[SOLVED] Difference between switching regulator with external and internal power mosfet?

[chicken_feet]

Hi all,
Would appreciate if anyone could provide some answers to the questions below -
1. There are switching boost/buck power regulator with internal switching mostfet and external ones. When and how should we decide if we should chose those build in or those external mosfet? My newbie thoughts as such -
Regulator with internal mosfet has limited output current capability, generates more heat, smaller overall PCB area compared with those where the switching mosfet are external. Are my thoughts on this valid?

2. If I have multiple power rails ( let say 6 power rails). Would it be better to have a multi channel regulator or get individual regulator for each power rail? What are the advantages and disadvantage of using a single regulator with multiple channels?

Many thanks.
Best rgds,
KaKiaYam

 

[malli_1729]

My views are for Buck converters/controllers
Yes, your views are correct,

1. Internal/External MOSFET switching regulators
Decision depends on Max current handling capacity,Area of the converter size, layout complexity,reliability concerns, type of application (aerospace,telecommunications,medical etc)
5 years ago, when we start design regulators for currents > 10A, we go for Controllers (External MOSFET), for lesser currents with we go with Converters(Internal MOSFET).
But, now i see Converters will handle currents up to & in excess of 40A.
Check your application requirements, whether requirement for higher current is continuous/transient also.

Layout complexity & Risk is more for Controllers, at the same time they are more flexible for good engineers. layout complexity & area is less for converters compared to controllers.
I see controllers are more reliable for long life & robust environments than converters.
Controllers are not good for people who doesn't have much experience in power component layout, as these will have extra loops to optimize.

2.Multi channel/Individual regulator
Space,current demand, over all power dissipation will decide the selection.

 

[dick_freebird]

BOM cost and board area are often the key design-tos.

An internal FET is going to be inferior electrically and
input-cost-wise (being produced in an over-elaborate
flow relative to its function), but the value of integration
will likely make -price- superior, and potentially enable
functionality that isn't so easy to get in a PCB / discrete
form.

You can pick and choose FETs if left outside, even mix
root technologies (like silicon PMIC, GaN switches).

 

[chicken_feet]

My views are for Buck converters/controllers
Yes, your views are correct,

1. Internal/External MOSFET switching regulators
Decision depends on Max current handling capacity,Area of the converter size, layout complexity,reliability concerns, type of application (aerospace,telecommunications,medical etc)
5 years ago, when we start design regulators for currents > 10A, we go for Controllers (External MOSFET), for lesser currents with we go with Converters(Internal MOSFET).
But, now i see Converters will handle currents up to & in excess of 40A.
Check your application requirements, whether requirement for higher current is continuous/transient also.

Layout complexity & Risk is more for Controllers, at the same time they are more flexible for good engineers. layout complexity & area is less for converters compared to controllers.
I see controllers are more reliable for long life & robust environments than converters.
Controllers are not good for people who doesn't have much experience in power component layout, as these will have extra loops to optimize.

2.Multi channel/Individual regulator
Space,current demand, over all power dissipation will decide the selection.

The line between controllers and converters are increasingly blurred now which is why it prompted the questions above. Would converters offers less flexibility in terms of improving the stability of the power supply over a larger range of current/load requirements? We understand that many switching power supply has stability issues under low load with high ripple.

As for the multi channel over individual regulator selection, I would assume individual regulator would offer better thermal but compromise on layout space.

Many thanks for your advise. Have a nice day!

KaKiaYam

 

[malli_1729]

The line between controllers and converters are increasingly blurred now which is why it prompted the questions above. Would converters offers less flexibility in terms of improving the stability of the power supply over a larger range of current/load requirements? We understand that many switching power supply has stability issues under low load with high ripple.

As for the multi channel over individual regulator selection, I would assume individual regulator would offer better thermal but compromise on layout space.

Many thanks for your advise. Have a nice day!

KaKiaYam

As per my understanding, Stability is more of Mode control used in converter/controller & the efficient layout,Loop closures. Both should work in the same way if layout is good.
If stability is concern take flow controls like D-CAP3,Current mode control, etc, make sure all critical loops are very small area and properly closed.

 

[treez]

internal/external fet doesnt have any bearing on stability...as long as you know the modulator transfer function then you're ok.

Regarding multiple power rails...it all depends on the allowed voltage tolerance, and the power of each rail.
Multiple power rails are a bit messy.....especially if when the regulated rail is on max load, and some of the others are simultaneously on no load.