SM320F2812-EP: Coefficient of Thermal Expansion (CTE) for V62/05601-03ZE (TI P/N SM320F2812-EP)

Hello Nico,

One of our experts will get back to you by Tuesday, since Monday is a US holiday.

Hello Matthew, 

Thank you for looking into this, I had a couple follow up questions:

1.) Is the plating thin enough that it's CTE contribution can be ignored compared to the copper? 

2.) Is the copper lead frame the primary expansion mechanism, or is it the epoxy encapsulation material or combination of both (where I would need to calculate an equivalent CTE)? If it is a combination, what is the equivalent x-y CTE for the entire case, not just the copper? x-y in this case is expansion in the plane of the part, from lead on one side to lead on opposite side.

Thank you,

Nico

Nico,

I'm going to need to pull in someone from our packaging side of the house to dive a bit deeper.  It may take a few days to get the right folks looped in, so I appreciate your patience here.

Best,
Matthew

Nico,

To make sure, you are using this orderable PN, SM320F2812PGFMEP?  I had failed to notice there was a BGA version of this device, SM320F2812GHHMEP, but that is a Tin/Lead construction.

Please confirm this is for the PGF(TQFP) version of the device and not the BGA version.

Best,

Matthew

Hello Matthew, 

Looking at the TI datasheet for the mil spec p/n V62/05601-03ZE it says the package type is LQFP PLASTIC QUAD FLATPACK and I believe the correct part number should be SM320F2812 PGFMEP. 

Thank you,

Nico

Thanks Nico, I've relayed this back to our process team.

Best,
Matthew

Nico,

Sorry for the length of time between responses on my side.  

The package CTE is not something that we have available/calculated for all devices.  From what I have gathered this is fairly complex set of simulations to account for both the material composition as well as the geometry.  I'm going to try and get a solid answer if the lead frame CTE would dominate the overall calculations for your purposes.

Best,
Matthew

Hello Matthew, 

I am just checking in. Were you able to get an answer from your team? 

Thank you again for your help. 

Nico

Nico,

I appreciate your patience here, it appears there does exist a modeling method to derive the package CTE, but I'm still understanding what exactly the parameters we need to supply into it to get out the CTE.  

I'm going to set up a call with the modeling team to understand the best path forward and/or the typical path forward when a customer requests this info.

As I mentioned this is not a typical request that I am familiar with at the device level.  I know that doesn't help you, but just trying to be as upfront as possible.

I will update tomm with the plan going forward.

If you could indulge me a bit as someone who is well versed in the electrical aspects of this device, but not so much the mechanicals; is the long term risk of an unknown CTE that the device may become unmounted over time due to the different thermal co-efficients of both the device, solder, and PCB we are dealing with? 

Is there a particular mission profile that requires the need for this info, vs applications that do not(considering that we don't provide this publicly).  

I'd like to drive some better efficiencies in this process moving forward, I think the answers to the above will be of help to do that.

Best,

Matthew

Hello Matthew, 

Thank you for getting back to me and I appreciate your help with this unusual request. Aerospace companies are increasingly asked to meet a temperature cycle life requirement for both military and commercial customers. For example, we might be asked for our electronics assembly to withstand 2000 cycles from -20 to 80C. That is not an actual mission profile, just an example. 

For each temperature cycle, CTE mismatch between the PWB, component and solder (to a lesser extent) cause plastic strains in the solder joint which accumulates fatigue damage over time. After a sufficient number of cycles the joint (usually corner lead joint) will break altogether causing a failure. There are empirical equations developed by the industry to predict the solder joint temperature cycle fatigue life, but they require the component in-plane CTE (one corner to the other) as an input for the model. 

There are nobs to turn to increase the cycle life for a design, such as selection of the PWB CTE that results in highest fatigue life for a problem part but to make those sorts of design decisions it is helpful to know what the component CTE is. 

Thank you for your help.  

Nico