TMP116: FPC Implementation

Zeming Kong

Part Number: TMP116

Hi team,

My customer is evaluating TMP116 and plans to implement it in FPC. Epoxy adhesive is used to fix the temp sense FPC on detection object.

There are 2 types of use cases.

a) Epoxy adhesive will cover all the package in 60C temp. condition when implementing, will this damage or erode the package? And operating temp will up to 80C, will this cause overheat problem?

b) We are concerned about the impact on long-term reliability due to distortion or deformation caused by stress when the adhesive adheres to the package or pin. Can this be used for more than 2 years lifetime?

c) Anything else needed attention in these use cases? or do you have any similar application? Could you provide recommended design?

Best,

Zeming

over 3 years ago

0 Ren Schackmann over 3 years ago

TI__Mastermind 35915 points

Hi Zeming,

a. The TMP116 is rated for 150C operating temperature. There's no issue with 60C or 80C.

b. We've observed encapsulation used in our packages causes an initial shift in temperature. This error is an offset that will not change over time.

c. Is the 'Detection Object' electrically conductive? Any electrical contact with the vias for the thermal pad must be at GND potential. Design 1 has better thermal conductivity to the sensor than Design 2. This will give a faster thermal response, but should not affect final accuracy.

thanks,

ren

0 Zeming Kong over 3 years ago in reply to Ren Schackmann

TI__Expert 3515 points

Hi Ren,

Thanks a lot for your reply.

a. Understood that no thermal problem here.

b. So you mean if we package this IC using epoxy adhesive, temp shift will happen but no erosion any physical damage?

c. Yes, it is a metallic case. I will suggest the customer to ground this.

And is there no similar recommended design we can refer to?

Additionally, could you tell me the material and coefficient of linear expansion of the TMP116 package?
Also, I would appreciate it if you could teach us about the material of the pins including Thermal Pad.

Best,

Zeming

0 Ren Schackmann over 3 years ago in reply to Zeming Kong

TI__Mastermind 35915 points

b. The temperature shift is caused by mechanical stress of the epoxy. I don't think I can comment on erosion; I believe this would be determined by customer process and contamination.

No, we don't have an published material for this type of design.

The mold compound is a variant of G700, which is a common semiconductor mold compound.

This page shows finish plating is NIPDAU: www.ti.com/.../search

thanks,

ren

0 Zeming Kong over 3 years ago in reply to Ren Schackmann

TI__Expert 3515 points

Hi Ren,

One more question, is there anything we need to pay attention when we use I2C to read the temp?

Can we consider that we can read the result at any timing?

Best,

Zeming

0 Ren Schackmann over 3 years ago in reply to Zeming Kong

TI__Mastermind 35915 points

Hi Zeming,

There isn't any special timing consideration for this application. If you intend to have a long wire run to the sensor, you might plan to run at a lower frequency or stronger pull-up due to the parasitic capacitance of the wires. The datasheet shows support for 1Khz to 400KHz, and 400KHz is already a pretty modest speed.

thanks,

ren

0 Zeming Kong over 3 years ago in reply to Ren Schackmann

TI__Expert 3515 points

Hi Ren,

Thanks for your support. The customer is now evaluating the reliability of TMP116.

Could you provide reliability test results (temperature cycle, bending test, etc.) of the solder joint between the temperature sensor IC and the mounting board?

Best,

Zeming

+1 Ren Schackmann over 3 years ago in reply to Zeming Kong

TI__Mastermind 35915 points

Hi Zeming,

All quality/reliability information is available from the TMP116 product page. Here is a direct link to TMP116 quality information:

https://www.ti.com/quality-reliability-packaging-download/report?opn=TMP116AIDRVR

thanks,

ren

Sensors

Sensors forum

TMP116: FPC Implementation