TSENS MPC5675K

[wittkop]

We have two date codes of MPC5675K devices.  Date code 1516 returns a temperature which is representative of other temperature sensors on the card and other assemblies in our unit.  Date code 1352 returns a temperature which is roughly 17 degrees higher than the other sensors under the same conditions and running the same software.  Was something changed to the devices or calibration constants between the two date codes?

[wittkop]

We have multiple devices of both date codes.  Here are the constants from a sample of each date.

Date Code 1352

P1 0xA99

P2 0x5ED

C1 0x5D6

C2 0x894

Date Code 1516

P1 0xAFB

P2 0x645

C1 0x59E

C2 0x863

[davidtosenovjan]

Can you share with me the full device part numbers of the devices, please? Based on this we can found which changes was done there.

Are you using equations from latest reference manual? I was told there were some changes in the past in production as well as in equations in RM. The latest device is more accurate. 

[wittkop]

The full part number is SPC5675KFF0MMS2R.

The equations we are using are from reference manual "MPC5675K Microcontroller Reference Manual, Rev 10".

Thank you,

Tom

[davidtosenovjan]

Just for sure - both device has the same part number, they only differ in date code. Is it so?

[wittkop]

Have you received any information about changes to the TSENS?

[davidtosenovjan]

I would need from your side all numbers and letters printed on both packages.

Without precise device identification I will not be able to find out anything.

[wittkop]

Have you received any information about the TSENS issue?

[davidtosenovjan]

I have just been informed new method is applied since date code 1451.

[davidtosenovjan]

Hi, finally I have got some information -

Devices are using different method how calibration constants P1/P2/C1/C2 are received in factory.

SPC5675KFMMS2 0N72D CTJR1516A is using new method.

SPC5675KFMMS2 0N72D CTKN1352A is using old method.

I will not publish more details here, moreover I am waiting for further clarification to exactly know from which datecode the change is being applied. Note that this is a difference in test and calibration software during manufacturing, device silicon is equal.

However the accuracy of the temperature sensor is +-10 degrees so from this point of view the device are in line with the DS because the maximum difference between two devices can be up 20 degrees.

[NicolasH1]

Hi David,
i am facing the same issue.
We have a device with those info:
SPC5675KFVMS2 0N72D CTKL1424D,
and another:
SPC5675KFVMS2 0N72D CFZ2108B.
For both we use the formula described in chapter 53.6.2 of reference manual rev 10.

A = PxC2 – P2Cx

B = CxP1 – PxC1

T = -40 + 185*(A/(A+B))
For the 1st one, we have a realistic temperature, not for the 2nd one, which seems to require the "new formula".
Would it be possible to have the formula for the devices whose date codes are higher than 1451?

How is it possible to read the info indicating which formula has to be applied?

Thanks in advance.

[wittkop]

Here are pictures of each device

[wittkop]

I believe so.  I have the question to our purchasing department to be completely sure.

Tom

[davidtosenovjan]

Could you read TSENS calibration constants on both devices and send it to us? Thanks