch up. Look at the old tube amps as a case in point. They ran at 60Hz and used big L-C's to handle the ripple. there wasn't any "Line regulation" or transient response, just output C to hold things up.
For Audio power Amps, choose a amp topology with an inherently good PSRR and do a reasonably good job on the power supply and you shouldn't have an issue.
EEM2AM have you actually built any LLC's or are you only doing simulations? This isn't a slam and I am not trying to be offensive. If I seem abrasive, I must apologize. I am just curious as to why you are so dead set against the topology. I have used this topology from 90W to 2KW and I do know what I am talking about when it comes to the topology.
All that being said, there are other topology choice one can make and achieve similar results. I do not knock any of the different topologies as each has their place in my toolbox.
Tony
Tony
i have done LLC for high power LED light which powered from PFC output voltage. -Its constant load with constant Vin, so LLC is the natural choice as its 150W too.
But for SMPS's requiring good transient response, resonant converters are not a wise choice.
Resonant converters are good for steady state....Class D supply is always going from no-load to max-load, and you totally loose the benefits of resonant operation in that case.
Its much better to supply a Class D amplifier with a smps with a fast transient response.....because its a lot harder to do a Class D amplifier whih can handle a lot of ripple on the supply rails.......Doing a Class D AMplifier with a good PSRR is a lot harder than doing one with less PSRR.
As you say, we already discussed LLC converters for class d supply in about early january 2012 on this forum.
for class d supply, you want Current mode, with low duty cycle so you get a high RHPZ frequency, and thence a great transient response.
But for SMPS's requiring good transient response, resonant converters are not a wise choice.
Resonant converters are good for steady state....Class D supply is always going from no-load to max-load, and you totally loose the benefits of resonant operation in that case.
Its much better to supply a Class D amplifier with a smps with a fast transient response.....because its a lot harder to do a Class D amplifier whih can handle a lot of ripple on the supply rails.......Doing a Class D AMplifier with a good PSRR is a lot harder than doing one with less PSRR.
As you say, we already discussed LLC converters for class d supply in about early january 2012 on this forum.
for class d supply, you want Current mode, with low duty cycle so you get a high RHPZ frequency, and thence a great transient response.
LLC extensive use in the amplifier of class D as its higt efficiency and no output inductaner,of course not for class A amplifier ,
Why care LLC?
1.Features of the resonant convertera. Normally soft switching(ZVS or ZCS)b. High current and voltage stress2. Semiconductor development: (Deceasing Rdson, falling time tf, while Cosskeep the same level)3. High power density, high frequency trend(Switching loss is the main issue)
i have one question why cannot upload picture?
i have the simulation of LLC model for transim tool ,if you needed pls call me @ leo@1kv.cc
thanks
Leo
Why care LLC?
1.Features of the resonant convertera. Normally soft switching(ZVS or ZCS)b. High current and voltage stress2. Semiconductor development: (Deceasing Rdson, falling time tf, while Cosskeep the same level)3. High power density, high frequency trend(Switching loss is the main issue)
i have one question why cannot upload picture?
i have the simulation of LLC model for transim tool ,if you needed pls call me @ leo@1kv.cc
thanks
Leo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.