UCD90120A: Wrong UCD internal temperature value

Hi Lorenzo,

We will look into this and reply to you soon.

Regards,

Weidong

Hello

The internal temperature is to measure the die temperature of the chip.

How do you program the device? Have you followed the steps mentioned https://e2e.ti.com/support/power-management/f/196/t/514452?UCD90120A-internal-temperature-sensor-accuracy to ensure that the calibration data is good?

Regards

Yihe

Hi,

We have followed the steps mentioned in such link, but we encountered some problems.

Can I send you UCD90120A_Trim_Data.zip and UCD90120A_Trim_Data_2.zip?

The first one was obtained isolating UCD device from other devices thanks to a switch implemented in our board. As reported in the Report.html file, Device was not found in this mode. So I followed the same procedures connecting UCD device to the other ones, obtaining UCD90120A_Trim_Data_2.zip. 

Device seems to be entered in ROM mode now and our board cannot be powered up. I tried to follow the previous procedure, but device cannot be found when UCD is isolated from all devices. When UCD is connected to other devices, instead, program shows these messages:

As you can see, UCD9XXX Device GUI found ROM UCD30xx, but we are implementing UCD90120A. How can I solve such problem?

Thanks,

Lorenzo G. Toscano

Hello

Please attached two zip files in the post so that I can review.

The device was stuck at the ROM mode. please refer the attached document to recover and make sure the checksum is correct before regenerating4834.recover UCD90120A from ROM mode.pdf

Regards

Yihe

Hi,

Ok, I attach such zip files.

I followed the steps described in the "4834.recover UCD90120A from ROM mode.pdf" file and now device works as before. Thanks.

UCD90120A_Trim_Data.zip

UCD90120A_Trim_Data_2.zip

Looking forward to hearing from you,

Lorenzo G. Toscano

Hi,

I repeated the steps to get trim data, getting the following zip file:

UCD90120A_Trim_Data_correct.zip

Looking forward to hearing from you,

Lorenzo G. Toscano

Hi

Thank you for the sharing.

The trim data looks good. Could you also share your project file and schematics?

Please be noted the internal temp sensor is to probe die temperature.

Regards

Yihe

Hi,

Hopefully trim data looks good. The problem is that internal temp sensor is measuring a value much lower than expected. In fact, when using our board inside our climatic chamber at -20 degree C, internal temp sensor measures around -26 degree C, but it should be at least equal to -20 degree C.

Which could be the problem?

Looking forward to hearing from you,

Lorenzo G. Toscano

Hi

I had checked with the team and the data sheet is not updated to reflect the internal temperature-sense accuracy:

The internal temperature-sense accuracy is characterized by design and not product test. The range of the sensor is between -10C/10C. Again,these values are not production tested.

We will update the data sheet to address this.

BTW,  what do you plan to do with this internal temperature?

Regards

Yihe

Hi,

Thank you very much for your feedback.

We would like to use this internal temperature to control switch on/off of our boards, in order to avoid eventually possible board damages in unpredictable and undesirable cases.

I wanted to ask you politely if the range of sensor -10C/+10 C is an offset or accuracy problem.

Looking forward to hearing from you,

Lorenzo G. Toscano

Hello

If you have other on board temp sensors with much better accuracy, is that a better optional to do this than the internal temp sensor.

UCD90120 can take external signal to control the power sequencing. Will this help?

The +/-15 accuracy is based on the characterized by design instead of production test.

Regards

Yihe