Capacitor based backup for brief loss of power in memory stick

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Anonymous
Not applicable
        I am using the CY7C68033 USB Controller Chip to create a simple USB memory stick (1-NAND chip). It has an additional function that it provides a One-Time-Password per power cycle. I want to stop a brief turning off of the power to the USB port to cause a power cycle. I am not an Engineer and I simply used the reference schematic using the TP378833 voltage regulator. I am “hoping” that I can simply add a RC circuit to the device to provide a brief (~3 second) guard against this. I was hoping someone could tell me first if this is feasible and second, do I put it before the regulator or after and will it effect power consumption? I am thinking of using a 300 uf Cap but don’t know what resistor would go with that. The memory stick is very thin as it looks like a House Key so batteries etc are not an option. … Or am I being paranoid 🙂 ? Thanks.   
0 Likes
4 Replies
Anonymous
Not applicable

 Hi,

   

 

   

Your description about the power cycle is not clear. Please give more details about this.

   

 

   

Do you want to power cycle the board without disconnecting it  from the host?

   

 

   

Regards,

   

Vikas

0 Likes
Anonymous
Not applicable
        I am trying to STOP a hacker from powering down the USB port on a PC briefly to cause my device to power down and when powered up recalculate a new One-Time-Password. If they do this they could harvest the One-Time-Passwords from the device. If I can increase the time the port needs to be without power before the device sees this as a true loss of power, I can reduce this threat. (I am assuming I can have a capacitor power the device briefly).   
0 Likes
Anonymous
Not applicable

 Hi ,

   

 

   

I don;t understand your implementation of one time password.How you are implementing this one time password?

   

 

   

Is it similar to the one in the link http://www.cypress.com/?docID=33264

   

 

   

If you just want to retain the power even if the upstream VBUS is disconnected by hacker /or some you cannot do it. Beacuse If you place resistor and capacitor then it wil get discharged based on its time constant (~R*C). If you just want to retain for specific amount of time then use appropriate resistor based on it time constant equation

   

 

   

FYI:For NX2LP it needs minimum 5ms of time constant. place this at the output of regulator.

   

 

   

Regards,

   

Vikas.

0 Likes
Anonymous
Not applicable

 Vikas,

   

         I am using a Patent Pending reprogramming of the Cypress Controller Chip to act as a USB Flash Drive where there are 20 "special" files stored on the FAT32 Drive. 10 Can be used to store HMAC-OTP counters,Identifiers and encrypted data (Code File). While there are 10 corresponding files when read return the Identifier, OTP Hash and encrypted data (PasswordFile). The CodeFile returns NULLs when read, the Password file returns NULLs the second and subsiquent times it is read. The rest of the drive can be used to store user's files.

   

       Obviously the flag that indicates that the Password file should return NULLs is lost after a power cycle of the Controller. So if a hacker had a program to turn power off to the USB port even briefly the next OTP Hash would be available from the Password File. I want to keep the controller powered during a brief power cycle of the USB port that a Hacker might do to harvest OTP-Hashes. Obviously I would like this to last as long as possible, but size constraints limit the capacitor (1-1.5mm height).

   

      Also I have no idea if the capacitor is placed on the USB side, will it discharge back into the PC, if placed after the regulator will it discharge back through the regulator. Do I need a diode to stop this. Will the charge of the RC circuit increase the load on the USB port too much. What is the minimum R value to avoid this. These are all the questions, my untrained mind has come up, I am sure there are others as I said in the first post, I am not an engineer but a computer programmer.

0 Likes