choose the most appropriate dc dc

[franticEB]

Hi,
i want to design a dc-dc converter that is able to convert the unregulated 48V input (40Vmin-60Vmax) in a programmable ISOLATED output voltage.

So 48V_in
--> 24V_out@1A
-->48V_out@1A
-->144V_out@1A

So i thought to use a single output Flyback converter and to set the right output voltage by modifying the error amplifier's voltage reference with a DAC.

My questions are:
Is the flyback the most suitable choice?
Have you another kind of smps converter to suggest me?

 

[goldsmith]

Is the flyback the most suitable choice?
Have you another kind of smps converter to suggest me?

Hi franticEB
Of course it is not !
in flyback there are a lot of disadvantages specially for your aim .

Forward or half bridge are much better in this task .

Best Wishes
Goldsmith

 

[franticEB]

What about push-pull?

Does the forward converter suffer the same disadvantages of flyback?

Could you suggest me any PWM controller from witch to start the design?

Thanks

 

[goldsmith]

What about push-pull?

Does the forward converter suffer the same disadvantages of flyback?

Could you suggest me any PWM controller from witch to start the design?

Thanks

Hi again
Push pull is much better in compare with flyback but still deals with some disadvantages .
Forward or half bridge converter both are fine for your aim as i have said this before .
Yes of course i can suggest you a proper PWM driver for start . go through UC3845 .

Best Wishes
Goldsmith

 

[treez]

Flyback is ok, though you will get more switching loss and rcd clamp loss. If you can tightly couple pri and sec then your losses of these will be much less.
Flyback ok because your I(out ) is low

Half bridge can be problematic...read the LM5039 datasheet to see this problem.
two tran forward....bad because you need hi side drive....which is a nuisance as it cannot be bootstrap.

 

[goldsmith]

Flyback is ok, though you will get more switching loss and rcd clamp loss. If you can tightly couple pri and sec then your losses of these will be much less.
Flyback ok because your I(out ) is low

Half bridge can be problematic...read the LM5039 datasheet to see this problem.
two tran forward....bad because you need hi side drive....which is a nuisance as it cannot be bootstrap.

I disagree ! either halfbridge or forward converter could be fine if designer knows how to design !


Best Luck
Goldsmith

 

[treez]

Thanks, but the description of the problem that is given about the half bridge in the LM5039 datasheet is very true. It must be accounted for in any half bridge design.
2 transistor forward...yes fine, go ahead and use it if you can sort your hi side drive. But poster must remember that gate drive txfmrs are highly susceptible to leakage inductance.
I think for the sake of simplicity , id just go with the flyback....not the most efficient solution in the world, but what is the volume of these psu's?....what is the usage pattern?...maybe they only get used for a few minutes per day.....is great efficiency really sought after, etc etc .

 

[goldsmith]

Thanks, but the description of the problem that is given about the half bridge in the LM5039 datasheet is very true. It must be accounted for in any half bridge design.
2 transistor forward...yes fine, go ahead and use it if you can sort your hi side drive. But poster must remember that gate drive txfmrs are highly susceptible to leakage inductance.
I think for the sake of simplicity , id just go with the flyback....not the most efficient solution in the world, but what is the volume of these psu's?....what is the usage pattern?...maybe they only get used for a few minutes per day.....is great efficiency really sought after, etc etc .

Hi Again !
Take it easy my friend ! you talk about forward converter as i guess you consider it as a rebellious horse ! ha ha
Great efficiency of what ? flyback ? who told that it deals with great efficiency ? indeed less than forward or half bridge !

Best Luck
Goldsmith

 

[treez]

https://www.ti.com/lit/ds/symlink/lm5039.pdf

..this tells of the half bridge woes of capacitor voltage imbalance.

I tell you about two transistor forward converter that uses no gate drive transformers, but uses two high side bootstrap gate driver ic's.
it is set up like a full bridge, but only two diagonal transistors conduct a power stroke......the opposite high side fet is missing , and just a diode is there instead, and the bottom side other fet is just for bootstrap capacitor refresh.

...so there you have it, two transistor forward converter with bootstrap high side fet drive.....I bet many will have thought that that was impossible.