HTS221 - difective products?

[AAssi]

Hello i have 45 HTS221, in everyone the temperature is correct but the value of the humidity is wrong, i's usually 4 to 30 point higher than the true value.

I have try everything suggested in this forum or in technical data, like ultrasonic wash, rehydration, internal or external heating.

last year ST replaced the component free for charge because they were defective.

The connection, the footprint and the soldering process is ok for ST too.

I think the HTS221 is a bad product.

but i need an answer, Had anyone managed to make them work properly?

I apologize for my bad english

[S.Ma]

Only way to check would be to use the X-NUCLEO ST MEMs shield and compare in same conditions (HW, environment and SW) to check if the parts are indeed defective (or uncalibrated).

[LFONT]

Where is the "true value" coming from?

Did you check the reference sensor you are using is not uncalibrated itself?

Anyway, like all humidity sensors, also HTS221 needs to be carefully used.

Placement is, for example, very important since heat conducted through the PCB may heavily impact the readings. 

One important evaluation could be to check for close heat dissipating devices, like MCUs, LCD backlight, power supplies (batteries), RF chips (Bluetooth) or the likes.

If nothing is evident, you can contact the distributor who sold you the devices to check if the batch (you need to provide the references to your material) from which your devices come from have been signaled as "non conformal" and request for replacement.

[AAssi]

The true value coming from a calibrated instruments. I take a new tool and the values are the same, i took an old board, with a Sensirion's SHT11 and the difference is only one humidity point.

All the sensors are in the same room, at the same environmental conditions.

So these are the value

Calibrated instruments : T 20.1 H 46.2

STH11: T 19.9 H 45.2

HTS221 on nucleo board: T 19.6 H 61.9

HTS221 on my board: T 19.3 H 50.1

I tested it with my program, approved by ST and with Arduino board, using one of the library example for HTS221.

[LFONT]

​Thanks for the reply.

Can you, please, share the PCB layout with indications of the different components (on both sides)?

Thanks

[AAssi]

This is an image of my PCB. I had follow the ST guidelines for the placement, so the source of heat are the STM32, the switching power supply and U5 transceiver. In my box, the air flow from HTS221 to the switching. The test was performed in open air.

There are nothing in the bottom sides.

[LFONT]

​Hello.

The PCB layout seems correct.

Assuming the soldering temperature profile and re-hydration are correctly executed according to the Section 4.1/HTS221 DS as reported hereafter:

 <<The HLGA package is compliant with the ECOPACK® standard and it is qualified for soldering heat resistance according to JEDEC J-STD-020. After soldering, the accuracy specification of the sensor can be guaranteed after re-hydration of the sensor element in a stabilized environment (25 °C / 55% RH) for 3 days or at 70% RH for 12 h. Otherwise the sensor may read an offset that slowly disappears if exposed to ambient conditions.>>

the potential root cause for the wrong readings could be:

• absorbed soldering process residuals or previous stress tests

Actions:

1.     If possible, run a microscope analysis of the HTS221 package and the sensing area to verify if they are damaged (soldering process residual over and around)

2.     Run a bake for 12h at 125 °C

3.     Run a test against the reference with the recommendation to put it as much as possible close to the sensor to verify if the sensor has recovered its performances

Thanks

[AAssi]

Hello,

for the re-hydration, one of the fault HTS221 was on ambient condition for month. Here, in Parma Italy, we have a lot of humidity, especially in summer.

my PCB assembler use a no-clean solder paste, they run a microscope analysis and everything is correct.

i can try to ask to my PCB assemble to bake the PCB

[LFONT]

​Ok, let's proceed step by step.

First let's do a bake as suggested.

In the meantime, please provide the reflow temperature profile that was used (not the theoretical one, but the real measured profile the devices have experimented).

Thanks