Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Half-bridge versus Full-bridge for Power Dense supplies?

Status
Not open for further replies.

idijoeteque11

Newbie level 4
Joined
Jun 20, 2019
Messages
7
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
93
Hello,

I have seen a few posts on here regarding whether the full bridge or half bridge circuit topology is best within a SMPS. However, not in regards to power density.

I am designing a power supply that requires a high step up (of 1:20) and must achieve the highest possible power density. Although FB is typically used in this domain, I have also recently seen some topologies for high step-up that use the HB, leading me to get thinking a bit more as to why.

My question is, if one was wishing to attain the highest power density, which would they choose as the main inverter bridge - the FB or the HB? One would assume the FB, because the HB presents half the input voltage to the transformer and thus it has to step up the voltage by an even larger margin. However, the addition of two extra switches means the requirement for extra gate drives, and larger heat sinks. Furthermore, if one was to use a resonant topology, there are smarter things that can be done with the half bridge to get high performance than with the full bridge, such as clamping etc.

Full bridge is usually used with high power, higher input current supplies - however mine is moderate power <1kW, and input current is not significant (<5A).

Therefore, although the obvious choice for me originally was the full bridge due to the step up aspect - I am now leaning towards possibly using the half bridge instead. Also oftentimes, when I see the HB, it seems to be switched at a much higher frequency than the FB - leading to reduced passives sizes etc.

I suppose the drawback of using the HB (lower output voltage) could be mitigated by boosting in a pre-converter, for e.g.

Your opinions on this matter are greatly appreciated! Thanks. :popcorn:
 

highest possible power density.
the highest possible power density ever ... ?

What is your max Vout ? - this will limit the power density

1/2 bridge is fine for resonant designs at 1kW, you cannot use current mode easily for 1/2 bridge, current mode for full bridge is very useful
 

the highest possible power density ever ... ?

What is your max Vout ? - this will limit the power density

1/2 bridge is fine for resonant designs at 1kW, you cannot use current mode easily for 1/2 bridge, current mode for full bridge is very useful

Hi! Maximum output is 5kV, at 150mA.

I didn’t know that about current mode control of half bridge, that it was different to full. But yes, obviously not the maximum power density ever but the highest possible for my application with that criteria placed as the most important. I thought maybe with less switches (I.e. a half bridge) the solution can be smaller. The voltage on resonant capacitor can Be limited in a half bridge also whereas in full bridge it cannot, so half bridge may have more inherent safety benefits than the FB.
 

transformer design will be critical here - better to rectify 5 x 1kV outputs and series, LCC 1/2 bridge would be my choice ...
 
transformer design will be critical here - better to rectify 5 x 1kV outputs and series, LCC 1/2 bridge would be my choice ...

Hello,

Thank you for your suggestion. Do you mind me asking though, why would you choose a half bridge inverter? The transformers turn ratio will have to double, because the half bridge chops the input voltage to half its initial value, whereas in a full bridge it is chopped but remains at the input supply voltage, therefore transformer turns ratio doesn't need to be as large. Also, current stress of MOSFET in half bridge is more - bad for a resonant design, no?
 

The transformers turn ratio will have to double
Turns ratio isn't critical, V per turn and absolute number of secondary turns will be the same.

Regarding transistor rating, your statement in post #1 also applies to resonant half bridge
Full bridge is usually used with high power, higher input current supplies - however mine is moderate power <1kW, and input current is not significant (<5A).
 
Turns ratio isn't critical, V per turn and absolute number of secondary turns will be the same.

Regarding transistor rating, your statement in post #1 also applies to resonant half bridge

Okay, roger that. However, if the turns ratio increases in the half-bridge, how does the number of secondary turns stay the same? Surely it should have to increase so that the extra voltage is induced in the secondary? Otherwise double the number of primary turns needs to be present, but then that infers the turns ratio hasn't actually increased after all.

FvM, would your personal recommendation be HB or FB for this application? And why? If you don't mind me asking.
 

you can't easily have current mode control with HB - so if you are seeking a tight dynamic loop - FB may be the better option.
 

you can't easily have current mode control with HB - so if you are seeking a tight dynamic loop - FB may be the better option.

Wait, I thought that current mode control (hybrid hysteresis control) was only possible in half bridge topologies? And not applicable in full bridges? Which would suggest if controllability is the biggest issue this would be chosen.

There is also a possibility of paralleling two half bridges and having two transformers connected in series at the output.
 

I thought that current mode control (hybrid hysteresis control) was only possible in half bridge topologies?
nope ... tisn't the rail splitting caps diverge quickly if you try to do current mode control and don't compensate for this ...
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top