LTC1261 controlled by SHDN discharge problem

[yefj]

Hello , I am trying to implement the response of DC-DC circuit shown in the attached link.
Its a pure simulation functionality issue.
I put a pulse into SHDN which in real life will be GPIO periphery of MCU.
As you can see when I put 3.3V the DC-DC is not responding.
When SHDN=0 its start working and after some time settles to -4V.

The problem starts when I put 3.3V back to SHDN ,the output just stays as it was before.
As you can see below I have tried to put MOSFTER to discharge quickly the output.
The discharge is working but I get -0.8V instead of -4V and i get a muc bigger ripple.
Where did i go wrong?
Thanks.

https://www.analog.com/en/products/ltc1261.html?ADICID=SYND_WW_P682800_PF-octopart
https://www.qorvo.com/products/d/da007774

 

[FvM]

Several issues.
- SHDN stops SC converter but doesn't discharge output. Without load current, output voltage doesn't decay.
- Your discharge transistor has wrong polarity. Need PFET or NFET with flipped D/S and a level shifter to drive it.

 

[KlausST]

Hi,

.. additionally
* I don´t think it´s a good idea to do a discharge of a capacitor with no current limit
* use a simulation tool like it´s meant to be: use current measurements to investigate current flow path(s)
* a look into the MOSFET datasheet (first page) shows the diode...

Klaus

 

[yefj]

Hello Klaus,what mosfet do you recommend me to use for this purpose ?
What precaution do you recommend to limit the current going threw the mosfet ?
Thanks .

 

[KlausST]

What precaution do you recommend to limit the current going threw the mosfet ?

The most basic electronics device. Hint: It has a constant relation between voltage and current.

what mosfet do you recommend me to use for this purpose ?

Why a MOSFET at all? Why no relay, optocoupler, BJT...

But OK, let´s look for a MOSFET.
There are only two different types of MOSFET: N-CH and P-CH. Both have the internal body diodes.
Let´s try the N-Ch one: The only possible way (for S, D) is D to GND and S to Negative_Supp(else the diode becomes conductive)
Then you need to drive the gate HIGHer than Source to make it ON. Higher than Negative_Supp.

This can´t be done with your SHDN signal connected to G, because it would make the MOSFET ON as soon as the supply goes negative.

****
Let´s try the P-Ch:
Here you need to connect S to GND and D to Negative_Supp.
Now you need to make the gate more Negative with respect to S. Negative with respect to GND.

This also does not work with your SHDN signal ... it never becomes negative.

****
None of the single MOSFET does work.
That´s why FvM recommended: "and a level shifter to drive it."

Klaus

 

[FvM]

Designing a discharge circuit without timing requirements is pointless.

 

[yefj]

found an error

--- Updated Apr 14, 2024 ---

Hello, Thank you for the good responces. There is no time requirements.
The device could be turned off even after 10sec.
I have chose a PMOS with Vt=-0.45 from datasheet shown below
I want the PMOS conducting using the formulas below:
Vs=0V Vd=-4V
conducting condition:
Vsg>0.45 --> 0-Vg>+0.45 Vg<-0.45

As you said i need level shifter to make the PMOS conducting.
I learned previosly the level shifting could be done using zener diode using its reverse voltage value.
How do I implement similar voltage level shifting in my situation?
Thanks.

https://www.vishay.com/docs/71683/si4403dy.pdf