I'm using the EP53A8HQA (buck from 5V to 3.3V) on a module board which is connected to a carrier board. The voltage to my module is applied after the voltage on carrier board and the following phenomenon accures:
The voltage to my MCU (and it's IO's on my modul) is applied after the voltage to the IO's on the carrier board, as a result the IO's on my MCU "see" the 3.3V voltage from the carrier board IO's before they get it from 3.3V Rail on my board.
In that time (about 200uS) there is a 2.1V created on the Output of 3.3V Rail on my module. As a result of that voltage, there is a negative inrush current of -6Amp from 3.3V rail via an internal Coil and N-Channel FET to GND of the regulator (for about 200uS).
Can this be a problem?
I would like to send you some pics, but I don't have an e-mail. If you can provide me an e-mail I'll send some pics.
I will send you an personal email, so you can share any confidential information through the email.
BTW, this case consider a pre-bias condition and its really serious where when the EP53A8's Enable asserted high the NMOS will turns on to discharge the output capacitors and will cause the pre-bias voltage to drop to 0V if it is a weak pre-bias.
in that case a very high discharging current will pass through NMOS and could damage it.
So to reduce the risk, you need to have a FB(ferrite bead) between EN53A8 and the load, and also reducing the total capacitance connected on the EN53A8 output till the load (this to reduce the total charge on the output side, so when the NMOS start to discharge the caps this will generate lower discharging current)
Thanks for the reply.
Please send me your e-mail so I can add some pictures.
I already implemented the solution you proposed - I put 10uF Coil in serias after the Power supply and inrush current was reduced to -700mA (down from -6Amp).
Can this be a solution? In other words, Is -700mA is accaptable?
What is the worse case negative inrush the power supply can tolorate?
So sorry for late response, unfortunately i was checking how i can support your account where Intel policy prevent me to get direct contact with MAG accounts through Forum.
I will forward your case to Intel TS in your region to give a direct support to your case.
Back to EP53A8 issue: Since the bead reduced the discharging current significantly to less than 1A, so this is a safe current level that EP53A8H can handle and you can proceed with that solution.