Temperature characterization - die v package

Thread Starter

scMicropt

Joined Apr 20, 2017
11
Just curious - are wafer die performance characteristics stable over temp and is it the package encapsulant that limits the performance characteristics over temperature? I have some encapsulated QFN parts rated to -40 C to +85C and want to characterize them at 100 C. Heat is not a factor in the part i want to take to temperature but it could be a factor for some parts. I can't get a straight answer from the application engineers. One suggested that the parts would survive but they had not characterized the part at 100 C. Fair enough. I will do so.But I am curious if there are other limiting factors.

The part in question is a DAC. It is rated for -40 to +85 and I need it to run at 100 C. Its thermal consumption is 1mW at full speed.

Any input appreciated.
 

dl324

Joined Mar 30, 2015
16,839
It sounds like you need to use the military version. Better suggestions with better info; like part number for starters.

Characteristics change with temperature. The datasheet will typically give parameters for low, room, and high temp.

What matters is the junction temperature. The datasheet will give the thermal resistance from junction to case and then you control case to ambient.

EDIT: Information for LMx58:
upload_2018-2-1_8-10-30.png
If it isn't included in the datasheet, you need to contact the manufacturer.
 
Last edited:

Thread Starter

scMicropt

Joined Apr 20, 2017
11
It sounds like you need to use the military version. Better suggestions with better info; like part number for starters.

Characteristics change with temperature. The datasheet will typically give parameters for low, room, and high temp.

What matters is the junction temperature. The datasheet will give the thermal resistance from junction to case and then you control case to ambient.

EDIT: Information for LMx58:
View attachment 144916

The part is a single DAC LTC2641-12. 12 bit DAC 50 MHz SPI bus. I can't get the die so I was wondering how it will perform at 100C. There is no military version of a DAC that will run at 50 MHz on a SPI bus.

Thanks
 

crutschow

Joined Mar 14, 2008
34,280
Understand that testing one unit at 100°C will not guarantee that other untested units will operate the same way.
Why do you need 100°C?
 

dl324

Joined Mar 30, 2015
16,839
The part is a single DAC LTC2641-12. 12 bit DAC 50 MHz SPI bus. I can't get the die so I was wondering how it will perform at 100C. There is no military version of a DAC that will run at 50 MHz on a SPI bus.
You need to calculate maximum power dissipation and contact LTC to get thermal resistance information.
EDIT: It's given on page 2 with the package outlines.

You can also inquire about parts specified for a wider temperature range (from page 3).
 
Top