STEVAL-ISV020V1

[Khoi Ly]

Hi ST Community,

I wonder if there is a way to know if the battery is being charged by the energy harvester. I connected the solar panel to the harvesting circuit and have the li-po battery on the output side. I don't see the battery being charged.

Based on the instruction, I need to supply 2.6V of voltage initially to the CN4 channel, but it does not work either.

Please let me know what I can do.

Khoi

[Peter BENSCH]

Welcome, @Khoi Ly​, to the community!

You probably misunderstood: the 2.6V are not to be applied to CN4, but to the PV input CN1, CN4 remains free. It is also the minimum voltage from which the SPV1050 works, so the solar cell must supply min 2.6V. What voltage does it generate in your case?

For the function of CN3, CN4 and the other connections to CN2 it is highly recommended to work through AN4397, the Getting Started with the SPV1050, as well as the data sheet of the SPV1050.

Regards

/Peter

[Khoi Ly]

Hi Peter,

Thank you. The circuit claims to harvester voltage as low as 150 mV, so I am curious how to make it work with my solar cell at 1.1V to charge my li-po battery. Do I need to modify the circuit somehow?

Thanks,

Khoi

[Peter BENSCH]

Well, that is indeed somewhat misleading and perhaps not presented in sufficient detail. It is important to know that with silicon-based devices it is not easy to work with voltages below the forward voltage of pn junctions (~0.65V). Accordingly, the data sheet of the SPV1050 also talks about two voltage ranges:

1. Input Voltage Range
2. Minimum input voltage at startup.

Let's start with the second range Minimum input voltage at startup, as that is important for switching on (start up) the converter. Here, a distinction is also made between boost and buck-boost configuration, which each have different startup voltages: in boost mode it is 0.55V (typ.), in buck-boost mode 2.6V (typ.) as mentioned before. At these voltages, the component can start working and generate an output voltage.

Now to the first range, Input Voltage Range:

When the SPV1050 then operates and supplies itself, the input voltage can drop down to 0.15V before the SPV1050 stops working.

Finally, please note that there is a second board STEVAL-ISV019V1 that is designed for boost mode. Unfortunately, when the Data Brief for the STEVAL-ISV020V1 was written, it was pointed out too generally that the board could be switched to 0.15V.

Yes, this is possible, but ultimately only if the board is put into boost mode. In effect, this turns the STEVAL-ISV020V1 into a STEVAL-ISV019V1:

• L1.1 to CN1.1 instead of U1.19.
• U1.19 and U1.20 to GND

Unless I missed something, these should be the only changes.

Is that any clearer now?

Regards

/Peter

[Khoi Ly]

Thanks Peter, It's much clearer now.

So in any cases (STEVAL - ISV019 or STEVAL - ISV020), I will need to supply to the input at start up ( 0.55V or 2.6V for boost and buck-boost config respectively) before I can lower the input voltage to 0.15V.

After the initial start-up time, the input voltage can go down to 0.15V for the energy harvesting to operate. Do you have a suggestion as to how I can use the same Li-po battery's output to step-down and feed the 2.6V back to the input to run the energy harvester?

Thanks,

Khoi

[Peter BENSCH]

We'll check it again internally and then come back here.

Regards

/Peter

[Peter BENSCH]

In fact, for such an application, where the voltage of the solar cell is not sufficient to start the SPV1050, a start-up aid would have to be provided. In principle, it could look something like this, whereby you would have to generate control signals 1 and 2 accordingly:

1. Enable the supply path from the PV panel (and disable the supply path from the battery) when sufficient voltage is available from the PV panel. 
2. Activation of the supply path from the battery (and deactivation of the supply path from the PV panel) if the PV panel does not supply enough voltage.

The SPV1050 provides two signals with open-drain output:

• /BATT_CONN reports whether the pass transistor between the STORE and BATT pins of the SPV1050 is
  • closed (/BATT_CONN=0=active) or
  • open (/BATT_CONN=1=inactive). 
• /BATT_CHG reports whether the DC-DC of the SPV1050 is
  • working and the battery is being charged (/BATT_CHG = 0 = active) or
  • not switching (/BATT_CHG = 1 = inactive)

If you like, you can test whether the two signals /BATT_CONN and /BATT_CHG can be used to switch the schematically shown logic level p-channel MOSFETs using the control signal.

Alternatively, you would have to build a voltage monitor (ultra low power, of course) that:

• is supplied by the battery without the solar panel being active,
• applies the battery voltage to the input as soon as the solar panel supplies a little voltage and
• disconnects the battery voltage again as soon as /BATT_CHG becomes active (i.e. 0)

Hope that helps?

Regards

/Peter

[Khoi Ly]

Hi Peter,

I was able to use the battery to feedback the start-up voltage to the input and it seems to work properly.

I have a separate question regarding battery voltage sampling during charging.

During charging, the harvester module will charge the battery for 16s and disconnect for 400ms to sample the voltage value. I wonder how I can extract/measure the battery voltage during this 400 ms period.

I would like to plot and collect data of the state of charge for this battery.

Thank you,

Khoi

[Peter BENSCH]

Well, what is stopping you from measuring at the battery during the time tSAMPLE?

But it seems to me that the original question has been answered. Since it is now another question that has to do with your project but not with the original question of this thread, I suggest opening a new thread for it. In this thread you can then also better present the framework conditions, e.g. your constructed circuit incl. the points at which you want to measure, etc.

Regards

/Peter