Current Mode control vs Voltage mode control for SMPS

[treez]

For a Full Bridge SMPS in current mode, as opposed to Voltage Mode, we can say that with Current Mode, we no longer suffer the double pole of the output LC stage, because current mode control cancels out one of these poles, meaning that we can more easily compensate the feedback loop to attain stability..is this correctly worded?

 

[CheeseToastie]

For a Full Bridge SMPS in current mode, as opposed to Voltage Mode, we can say that with Current Mode, we no longer suffer the double pole of the output LC stage, because current mode control cancels out one of these poles, meaning that we can more easily compensate the feedback loop to attain stability..is this correctly worded?

That's one of the suggested advantages. More importantly current mode control, peak or average, avoids flux walking in the transformer's primary magnetising inductance preventing saturation.

This is a problem in voltage mode control for both push-pull and full-bridge converters and it has been known about for some time.

I really have to question why people repeatedly 'insist' on presenting circuit solutions for push-pull/full-bridge converters using voltage mode controllers when there are better solutions available.

Can we move on from the TL494/UC3525 because that sort of stuff only serves to break other people's hearts when they try to recreate something someone else got lucky with.

 

[treez]

Thanks, as you know, what you speak of is even worse with the half bridge, and in reality , the half bridge should be banished.
Instead, do Two transistor forward, full bridge or LLC , or active clamp forward, or interleaved forward or flyback.

 

[CheeseToastie]

Yes.. A half-bridge suffers from the opposite problem. Unless you include flux-balancing windings of take other actions then when trying to use current mode control the 'soft' capacitor centre return is subject to drift. With voltage mode control that drift acts to compensate against flux walking.

However to implement current limiting you have to force a hard shutdown and soft restart. There is no reliable way to run the converter continuously in a current limited condition unless you have implemented something like flux-balance windings in which case you will like as not be running current mode control anyway.

I'm not sure I would banish it. Different things require or have preferred methods of control. I'm not suggesting that the topologies themselves are banished. Rather that in the case of push-pull, full-bridge and others voltage mode control has been superseded by current mode control.

It's just slightly depressing to see people following the tired old path down to grief using something that has or should have been superseded.

 

[mtwieg]

CMC doesn't completely eliminate one of the poles, it just moves it to a higher frequency. The exact frequency is hard to solve, but it comes down to solving the SSA model of the circuit.