BQ20Z65-R1 Gas Gauge: AverageTimeToEmpty (Remaining Time)

Hi Manfred,

did you see the descriptions under standard  SBS Commands, for  the  AtRate (0x04) function  in section A.5 on page 68 of sluu386 and also AtRateTime toEmpty (0x06)  description on page 69. Basically you tell the part the intended AtRate using 0x04 command and then send the 0x06 to return the predicted remaining operating time. The part is using the remaining capacity value it has and the rate information to work out when the pack will hit the terminate voltage.

Mike

Hi Mike

Is it the recommended way to calculate the remaining time by using AtRate and AtRateTimeToEmpty?

The description of AverageTimeToEmpty (0x12) states "The AverageTimeToEmpty() displays state-of-charge information in a more useful way. By averaging the instantaneous estimations, the remaining time will not appear to "jump" around as it does on many of today's systems". So we decided to use the AverageTimeToEmpty instead of any other value. Is there a detailed description available, of how this value is calculated?

Many thanks in advance.

Regards,

Thomas

Hi Thomas,

the description you quote sounds reasonable but I did not see those exact words in TRM , can you give me a reference to where you saw those ?

Just reading the TRM it seems there are three different choices for a  prediction of run time to empty

0x12 Average time to Empty based on the discharge Averagecurrent value over what seesm to be a minute which is returned by 0x0b

0x11 Run Time to Empty based on the discharge current value over what seems to be a second returned by 0x0a

0x06 At Rate time to empty based on the discharge current you specify using the AtRate command 0x04

So you essentially have the choice of predictions based on what seems to be 1 second , 1 minute or some custom rate you specify. The problem with the first two is that neither may reflect your average over a much longer period which is where the At Rate option is good if your load tends to be fairly constant over longer periods, plus it is repeatable, i.e. when you look at the returned value you know it is for the rate you are interested in rather than based on whatever current happens to have been flowing in past second or minute. Clearly 0x12 will be less dynamic than 0x11 which is what the ext you quoted is referening to but it is still varying to some degree based on the current load whereas the AtRate value isn't. You can change the At rate value at any time but if you want to always base your estimates on say 0.5A everytime then that is what you can ask it to tell you.

I am not aware of a detailed description of how the calculation is done but based on the playing around I have done in past and just repeated it just seems to be a simple divide of whatever current is being used into the remaining capacity value .

Here are three examples using different  At rate values which is easier to manipulate with the EVM I have connected on my desk at the moment , it is a z45 but it should be same as z65. The remaining capacity was showing 284mAh so I did three simple tests of entering 284mA, 28 mA and 2840mA value for At rate into the EVM value filed for At rate and  as you can see the At rate time to empty values returned were  60 minutes, 609 minutes and 6 minutes which is what you would expect. Note it does not seem possible to enter exactly 28.4mA for the At Rate , i.e. you can' enter point something so I had to round down and enter 28mA hence I saw 609 minutes rather than 600 minutes.  But it seems a pretty simple calculation it is doing. I believe for the other time to empty functions it will do a similar divide of whatever current is being measured into the remaining capacity value to calculate an answer.

,

Mike

I have another example here where I set up an EVM with a discharge current of 100mA being sensed. I left it a few minutes so that the current and average current values settled at 100mA , plus I entered an At Rate value of 100mA. The Remining Capacity was 378mAh  so divide by 100mA is 3.78 hours which x60 is 226.8 minutes and as you can see below all the 3 Run Time to empty values I reviewed above are saying 227mA  :)

Note the one gotcha you have to watch for is make sure you have CapM set correctly otherwise the reported capacioty value is in mWh and then the run time to empty values obviously don't  correlate so obviously with a mA load.

Hi Mike

Thank you for your answer and the examples you provided.

The quote I posted is taken from the Smart Battery Data Specification Rev 1.1 (Dec. 1998).

Well - generally, my question is answered, as you're not aware of a detailed description. Maybe you could give this as an input to those responsible for the documentation of the gas gauge. I think that there should be an easy way to get more detailed information about the calculation of the values provided by the gas gauge.

Anyway, thanks for your Support.

Regards,

Thomas