Post Go back to editing

Is it possible to use ADALM1000 to discharge Nominal 1.2V Ni-MH battery?

Category: Hardware
Product Number: ADALM1000

Hi,

Recently, I attempted to use the ADALM1000 (M1K) to discharge Ni-MH batteries and obtain their discharge curves.

The experiment setup was very simple: I connected the battery’s positive terminal (Bat+) to CHA and the negative terminal (Bat-) to GND, without any limiting resistor in between.

However, I encountered an issue where the discharge process seems to stop when the battery voltage drops to approximately 1.23V. The battery I used was a Panasonic Eneloop AA size Ni-MH battery with a nominal capacity of 1900mAh. As you can see in the attached data, when I was using a -200mA discharge current to sink the battery, the datalogger logged samples once per second. The discharge curve ran for about 18 hours (65485 seconds), which is obviously much longer than the expected duration based on the nominal capacity.

My goal is to fully discharge the Ni-MH batteries to understand their complete discharge characteristics. I understand that the current sink loop circuit should not be able to sink the voltage to near 0V, as there seems to be a load resistor of about 4 ohms, which would drop about 0.8V at -200mA.

I wonder if there is any way to achieve this goal? There are hardly any similar applications available. I only found one example, “Active Electronic Loads,” but it seems not set up for low voltage applications.

Any suggestions would be much appreciated.

Thanks,

Calvin

Parents
  • I have a few questions about how you conducted your experiment. Can we assume you had configured the M1k SMU to source current / measure voltage (SIMV) mode?

    You did not seem to also data log the measured current. So we don't know if indeed the SMU was sinking the requested (-200mA) current. If we have both the voltage and current data we would know the internal resistance of the cell as it was discharging.

    Can you redo the experiment data logging both the cell voltage and the current flowing out of the cell (current being sunk by SMU)?

    Another approach to doing this experiment would be to use CHA in SVMI mode set to 0 V with a small series resistor (say 5 to 10 ohms) across the cell so as discharge it to GND through the resistor. Then use CHB in Hi-Z mode connected to the cell to measure the voltage of the cell vs time. The CHA current measurement vs would be the cell discharge current vs time.

    If  you are writing your own control / capture program script in Python you could even monitor the CHA discharge current and adjust the CHA voltage (each measurement cycle) to maintain a more or less constant discharge rate.

    Also a user has posted some battery measurement scripts in this GitHub Repo:

    https://github.com/StefanMack/M1K/tree/master/BattMeas

  • Hi, 

    I finally obtained a complete discharge curve of a Ni-MH battery using the method you shared. Please see the attached data below. The experiment was conducted by setting CHA to SVMI at 0V to drain current via an external resistor of about 4.7 ohms, while CHB monitored the voltage variation of the Ni-MH battery. The battery was fully discharged from about 1.33V to below 1V in approximately 9.4 hours (33800 seconds). The results closely match the nominal capacity, so I think this experiment should be valid.

    However, I also tested a few times by directly connecting the Ni-MH battery to CHA and GND in SIMV mode, with various currents of -50mA, -100mA, -150mA, and -200mA. The servo loop worked well for currents under -100mA, but above -150mA, AD8018 didn’t seem to drain enough current (measured by the voltage drop of R46, R47), which may have caused my initial experiment to fail. I’m not sure if my M1K is broken anyway.

    Regarding your point that “The M1K cannot supply the full 200mA of current near 0V,” I thought the AD8018, with a negative supply of -1V, should be able to drain the voltage to near 0V. If this is not possible, what might be the reason?

    Datalog: docs.google.com/.../edit

Reply
  • Hi, 

    I finally obtained a complete discharge curve of a Ni-MH battery using the method you shared. Please see the attached data below. The experiment was conducted by setting CHA to SVMI at 0V to drain current via an external resistor of about 4.7 ohms, while CHB monitored the voltage variation of the Ni-MH battery. The battery was fully discharged from about 1.33V to below 1V in approximately 9.4 hours (33800 seconds). The results closely match the nominal capacity, so I think this experiment should be valid.

    However, I also tested a few times by directly connecting the Ni-MH battery to CHA and GND in SIMV mode, with various currents of -50mA, -100mA, -150mA, and -200mA. The servo loop worked well for currents under -100mA, but above -150mA, AD8018 didn’t seem to drain enough current (measured by the voltage drop of R46, R47), which may have caused my initial experiment to fail. I’m not sure if my M1K is broken anyway.

    Regarding your point that “The M1K cannot supply the full 200mA of current near 0V,” I thought the AD8018, with a negative supply of -1V, should be able to drain the voltage to near 0V. If this is not possible, what might be the reason?

    Datalog: docs.google.com/.../edit

Children