Problem with the ADP1613

Hello Nasib,

in all basic Boost-Converters Vin path is not isolated from Vout, even when EN/Shutdown applied.
Current flows over Inductor and Diode to Cout, thus Vout will be ~ Vin. 
Again this is valid for all known basic boost converters, a physical fact of the structure.

To prevent this you need to make an external switch. A single MOSFET is known for this, but you have still Vin<--> Vout connection in certain cases over MOS body diode (like for PMOS when Vout >Vin during Vin turned off. Thus two MOSFETs back-to-back a better solution. But more components you need then.   
Also logical signal driving these two MOS must be arranged accordingly.
To make this design problem easier, we have parts like LTC3122 or 3123. Where such switching is integrated.
It is called output disconnect.

Or you can also use a Buck-Boost converter, in your case surely it will be only in Boost mode (5Vin, 15Vout). 
They have 4-switches making output disconnect also from these. LTC3111 or 3114-1 are very popular parts. 

I hope these can help you,

Hi my main concer in not the 5V at Vout when EN=0. My main concern is that if i connect a 30ohm resistor in the output. the system start to limit and the output drops to 10V while the current rises up to a short circuit...

Probably you hit the overcurrent/short-circuit  protection,  as your switch current will be around this  that value.

With Vout=15 @ 0.5A and Vin=5V and about 90% Efficiency assumption (Fig.1), you have about 1.2 * 15V*0.5A / (5V*0.9) ~ 2Ain. 
Figure 16 shows overcurrent threshold is about this value for 15Vout. 

You can test first with e.g. 40ohm. If it functions you can decrease Rload to 30Ohm and observe if the voltage drops again to e.g. 10V.
If yes, that is a sign overcurrent protection jumps in.

Possible solution would be using a higher switch current part; like ADP1614 or LT8333.

Hi. Thanks so much for your response. 

I will make a couple of experiments today and get back to you here again. 

Also I experimented another interesting behaviour.

If I connect a lipo battery charger (that provides a constant current of 170mA when I connect it to an independent power source, because of the current status of the battery). The current consumption of the system rises to 700mA more or less (and 15V).... I can not understand this...

Thanks! 

Battery charger connected to  15Vout, right?
Such chargers have their own input capacitors (MLCC and/or Electrolytic), which will be additional Cout for boost converters.
If this helps Vout getting back to 15V, I can think of help of such capacitors.

Also meaning your 10uF Cout was not enough. Is that an X7R cap?
Do not forget MLCCs lose significant capacitance under high bias voltages.
Also across temperature, lifetime and similar effects.

You can add  some uF parallel to 10uF, when you test ADP1613 back alone. 

Battery charger connected to  15Vout, right?
Such chargers have their own input capacitors (MLCC and/or Electrolytic), which will be additional Cout for boost converters.
If this helps Vout getting back to 15V, I can think of help of such capacitors.

Also meaning your 10uF Cout was not enough. Is that an X7R cap?
Do not forget MLCCs lose significant capacitance under high bias voltages.
Also across temperature, lifetime and similar effects.

You can add  some uF parallel to 10uF, when you test ADP1613 back alone.