[SOLVED] inductor current exceedes its saturation rating durin buck startup

[Gouwomb]

Hello,
I am designing buck converter using MCP16301
Vin 14V
Vout 5V
Iout 125mA
https://ww1.microchip.com/downloads/en/devicedoc/20005004d.pdf


I have chosen inductor for it:
SRN3015TA-220M
22uH, saturation current 550mA
https://www.bourns.com/docs/Product-Datasheets/SRN3015TA.pdf


I have a concern about exceeding inductor saturation rating during buck converter startup (probably caused by charging of output capacitor).
Application note do not address this condition.
It only suggest to pick an inductor with current rating above peak inductor current in steady state (300mA in my case).

Is there something to worry about?
I guess controller will limit peak current cycle by cycle to about 1.6A until output capacitor is fully charged.

Simulation are from microchip’s MPLAB.

 

[KlausST]

Hi,

a quick view into the datasheet says:

Search for "soft":
The device includes a softstart feature, usually desgined for charging the ouput capacitor slowly, while reducing the input (and inductor) current.

Search for "saturation":
"An inductor saturation rating minimum of 760 mA is recommended. "

Klaus

 

[Gouwomb]

Hello Klaus,
thank you for reply,

I might be wrong,
but soft start is seen on the plot as increasing peak current limit during first couple of cycles.
So it looks like to me as soft start is too "short".


"An inductor saturation rating minimum of 760 mA is recommended. "

Yes it is true, but sentence above applies for applications using full 600mA output current and I need only 125mA.
Also datasheet recommends 20uF output capacitor.
For plots in first post I am using 10uF.

plots 20uF output Cap


As you can see inductor current exceeds even suggested 760mA saturation current.



It is high volume low cost project, so smaller inductor I choose is better.

 

[KlausST]

Hi,

I might be wrong,
but soft start is seen on the plot as increasing peak current limit during first couple of cycles.
So it looks like to me as soft start is too "short".

Yes, the time is short. But usually one wants fast startup.
You can´t have all in one: fast startup, low inrush current, high ouptut capacitance....

First decide your
* start up time and your
* inrush current limit
Then decide if the IC can fulfill your requirement.

Yes it is true, but sentence above applies for applications using full 600mA output current and I need only 125mA.
Also datasheet recommends 20uF output capacitor.
For plots in first post I am using 10uF.

edit: Sorry I misinterpreted the your sentece....
(I can't find where you see that there is only 125mA saturation current needed for 600mA output current. This is impossible in my eyes.)
Now I know what you mean.

The start up current is almost independent of load current...because most of the energy is needed to charge the ouptut capacitor.
After the output capacitor is fully charged then the ripple current and peak current depends on load current.

You can´t ignore physics.
To charge the ouptut capacitor in a short time it needs high inductor current.
* The higher the capacitor value, the higher the current
* The shorter the time, the higher the current.

Klaus

 

[FvM]

So it looks like to me as soft start is too "short".

Yes. If you worry about the inrush current, you need a switcher IC with programmable soft start or current limit.

 

[asdf44]

On a really quick look this inductor has a pretty soft saturation profile that may be ok in a high inrush scenario above its 'saturation' rating (inductance doesn't drop steeply above its saturation rating).

Soft start is a better solution but many 'soft' saturation inductors can handle transients and surges.

 

[Akanimo]

Unfortunately, linear inductor models cannot tell whether an inductor is saturated. You need to model the core properties to be able to tell whether the inductor is saturated or not.

 

[asdf44]

Though saturation is a simple function of the instantaneous current. So the simulation will do a decent job of showing you whether current approaches or exceeds that value. But make sure to put in a derated value for the inductor (-30%) since that's the value the inductor will have at its saturation current.

 

[treez]

You can actually put an RC across your "RTOP" to make it start up more gradually aswell.
Also, i doubt it may not matter anyway...as those inductors are gapped, and so you will not get full inductor saturation.
Having said that, can you not find a control chip with a lower peak current rating?

If you have a fet current sense resistor , you can increase that to keep startup current low...but your controller doesnt offer you
the chance to do this...you are stuck with the peak current of that controller.
Find another controller is my advice....one with a lower peak current rating.