Hi,
I have been recently working on configuring a bq27520 chip in a project I am working on. I have been using the EVM to run chemical ID tests and learning cycle tests in an effort to generate a golden binary for programming the fuel gauge with prior to deploying it in an embedded system.
I was able to run several successful chemical ID tests. I decided on the ID 3171. From here, I ran a complete learning cycle that failed. About 1 day later I ran another learning cycle that got "successful" results. I saw that Data Memory-->Gas Gauging-->State-->Update Status 1: changed to 0x02.
I also saw that data memory Ra1 Table had been updated. Although the other "update status" was 0x02, the Data Memory-->RaTables-->Ra1 Table--> Ra Status was 0x05 at the end of the run. Should this be considered a successful learning cycle? The Ra0 table that had been updated during the first "failed" cycle status (Data Memory-->Ra Tables--> Ra0 Table --> Ra Status) changed to 0.
There is also another "update status" in the Data Memory-->OCV Tables-->OCVa0 Table-->Update Status and OCVa1 Table-->Update Status. These values seemed to be at 0x00 the whole time.
I have read that I am supposed to "copy the table that has the learned, ie the qmax and Ra values to the one that are unlearned so both tables will now have the same values" in preparation for creating a golden binary (slua902). Does this mean I am supposed to copy the Ra1 Table (including Ra Status, Ra Flag, Ra Base, and Ra Gain) to every single one of the Ra Tables: Ra0 Default, Ra1 Default, Ra0 Table, Ra1 Table, Ra0x Table, Ta1x Table? It is a little unclear due to their being 6 tables and instead of the implied 2.
Also, do I copy Data Memory-->Gas Gauging--> State --> Qmax Cell1 to Data Memory-->OCV Tables-->OCVa[0/1]-->Qmax Cell?
I am attaching my log file from the run as well as the exported settings before and after the learning cycle.
Thanks for the help,
Eddie
