Hi,

I have a microcontroller interfacing with an SD card. I want to be able to remove the power supply from the SD card when the device goes into "sleep" mode. I use the following schematic to connect/disconnet the sd card power supply:


When pulling the base to GND the sd card turns on. But when i try to turn the transistor off, the collector only drops to 2.8V and never disconnects the SD card from the power supply(VCC).

I dont understand what is wrong with this circuit, when removing the SD card and connecting a load to the transistor collector pin the transistor works as it's supposed to do so it looks like the SD card is causing this behavior. Any advice?

Regards,
Sisco

 

The base of the transistor needs to be pulled up to Vcc with a resistor. When there is no current through R141, T23 will be turned off.

 

The base of the transistor needs to be pulled up to Vcc with a resistor. When there is no current through R141, T23 will be turned off.

Hi Papabravo,

Thanks for the quick reply. I should have said it right from the beginning but i already tried using a 10K pull-up resistor between the base and VCC without any success. Also the microcontroller pin and the SD card power supply are both 3.3V, so when the controller is driving the base high no current should be able to flow right?

Sisco

 

What about control / control / data signals to the SD card?

If these portbit is not set low or tristated, there will probably run current "backwards" from the the SD card data to its supply.

 

What about control / control / data signals to the SD card?

If these portbit is not set low or tristated, there will probably run current "backwards" from the the SD card data to its supply.

Where would this "backwards" current come from? The SD card reader has no other power source.

 

Where would this "backwards" current come from? The SD card reader has no other power source.

What I suggest, is that the U-controller sends current "backwards" through the SD card data and control connections ESD protection diodes to SD card supply.

And I see now that there are some pull-up resistors of the same potential current path.

 

Try putting a 470Ω resistor from Vcc to the base of the PNP transistor.

 

What I suggest, is that the U-controller sends current "backwards" through the SD card data and control connections ESD protection diodes to SD card supply.

And I see now that there are some pull-up resistors of the same potential current path.

Sorry, I see what you mean now, I think this is a possibility.

 

If you let the base float open does the SD card turn off?

 

Hi

Try putting a 2.2k resistor from the PNP transistor collector to ground so that the 2.2k resistor is in parallel with C127 and C128.
Like this:

 

Back feeding through the SD card is an issue with the design. All of the control signals must also be 'disconnected'

 

What I suggest, is that the U-controller sends current "backwards" through the SD card data and control connections ESD protection diodes to SD card supply.

And I see now that there are some pull-up resistors of the same potential current path.

This was indeed the problem, setting these pins to low before disconnecting the power allows the card to be reset properly.

Thank you all so much for the help!

 

Should consider using MOSFET because the saturation voltage drop on the bipolar takes most of your headroom.

 

Should consider using MOSFET because the saturation voltage drop on the bipolar takes most of your headroom.

No. I have used a PNP BJT in a same situation powering a 2732 UVEPROM and the voltage drop is 20mV.

 

Should consider using MOSFET because the saturation voltage drop on the bipolar takes most of your headroom.

That's not an issue.
The FMMT717 has a 72mΩ ON resistance and a 17mV max. Vce saturation voltage at 100mA collector current.

 

Learn something new everyday. I was still thinking in terms of the oldi devices with ~.3V Vce.

 

"Try putting a 2.2k resistor from the PNP transistor collector to ground"

+1

maybe also try base to ground with 2.2k.

germ

 

"Try putting a 2.2k resistor from the PNP transistor collector to ground"

+1

maybe also try base to ground with 2.2k.

germ

Are you sure about that, do you miss something?

 

Are you sure about that, do you miss something?

the 2.2k will bleed the current off.

 

the 2.2k will bleed the current off.

You still missed the point -- what you said was for NPN not for PNP ...