Max error 143%

Max Error should correct when you next get a Qmax update. Internally we keep track of the last cycle count number  when a Qmax update occured and at the moment that interbal count must be higher than the general count hence the 143% max error calculation. In the final  step to make production golden file  we recommend to load a fresh senc which will have the interbal Qmax cycle count equal to zero and copy the learned golden gg file into that , editing the external cycle count in the learned gg file which you import to be zero also, so they are both zero.  This is explained in section 3 point 5.

1067.slua334b preparing optimised default flash constants for specific battery types.pdf

On the next Qmax update cycle the internal Qmax cycle count will get set equal to the actual cycle count andMax error will become 1%.

Mike,

I ended up using the Wizard "5C. Update Golden Pack" to automate the export, clear hidden data flash, then import and overwrite DF.  It wasn't immediately obvious, however, that the source of the import file was the same file exported immediately after writing the Chemistry file during step "4C. Select Chemistry Manually".  I believe this is what has tripped me up previously and left me with a Golden File contaminated with the wrong chemistry file.  For others reading this, to check the chem ID, go to the SBS tab, Uncheck the "Keep Scanning" box, enter 0008 into the "Manufacturer Access" field at top left and click Refresh.  The Chem ID will be appear in the Manufacturer Access field.

Although I now have a Golden file that gives me 1% Max Error as soon as IT is enabled, I'd like some clarification on when the Wizard writes the Senc file and how to manually produce one, if I were to instead follow the instructions in the document you linked to above, using the Srec Programming function in the "Pro" screen.  Is p211 of SLUA404 a good guide for this?  It appears that one needs to toggle between ROM mode and GG Mode before and after Senc programming.

Thank you,

Jeremy.

Hi Jeremy,

yes SLUA474 seems to describe going into ROM mode to export a senc very well in section 3 although quite why you were doing that lost me slightly. However It sounds like you got there in the end but to be sure please export a gg file from your final  new golden device and double check that the Ra table values and Qmax values shown agree with what you had in the learned device after the learning.

e.g , essentially per slua334b , after you reload* a default senc, you then load the correct CHEM ID info and then reload the edited learned gg file back toresyore your settings and learned parameters.

*Note to load a senc uou also go into ROM mode and use the srec programming "program" buttom on the PRO screen having selected the right default file.  

So usually we do this by loading a fresh default senc. What you described concerned me slightly since it sounded like you were exporting a senc and loading it back . If you have one saved from the start of the learning process on a new uncycled part after you had made your settings and loaded the chem ID then that should work I would think but any exported senc from a cycled device will have the interbal cycle count issue.

If you need me to send you a default senc for your part let me know, you will have to remind me which exact 20z device as I don't see it it named on this Max error 143% thread.

 

Mike

 

 

 

 

 

Hi Mike,

I think the phrase "Write a .senc file with the correct chemical ID to clear the hidden data-flash constants" was the main source of my confusion. As you have suggested, the only way to obtain a Senc file with the correct chemical ID is to first program the gauge with the appropriate Chem file and then export a senc, as it will also contain the chem file. The alternative, as you say, is to load a default senc and then follow up with a chem file.

Instead of "Write a .senc file with the correct chemical ID to clear the hidden data-flash constants", I'd like to see this part of the documentation updated to say "Obtain a .senc file from the web, as it is not provided with the bqEVSW, navigate to the Pro screen, enter ROM mode by issuing SMB Command: 00 to 0x0f00, program the senc file, exit ROM mode by sending SMB command to 08 and then navigate to the bqEasy Screen, 4B Select Chemistry Manually, choose your cell and then click OK-Update Fuel Gauge Data Flash. Perhaps this lack of clarity comes from the way in which the technology has evolved over the years and the far greater number of cells supported today.

An alternative, it appears, is to instead write the default DFI file using the bqEasy 1. Setup, 1B. Load/Read .DFI or .ROM file Screen, followed by the Chem File in 4B. This seems to achieve the same objective, perhaps more cleanly and without the need for manually selecting ROM mode.

Can you confirm that the latter approach clears the temporary registers, as per writing the Senc file? My gauge appears to start off with 1% Max Error using this approach, so I suspect that it does.

Thanks again,

Jeremy.

Thanks Mike - I'm glad I now have this clear - it's been quite a steep learning curve. Next time I'll grab a fresh gauge and before I've done anything else, I'll write the relevant chemistry file and then export the senc, so that I have it available later on, when I need it.

Note that on page 8 of SLUA474, it says;

"6. In the Send SMB Command area set the SMB Command box to 00 and click the ‘Send’ button. This

will restart the firmware"

This should read;

6. In the Send SMB Command area set the SMB Command box to 08 and click the ‘Send’ button. This

will restart the firmware

Thanks for catching and documenting that