Interleave phases , multiphase converter..

[Trishool]

6- and 12-phase circuits are quite common in power electronics, feasibility isn't the point.

........ However, paralleling smaller inverter's in- and outputs still reduces ripple currents and lower PWM frequent harmonics, when the PWM generators are synchronized and phase shifted. In so far it's the next best solution.

What type of controller/control system is suggested here. If we dont synchronize the isolated section what potential problems can I face apart from inductor size , will it hinder the efficiency.

Thanks everobody for your valuable help....

 

[FvM]

No phase shift without synchronization.

 

[Trishool]

No phase shift without synchronization.

Yes only between HF/Isolated phases.

 

[dustmote]

I was recently considering the implementation of what would seem to be a similar idea {in my case, a 4 phase interleaved buck converter--- the reason for it's creation being to lower output ripple --which would lead to lower filter component values -- and enable a much larger total power throughput with less design problems than a correspondingly larger single phase buck converter}

Although there are multiphase buck/boost controller IC's in existence, such as those from Texas Instrument like the TPS40090.... It is often advantageous to create a more appropriate solution for your particular needs.... However it can be very useful to look into the operation and design of the existing solutions in order to gain perspective on what important design considerations and topologies you may need to be aware of.

In my particular case, I realized that generating the interleaved clock signals via the use of flip-flops arranged into a twisted ring counter, also known as a johnson counter, ~Please see wikipedia about this subject~ became somewhat problematic in light of the desire to be able to implement dynamic current balancing among the various parallel buck/boost units. In my case, it ended up being an acceptable enough solution to simply test and sort the components that I was using and then to mount all the MOSFETs on a heatsink together so that their temperatures could equalize among themselves..... Trying to modulate each phase's MOSFET duty cycle based on current feedback from that particular phase would have been much more complex rather than the normal method of simply modulating the overall system duty cycle based on a single feedback source.

It is often found to be helpful to mount parallel semiconductors in thermal contact with one another so as to help equalize current sharing between them. There has been much writing done about the subject of load balancing in situations such as these that I can link to if you require.

You may also find a simple circuit demonstrating the principle of this technique of johnson counter mediated multi-phase clock signal generation {in this case a 4 phase implementation} at this website: https://www.theremin.us/Circuit_Library/four_phase_clock.html

Good luck!

- - - Updated - - -

Whoops! That last link was not the one I intended to share.... Try this instead: http://www.electronics-tutorials.ws/sequential/seq_6.html