This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ27Z561-R1: Load Select

Part Number: BQ27Z561-R1
Other Parts Discussed in Thread: GPCRB

Dear expert,

Our default "Load Select, Load Mode " value is 1,1? = "Present average discharge power"?

So "Avg P Last Run" doesn't take effect?

How long duration for "Present average discharge power"?

It's very strange "Avg P Last Run" still take effect in my system even "Load Select, Load Mode " value is 1,1.

If "Avg P Last Run" is very big , like -2000mA / -760cW, FCC will drop to 0 when temperature decrease to -5C in my system. (RBteak is already done.)

After I limit  "Avg P Last Run" within -1000mA / -380cW, FCC and discharge are all OK at even -10C.

Why my experiment doesn't match "Present average discharge power"?

Great thanks

  • And why we recommend "load select" default "Present average discharge power" instead of "Avg P Last Run"? Is there any benefit for it?

  • The gauge will use Avg P Last Run for a capacity simulation/prediction at the start of discharge, where it doesn't know yet what the Present Average Discharge Power is.

    The Present Average Discharge Power is calculated during a discharge and it will take effect when the gauge triggers another simulation (e.g. at specific DOD points or if temperature changes significantly) so this is delayed, hence you will see a change in behavior at the beginning of a discharge with different values of Avg P Last Run even if you select a load prediction of Present Average Discharge Power.

    The reason why capacity can drop to 0 at -5deg.C with a large Avg P Last Run value is that the gauge will use this at the start of discharge to predict capacity and if temperature is low, cell resistance is high and a high load can cause the predicted voltage to drop below Terminate Voltage quickly, resulting in very little or no capacity.

    It's possible that the real cell heats up more than the gauge predicts with a 2000mA load at -5deg.C therefore it's important that the thermal model (temp a,k) is correct. If this is not learned or configured correctly (e.g. in the golden image as a starting point), then the gauge will not adjust cell resistance like the real cell would as it heats up and the real cell may be perfectly capable discharging significant capacity while the gauge predicts that it can't.

    That's why it's important that temp a and k are derived from a test setup where the cell is in the final enclosure without air forcing cell temperature to a specific value. For example, if GPCRb is done with a low temperature log file from a bare cell (not in the final enclosure) in a temperature chamber where a fan blows -10deg.C air onto the cell, then GPCRb will return temp a and k that reflect almost no cell heating and you will observe a loss of capacity in a test.

    The gauge will learn temp a and k during discharges so this will adapt eventually but it will change based on the load. So if the load is light and the cell won't heat up, then temp a and k reflect this situation. If you follow with a high load discharge that heats up the cell significantly, then temp a and k won't tell the gauge that this will happen and it will take another discharge cycle for the new values to take effect. It's possible to mitigate this by setting boundaries for temp a and k.

  • Thanks, Dominik,

    then how long duration for present average P?

  • It will initialize an IIR averaging filter at the start of discharge with the measured current and then keep calculating the average once a second. The time when this actually takes effect depends on the next simulation trigger (temperature change or DOD grid point)