topic Re: FRDM-K66 back drive protection in Kinetis Microcontrollers

FRDM-K66 back drive protection

rinipatel — Mon, 03 Oct 2016 08:52:16 GMT

Hello,

Why there is no back drive protection (Schottky diode) for P3V3_VREG in FRDM-K66F?

When board power supply (in my case P5V_SDA_PSW) is OFF, and external GPIOs are attached to Arduino headers (J1-3, J1-4, J2-GND), I'm getting negative current through jumper pins J20 and J18. And after this when board is power cycled, it goes directly in bootloader mode.

Could this be because of the absence of Schottky across J19? What could be the get around for this issue?

Thanks.

Re: FRDM-K66 back drive protection

jorge_a_vazquez — Tue, 04 Oct 2016 21:24:55 GMT

Hi Rini Patel

What voltage level (logic level) do you have in J1-3, J1-4?

I tried to reproduce your issue, I connected J2-GND and J1-3, J1-4 to 3.3 V, with the board off. Then I connected SDA MICRO USB with j20 shorted and j18 open. But I could not see your issue, my board power on correctly (without entry in BOOTLOADER mode).

Could you try with the jumper configuration that I mentioned to see if the problem still is happened?

Regards

Jorge Alcala

Re: FRDM-K66 back drive protection

rinipatel — Wed, 05 Oct 2016 05:48:28 GMT

Hello Jorge Alcala,

Thanks for the reply.

I can confirm your observation. By providing 3.3V to J1-3 J1-2, J1-4 and connecting J2-GND, board doesn't enter the bootloader mode on applying power supply.

But in my application, I've configured J1-2 and J1-4 as UART1_RX and UART1_TX respectively, and connected these pins with USB-to-UART FTDI breakout board's TxD, RxD and GND. (3.3 V logic). Now the TxD pin of FTDI chip is providing some current to J1-2 (few mA, which is 0mA if I directly apply 3.3 V Vcc) . So when board is powered on, there is already some back current flowing through J20.

With the similar connection on FRDM-K64 board, the back current was not there, which in my guess is because of the presence of Schottky across J19 (?).

What can be done to counteract the effect of reverse current and prevent the board from entering bootloader mode?

Thanks,

Rini

Re: FRDM-K66 back drive protection

jorge_a_vazquez — Fri, 07 Oct 2016 15:55:56 GMT

Hi Rini Patel

I tried some other test in the pins that you mentioned, and I don't see your issue yet, but I have checked both Schematics (K64 and K66) and I noticed this absence of Schottky diode in J19. So I will ask to design team, why this component was omitted and I will inform you any update as soon as possible.

Regards

Jorge Alcala

Re: FRDM-K66 back drive protection

rinipatel — Fri, 14 Oct 2016 03:45:22 GMT

Hello Jorge Alcala,

Any findings on your side?

Re: FRDM-K66 back drive protection

jorge_a_vazquez — Fri, 14 Oct 2016 21:41:53 GMT

Hi Rini Patel

No, I don't get an explain of the absence of this diode in the schematic of FRDM-K66, But I have noticed something.

First time you mentioned that the headers were connected in J1-3 and J1-4, witch are PTE11 and PTC4. If you check the default configuration for these pins, they are disable, witch mean high impedance, so I didn't understand why you have a leakage current. But then you mentioned that are J1-2 and J1-4, witch are PTC3 and PTC4, and in this case there is a problem, because PTC3 has as default configuration CMP1_IN1 witch is a analog pad and could be a leakage current here. This current could be the negative current through jumper pins J20 and J18 that you see.

Could you made a test without connecting PTC3, only connecting PTC4, so we can determined that this is the problem? if this is the problem, you could use a current limiting resistor in this pin, or you could put the diode that is missing. It 's not a good practice have power in pins of the MCU while this isn't property powered, so I would recommend you not to do this. If the MCU is not property powered, it could damage flash memory.

Hope this information helps, please tell me if this is the problem that you have.

Best Regards

Jorge Alcala

Re: FRDM-K66 back drive protection

rinipatel — Tue, 18 Oct 2016 05:22:41 GMT

Hello,

I made a test without connecting PTC3(UART1_RX) which is connected to TxD pin of my USB to GPIO breakout board and did not encounter the issue. But I think it has more to do with TxD pin of breakout supplying some current than with what the default state of pin is, because this I've tested with PTC3/PTC4 (J1-2/J1-4), PTE24/PTE25 (J2-11/J2-13), PTC14/PTC15 (J199-3/J199-4), PTD2/PTD3(J2-6/J2-8), and PTD6/PTD7(J6-6/J6-7). Same issue for all these pins.

Following is the exact sequence of my test:

1. Provide power supply to board.

2. Attach RxD, Rxd and GND (or only TxD) of USB to UART breakout with board UART pins.

3. Disconnect USB power supply. Getting -ve current through J20.

4. Reattach USB power supply, UART pins are still connected, board goes into bootloader mode.

The same test with FRDM-K64 works fine. No negative current through J20.

I'll try to place the diode that is missing, but upon close inspection it is shorted, so will it make any difference?

Thank you.

Rini

Re: FRDM-K66 back drive protection

jorge_a_vazquez — Thu, 10 Nov 2016 19:59:46 GMT

Hi Rini Patel

I haven't had luck in finding information about this issue, still they told me that I shouldn't try to made tests like this, because it could damage the MCU. So as I mentioned, there is no recommendable to operate the MCU with this procedure, your MCU could get damaged by getting power when VDD has no power, so I recommend you no to do it.

Sorry if this could cause you problems

Regards

Jorge Alcala

Re: FRDM-K66 back drive protection

rinipatel — Fri, 11 Nov 2016 03:48:13 GMT

Hello Jorge Alcala,

Thank you for looking into this. I'll see if I could find an alternate test procedure to avoid damaging MCU.

-Rini