MSP430G2553: How to confirm if the fuse is blown for failure analysis

Fuse on G2xx family can be only blown if VPP (6-7V) is applied to device pin during SBW / JTAG fuse blow shifting sequence. FET should now if device fuse is blown, because during fuse checking SBW / JTAG sequence (inside get device) device will respond with 05555h value. If I remember right, Katie Pier answered few weeks ago, here on e2e, how this (if fuse is blown or not) can be checked by measuring current on TEST pin, without FET.

Hi Wayne,

I'm not sure how to check the physical fuse by X-ray analysis, but there is a method for checking whether the fuse has been blown as explained in SLAU320. SLAA685 also further explains attributes of the physical fuse.

Regards,
Ryan

Ryan,

Did you mean Sec 1.3.1.2 in SLAU320 shows how to check if the fuse is blown?
www.ti.com/.../slau320ab.pdf

I am still confused.
Would you please explain in more detail how this works for G2553?
What should we do to test if the fuse is blown?

Also, it says "Under certain circumstances (for example, plugging in a battery), a toggling of TMS may accidentally occur while TDI is logical low. In that case, no current flows through the security fuse, but the internal logic remembers that a fuse check was performed. Thus, the fuse is mistakenly recognized as programmed (that is, blown)."
Does that mean under certain circumstances, it is possible for the fuse to be mistakely recognized as blown on G2553 which does not have CPUX?
That is, even if the factory operators do not accidentally blow the fuse, but the device is viewed as fuse blown and could not be programmed again??

Is there any physical way to check if the fuse is blown, like using X-RAY, C-SAM, De-Cap?


Because this is production issue, the customer requests us to make it very clear.

Wayne

Wayne,

All programming software tools, such as CCS or FET-Pro430, automatically checks the fuse as part of the JTAG entry sequence and will therefore notify you if the JTAG cannot connect due to a blown fuse. What is the specific error message you are receiving on your system? Your analysis of the fuse check programming error is correct and therefore makes it very important that the conditions are not met. I will check with the Quality Team to determine how an x-ray/de-cap can verify the fuse status.

Regards,
Ryan

Ryan,

Now, we suspect our defect devices are with fuse blown because CCS cannot connect to them, but the customer wants to verify. This is why we are asking on this post.

"Under certain circumstances (for example, plugging in a battery), a toggling of TMS may accidentally occur while TDI is logical low. In that case, no current flows through the security fuse, but the internal logic remembers that a fuse check was performed. Thus, the fuse is mistakenly recognized as programmed (that is, blown)."

Here, the certain circumstances mean unstable power supply?
Even if the fuse is mistakely recognized as programmed (blown), is this permanent or temporary?
When we power cycle it with a stable power supply, it should work correctly again so that JTAG can be connected.

Is my understanding correct?


Wayne

The TI guys can correct me if I am wrong, but "fuse" is a relic from the old PROM days when you actually burned out metal traces to program the chip. Now it is all EEPROM and almost impossible to see. You *might* be able to use a voltage contrast SEM to see it, but that is not my area of expertise. You would have to know exactly which EEPROM cell you need to check, and I would think that TI would consider that proprietary.