Inoncsistent magnetic sensor sensitivities

Thread Starter

kellogs

Joined Aug 13, 2017
7
Hi,

I am using this sensor with half success rate

Parts used around the chip:

Soldering procedure:
  • bake the sensors for 9h @ 125 deg C
  • apply leaded solder paste through stencil, hand place the chips
  • bake in oven according to JEDEC J-STD-020.

Testing:
  • wait for all devices to cool down
  • hover the same magnet over each PCB placed in the same testing spot

Result:
  • some of the PCBs will light up their led sooner than others. That is, i need to bring the magnet closer to some of the boards in order for them to light up their led, which effectively makes some sensor chips being more sensitive than others

How do I mend the situation for maximizing sensitivity for all boards ? I have tried rearranging a tray of PCBs after first bake and then baked them again - no noticeable differences sensitivity-wise.

Something else - the datasheet specifies at page 35:

>The following recommendations apply to capacitor C1:
>It is highly recommended to use low ESR (max 200 mOhm)

It does not specify a frequency (range) of interest. The 220 nF cap I am using only has that ESR from around 3.5 MHz onwards. Could this be related to my issue ?
 

jpanhalt

Joined Jan 18, 2008
11,087
The datasheet only mentions "Pin 1 Indicator" once, and that is the quote. Pin 1 is SCL/SPC, which is a clock pin. So what's "indicator?"
 

Thread Starter

kellogs

Joined Aug 13, 2017
7
Oh, right, I have missed that line in your answer.

Well, from LSM303C's datasheet, a predecessor of the chip I am using:

>>3. The "Pin 1 Indicator" is identified on top and/or bottom surfaces of the package.

I have never seen anything else other than the 12 pins on the bottom of these chips; they must have had it on the bottom at some point in time.
 

jpanhalt

Joined Jan 18, 2008
11,087
All chips have some sort of indicator for Pin 1, but it is not an electrical connection. It is usually a dimple or other marking (e.g., bevel) on the epoxy.
 
Top