Hello,
I'm evaluating the LIS2DE12 accelerometer by using the combination of STEVAL-MKI175V1 and STEVAL-MKI109V3 along a Power Profiler Kit II from Nordic Semiconductors for current measurement.
Method of test:
I removed the jumper that source VDD, bridged the Power Profiler Kit II (2 red circles) and connected GND (yellow circle)
Placed the STEVAL-MKI175V1 and connected it to the computer.
I proceeded to open UNICO-GUI and configured it to run at 200 Hz ODR with a range of 2g. Also enabled bit I1_ZYXDA from CONFIG_REG3 (22h) as I'm planning to catch the data available coming from INT1 in my application.
The datasheet describes that the expected consumption is 18 uA at 200 Hz ODR.
But the result leaded to a consumption of 41.36uA with an average consumption of 42.2uA. This is more than the double expected consumption.
Am I missing something to configure in order to achieve the expected consumption?
Hi @sch0bert ,
As explained in the STEVAL-MKI109V3 user manual, jumpers J13 (VDD) and J14 (VDDIO) allow the user to measure the sensor current consumption by connecting a multimeter in series with their terminals, please try again to measure the current consumption following this guideline.
I did a test using the STEVAL-MKI175V1 and STEVAL-MKI109V3, without removing any jumpers, and I measured the current consumption with the command *get_i (pag 17 of the User manual), I obtained exactly 18uA:
It called my attention that you were getting "2526 uA" in IddIO so I decided to repeat the commands you shared by having both jumpers (J13 and J14) and I got 25.47 uA. So could you point me which is correct? Is it 2.5 mA or 25 uA? I'm using FW version 3.60.7
Also, why is not specified in the datasheet the consumption values of IddIO? In that case, by having I1_ZYXDA set, you get an extra 25uA/25mA (to be confirmed) so the total consumptions is not what someone would expect from that it says in the datasheet.
I reviewed again the manual and indeed I did the test correctly as I'm placing the meter in serial to Vdd. By reviewing the schematic, I can see both power sources are completely independent from each other so there is no conflict if I measure just one at a time.
I proceeded to test again having the power meter in the configuration from the previous image but this time with the commands and this were my findings:
On the first scenario, the current meter in the STEVAL-MKI109V3 says 33.85 uA but in reality with the Power Profiler Kit II is 17 uA on the average. Seems like the development kit is measuring just the peaks and reporting that.
The second scenario, where I enter into power-down mode, the STEVAL-MKI109V3 reports 17.14 uA but on the Power Profiler I get 0.31 uA and IddIO remains consuming 25.30 uA. Isn't supposed to turn off also? Vdd_IO pin is not manipulated, is supplied directly from the eval board.
The third scenario, I decided to clear I1_ZYXDA bit and that brought down the IddIO consumption to 0.1615 uA
And forth scenario, I came back to 200 Hz ODR but with I1_ZYXDA disabled. The values from the Power Profiler were the same, an average consumption of 17 uA and a small increase in the IddIO consumption which I don't know is expected (not documented).
Reviewing again what I reported by using UNICO-GUI, I noticed something that I didn't catch it on the first time. In the image below, I start in power-down mode and then change it to 200 Hz ODR. As you can appreciate, it starts with a correct current consumption and then jumps into what I reported initially. Why is this happening?
Unfortunately there is no option to check the current consumption in UNICO-GUI for LIS2DE12 as I have seen for the LIS2DUX12.
In order to discard issues with the power supply of the STEVAL-MKI109V3, I decided to use the power supply mode in the Power Profiler Kit II at 2.5V and source directly Vdd. I controlled it using UNICO-GUI and I got the same results as using the STEVAL-MKI109V3 power supply and the Power Profile placed in serial measuring the current.
I'm planning to connect to the same power supply both Vdd and Vdd_IO with the recommended capacitors, is this actually possible?
Regards,
Pascal.
