LMZ14202H: Not working 30% failure

Hi Emanuel, 

I am sorry to hear that. I am not aware of any issue with this part recently. 

You may be able to contact either the distributor where you got the parts from or a local TI Field Application Engineer that may be able to assist you. 

The part is a family of 1A to 3A module which is pin to pin compatible

LMZ14203H/02H/01H

Another part that is not pin to pin compatible is 

https://www.ti.com/product/LMZ34202

Thanks

-Arief

The parts were purchased directly from TI.

I don't exactly understand what an application engineer will be able to help with. I explained the problem clearly.

An unchanged process, no problems until now, this specific 500pcs reel so far with 30% failure.

Sorry but I also don't exactly understand your second part of the answer - we are using all 3 versions in different products depending on the power consumption.

The problem is only with the recently purchased 2A version.

What all this has to do with the LMZ32202 I don't understand

Hi Emanuel, 

What i meant is that the field application engineer may be able to assist you to fill up a form to do Failure Analysis (FA) on the part. 

On the issue that you encounter, what was the issue? Is it the part does not startup or is there a damage to the pins?

We can also review your schematic to see if there is anything that potentially contributing to your issue. I understand that the application have been implemented for years but it's also good to double check.

Thanks

-Arief

Dear Arief,

Please allow me to explain again with more details.

The problem is indeed no output or no startup and it happens at the initial test after the reflow. No physical damage.

The reflow profile was checked and we use leaded temperatures which are lower than the lead free.

Today we assembled the same product that uses the 1A LMZ14201H, 100% of the boards passed the test with no problems.

We make the same product with the 3A too, never had any issues, in fact as mentioned never before we experienced such problems with the 2A.

It is clearly not a design problem.

I can add also that our engineers tried to replace 3 damaged parts using hot air, 2 out of 3 didn't survived the process, the third worked, showing that the only problem is with the LMZ.

I read in another thread that at some point TI made  a structural change in this part.

https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/486868/lmz14202h-failures-different-substrates?tisearch=e2e-sitesearch&keymatch=LMZ14202H#

It seems that it became very sensitive to heat.

I asked in my first message if there is a "replacement option", the intention wasn't to replace the part, but if there is a policy of returning parts left in the reel and receive others form a different production lot.

There is one point to mention, on the batch assembled where we had the damaged parts we changed the output capacitors from tantalum to ceramic, same value.  

Hi Emanuel, 

The fastest way is to enter a return request on the customer return portal.  Please see the instructions below for the FA return process.  
 
For products purchased directly from TI or the TI store the customer follows this return request procedure:

• Request analysis of suspected nonconforming product using the return request submission form.  Be sure to include picture of the top of the unit (with TI symbol) in the submission form attachment
• Upon TI acceptance of the request, return instructions will be provided. Please do not ship products until instructed with a CPR\QEM number.

If we can get this part still mounted on the PCB for mechanical removal that would be best.

Secondary reflow may cause additional damage internal for these modules.

Thank you and i hope that helps

-Arief

Hi Arief,

It is interesting why the post I refereed you to, where another user discovered a structural change in this component was quickly locked and the discussion was taken to direct contact.

We assembled yesterday a different board with the same part, exactly the same failure percentage.

I am expecting a more serious attitude to such a complaint from a reputable company such as TI.

This is not a cheap part and repairing damaged boards is very costly, our damage is huge, not to mention the reliability concerns about the parts that apparently are working.

I am very disappointed, perhaps after all these years I will open the door to Wurth Electronics that offer a similar part.   

Hi Emanuel, 

Most of the time, the thread or post that is old may get automatically locked by the system.

I apologize if we do not provide a satisfactory customer support for you.

I also understand that it may not be a design problem since you have used this application for a while, but i would be more than happy to check your schematic or layout if you like. 

One thing that may assist you in troubleshooting is to do A to B swap between the module and the working PCB and see if the failure follows the module.

Otherwise, if you suspected that the part may have some issue. You may be able to return the part for Failure Analysis based on the link above. 

Thank you

-Arief

Hi Arief,

I opened a FA request, so far no answer.

To summarized the events:

We assembled the LMZ14202H on both boards shown below and had the same failure rate.

On the white board we have versions that use the 1A and 3A too, no problems with them, only with the 2A.

No changes were made on the reflow profile which was tested with a profiler, same conditions were used for the 1A and 3A LMZ.

Earlier this year we purchased another reel of the LMZ14202H and it was assembled with 0 failure on the same board.

The white board is aluminum substrate double sided and the blue is regular FR4 double sided too.

The reels are stored in a dryer cabinet at <10% humidity.

As mentioned several times already, we use the LMZ family for more than 8 years, more thousands assembled and never had any issues before.

I think that all the logics point to a bad batch or worse, a change in the product's manufacturing method or structure, and I am unable to find any notice about this.  

Hi Emanuel,

Yes, i will check again with our internal quality team to see if there is anything change on the substrate. 

Once you have submit the part to FA, we may be able to analyze it more. 

By the way, you may have done this already but on the suspected 2A part that have an issue, is it possible to reflow it into the white board or vice versa (part from white board into the blue board). This just to make sure that the fault follows the part. 

Thanks

-Arief

Arief,

We removed from several boards the part, not an easy task specially on the aluminum board, replaced with new parts and everything is ok.

Except for 1 out of 8 boards, after all the failure rate is valid also for repairs.

By the way, the parts look intact, no signs of a humidity problem like the popcorn effect.

Hi Emanuel, 

Thank you for your clarification. I presume in this case the next thing to do is to submit the part for failure analysis. 

Thanks

-Arief

It seems that nobody is in a rush at the failure analysis dept.
Although I marked the case a "critical" nothing happens.

Aren't they supposed to send me return instructions?

This is really very bad customer support, I have no words 

Hi Emanuel,

I am sorry for the time delay.

The QA department may also have been back logged with the Failure Analysis queue and in the process of catching up with the request that they had during pandemic while we were advised to work from home. 

We have also already raised the concerns to them in regards to this issue. 

Thanks

-Arief