Other Parts Discussed in Thread: BQSTUDIO
Hello,
this post will be a big one in which I raise many questions, my apologies. But the overall process of setting up EOS configuration for production on the bq35100 is not clear to us.
The first step is to run the EOS draining experiment (as described in 5 Testing Procedure ) and capture the scaled resistence and impedance until a complete discharge on every 1% discharge step.
After that, we want to set up other devices for production with the knowledge we gained from the experiment.
For this, we are following the instructions that were published here
Regarding this document we are not sure what exactly is done automatically by the gauge and what needs to do be done manually. The following points will be a mix of questions and our interpretation of the instructions.
- "The only parameter that needs to be updated for initial EOS learning is the New Batt R Scale Delay"
- Question: What is meant by "initial EOS learning"?
- "this needs to be set to a value after the passivisation layer is expected to be gone."
- Question: The delay is given in "readings". What exactly is a "reading"?
- "The passivisation layer can be seen as the initial resistance measurement being higher than expected, and then decreasing after subsequent pulses. Once resistance stops decreasing and is flat, the passivisation layer has been removed."
- Question: Is it this point in the graph?
- "After the initial learning is setup, the most important value to configure from the resistance data is the EOS Trend Detection"
- We choose a value for EOS Trend Detection in such a way that the EOS flag will be set a few time units before the battery is completely depleted
Formula: Short Trend Average > Long Trend Average × (1 + EOS Trend Detection / 100)
- "To estimate the long and short filtered values we can use the following formulas from the TRM."
- The calculation is based on the set of values captured in the EOS experiment?
- Short Trend Average = Impedance × 1/DF1 + Previous Impedance × (1–1/DF1)
- Long Trend Average = Impedance × 1/DF2 + Previous Impedance × (1–1/DF2)
- Previous impedance for the first data point is assumed to be zero?
- How can we use the same formulas when short trend average and R short trend filter have completely different data ranges (max value 8355712 vs. 255)
- More questions: What about all the other values:
- R Table Scale
- R Table Scale Update Flag
- EOS detection pulse count (will be set automatically by the gauge I guess and is set to 0 when the NEW_BATTERY command is issued)
- EOS detection pulse count Thrhd
- Short trend average
- Long trend average
- Question: What do we need to do about the Chem ID? We use a battery type (Tadiran SL-2870T) which is not covered by the available presets within bqStudio.
Sorry for asking so many questions, and thank you in advance for taking the time,
Hendrik
