MSP430FR5738: Problem of Low Power Mode(LPM 4.5) and Draw More Current than Expected

Hi,

Can you please provide photos of the markings on the top of the chip? Where were they purchased? I will try to trace them internally.

To confirm that there are no other factors at play can you confirm that there have been no software or hardware changes between the "5AK CNKJ H" (working) and "86 I AV45 J" (not working) systems?

Are you able to perform a device swap between working and non-working systems? Steps:

1. De-solder FR5738 device (call it device A) from working system
2. De-solder FR5738 device (call it device B) from non-working system
3. Solder device A in non-working system and device B in working system
4. Ensure device A and B are running the exact same software.

Does the issue follow the board or the FR5738 device?

Regards,

Evan

Hi Evan,

Thanks for your reply. We can confirm that there have been no software or hardware changes between the systems. We have tried your suggestion as well. We have swapped the MSP's between working and nonworking systems. The result is the working one still works as expected on the other system(same current draw), and the nonworking one(the one that draws too much current) still behaves as the same,too much current draw. We have been told that all MSP's to be produced from this FR5738 series will have the new marking code. The old marking codes that work properly with us are will no longer be produced.

In the attachments below, you can see three photos of the markings on the top of the chip. Chip1 and Chip3 corresponds to the newer marking codes, where the LPM4.5 expected current draw can not be seen. Chip4 corresponds to the old marking code MSP, which we have used it in our projects so far. We can satisfy the LPM4.5 minimum current draw requirements with this card. As i mentioned, we have used exactly the same hardware and software design for those MSP's. If you trace these MSP's and marking codes, that will be very appreciated.

Thanks again for your help and interest.

CHIP 1

CHIP 3

CHIP 4

Hi Berkay,

Thanks for sharing those images. Can you share the origin of these parts? Are they from an authorized TI distributor?

Regards,

Evan

Hi Evan,

The origin of the MSP's are from Maleysia, and they were bought from an authorized TI distributor.

Regards,

Berkay

Hi Berkay,

I'm unable to track those device lot trace codes (LTC) using my normal tools. I'd like you to file an inquiry with the TI anit-counterfeit team. Please report the results.

Let me know if there is anything else I can assist with.

Regards,

Evan

Hi Evan,

You can find the photo of the case of the MSP and the related shipment information in the attachments below. The origin of it is from Philiipines, not Maleysia. In addition to that, one of our colleagues have also opened a case about this topic before. You can find it as well with the case number CS1336774. The MSP's we currently use and subject to our topic are the ones that were sent to us by the TI distributor as a result of the case we specified above. We then experimented our system with those MSP's that were sent and the results are the ones that i explained in this post previously. If the old marking code and the new marking code MSP's are the same, why does our code only give the desired values ​​in the old one, is there a technical explanation or solution other than filling out the anti-counterfeit form? Thanks a lot for your interest .

Regards,

Berkay

Ok understood. Thanks for explaining the history. 

Can you run the LPM4 example (linked below) on both old (working) and new (non-working) devices? What are the results?

I haven't found any documents indicating that there are significant differences between the devices in your possession. Regardless, the LPM4.5 Iq you have reported is not within spec of the current datasheet. 

Regards,

Evan

https://dev.ti.com/tirex/explore/node?devices=MSP430FR5738&devtools=MSP430FR5738&node=A__ADfZdck-rmqTk.VQ5TmrMA__msp430ware__IOGqZri__LATEST

Hi Evan,

We have tried your LPM4 suggestion. On datasheet, we have seen that at 25°C, it should draw 5.9 uA. We have observed that the old marking coded MSP draw 4.5uA,  which is very acceptable,  however, the new marking coded MSP's(CHIP1 and CHIP3 at above pictures) still works quite bad. It draw 180uA on one MSP, and 260uA on the other one. So, i think we have proved that there is nothing we can do on the software or the hardware side. It would be definitely some sort of MSP problem, maybe sourced from production.  We couldn't think or find any other solution for this problem. Is it possible for you to send us both new marking coded and old marking coded MSP samples which is guarenteed to be original and come from authorized TI distributor ? We will do the measurements again and compare the values and let you know the new marking codes would draw the expected datasheet current values or not. Or if you have any other suggestion for this problem, it will be appreciated.

Regards,

Berkay

Hi Berkay,

Thanks for performing that test. It does appear the problem is related to the MSP430. Would you answer the following questions below? I'm actively working internally to determine the next steps. 

1. How are you performing these current measurements? Can you described the test setup?
2. What were the results of the counterfeit assessment? You said they shipped you new parts, but were the devices deemed to be counterfeit or not?
3. How many devices have you found to have this issue? Do all the "new" device have this issue?

Regards,

Evan

Hi Evan,

1. We have used multimeter and power supply in order to measure the current. We did the measurements under normal room temperature conditions. We confirmed all the tests with different devices with different marking coded MSP's. If something wrong with any of the setups or measurements, we never get any correct values. However, we have obtained the required expected values for old marking coded ones, which proves our setups and measurement methods are correct.

In the previous post, i have already mentioned that we have tried your code for both scenarios. The scenarios are;

a. The code loaded on old marking coded MSP system

b. The code loaded on new marking coded same MSP system

Result : --> Old marking coded current draw is appropriate, as in the datasheet, however, the new one draws too much.

2. The MSP's we have used so far comes from Turkey TI distributor(Yunus Karabörek).  We don't think anything comes from him would be counterfeit.

3. We have observed this issue on all new marking coded MSP's. As we have said earlier, not only LPM4.5 Mode, but also even LPM4 mode draws too much current.

How are your measurement values for LPM4 and LPM4.5 modes for new marking coded MSP's ? Do you obtain the same values as in the datasheet ?

In addition, have you found any solution or do you have any other suggestion to solve this problem as soon as possible ? Unfortunately, It affects our ongoing projects schedule as well. If not, Is it possible for you to send us both new marking coded and old marking coded MSP samples which is guarenteed to be original and come from authorized TI distributor ?

Hi Berkay,

Thanks for your detailed response. I have no reason to doubt your test setup or findings, nor do I suspect your parts are counterfeit.

At this point, I think that a return is necessary to figure out what the root cause of the issue is. Please initiate that process through the TI distributor you bought the parts through. 

I hope we can resolve this quickly.

Regards,

Evan