Does discharging the battery to 8% really do anything?

[dannybgoode]

Airdata flagged one of my batteries as due a basic service which involves running it down to 8% and then charging it.

According to their blurb taking it down to 8% resets the battery's internal software or something. Is this actually the case and can anyone give me the full technical low-down on it all?

I've done it as a matter of course but would be interested to know the finer details.

 

[old man mavic]

i believe it has to do with if one of the cells differs in voltage from the rest and as you say it resets the balance part of the batteries charging circuit also DJI do mention it in the manual i have well over 40 charges on my batteries, and have never done it and when i check my batteries in the app the cell voltages are all similar,if i ever got a message to do it i would give it a try but until then i am not going to bother air data are just following best practice in their opinion i would think

 

[dannybgoode]

i believe it has to do with if one of the cells differs in voltage from the rest and as you say it resets the balance part of the batteries charging circuit also DJI do mention it in the manual i have well over 40 charges on my batteries, and have never done it and when i check my batteries in the app the cell voltages are all similar,if i ever got a message to do it i would give it a try but until then i am not going to bother air data are just following best practice in their opinion i would think

Thank you. Yes that makes sense. To be honest if it wasn’t for AirData I wouldn’t bother but I can’t cope with having outstanding maintenance tasks so I just do it .

Was out flying today so the two batteries that need doing are low anyway so it’s not much of a chore.

 

[The Editor]

Airdata flagged one of my batteries as due a basic service which involves running it down to 8% and then charging it.

According to their blurb taking it down to 8% resets the battery's internal software or something. Is this actually the case and can anyone give me the full technical low-down on it all?

I've done it as a matter of course but would be interested to know the finer details.

The reason for doing this (and this has been discussed previously on the forum) is as a pack ages the internal resistance of the cells increases as well as the overall capacity decreases.

By periodically running the packs down to Low Voltage Cut off this effectively resets or tells the firmware where ‘zero percent is’ since this changes over time.
The algorithms do a guestamation on remaining capacity displayed as a percentage which is never accurate (although recent Texas Instruments chips have got better at it).
It’s never been great and DJI took this approach to try and make reading a batteries capacity/depletion easier for beginners or novices at RC.

Unfortunately percentage is not absolute, voltage is, which is why I never take any notice of the percentage readout on the screen and always go by lowest cell voltage under load whilst I’m airborne.

 

[Elude]

What is the lowest safe voltage ?

 

[GaryR.]

What is the lowest safe voltage ?

3.65 volt

 

[Deleted member 94047]

3.65 volt

That seems rather high; especially if it is still under load. I usually aim to be on the ground or near it when one of the cells is just below 3.60V which is usually around 20 - 25% of battery.

Edited for clarity.

 

[DanMan32]

The percentage displayed I believe is based mostly if not entirely on voltage.
Though I plan to do a map of displayed percentage to voltage foe my M2 and P3.

 

[GaryR.]

That seems rather high; especially if it is still under load. I usually aim to be on the ground or near it when one of the cells is just below 3.60V which is usually around 20 - 25% of battery.

Edited for clarity.

For long time storage of lipo batteries the voltage should be more than 3.65 volt and under 4 volt.
Under load, a lipo must never be discharged below 3.3 volts per cell, otherwise this will destroy the lipo completely.
This can be red on every lipo guide.

Edited for complete clarity!

 

[The Editor]

The percentage displayed I believe is based mostly if not entirely on voltage.
Though I plan to do a map of displayed percentage to voltage foe my M2 and P3.

Good luck with that. In any case, it will change over time/age of pack.

There is no direct correlation between voltage and percentage. If you look at 20 graphs you’ll get 20 different answers/results.
Percentage is a guess.
Voltage is absolute.

 

[Thomas B]

Maybe just get a charger that can charge to 100%, charge to 60%, or discharge to 60%. Then it just happens!

 

[DanMan32]

Good luck with that. In any case, it will change over time/age of pack.

There is no direct correlation between voltage and percentage. If you look at 20 graphs you’ll get 20 different answers/results.
Percentage is a guess.
Voltage is absolute.

Actually no. The Texas Instruments chip they use goes by several variables to determine SoC. Voltage deviates with load, due to battery internal resistance so voltage is not absolute.

 

[The Editor]

Actually no. The Texas Instruments chip they use goes by several variables to determine SoC. Voltage deviates with load, due to battery internal resistance so voltage is not absolute.

Yes it is - it is the absolute potential difference between the cells at any given point they are measured.
Thus if you undertake a full throttle climb out, the pack will be put under maximum load. This will instantly show any sag in the cells due to higher internal resistance and alert you immediately of a potential problem with a cell or the pack in general. You are getting an absolute voltage reading instantaneously under maximum load.
I trust that and 30 plus years of flying RC that way far more than any algorithm in a chip.
I have NEVER had a crash due to not properly monitoring or understanding my packs.
The same can't be said for the DJI idiot/smart monitoring software giving a guess shown as percentage/fuel gauge.

I'll carry on doing as I have done thank you.

 

[DanMan32]

The thing is, estimated percentage attempts to help determine time remaining. Voltage might not do that as the voltage may go down quickly at lower SoC.

 

[WithTheBirds]

Yes it is - it is the absolute potential difference between the cells at any given point they are measured.
Thus if you undertake a full throttle climb out, the pack will be put under maximum load. This will instantly show any sag in the cells due to higher internal resistance and alert you immediately of a potential problem with a cell or the pack in general. You are getting an absolute voltage reading instantaneously under maximum load.
I trust that and 30 plus years of flying RC that way far more than any algorithm in a chip.
I have NEVER had a crash due to not properly monitoring or understanding my packs.
The same can't be said for the DJI idiot/smart monitoring software giving a guess shown as percentage/fuel gauge.

I'll carry on doing as I have done thank you.

No doubt you aren't alone in your obvious suspicion of the algorithms employed to monitor battery performance/condition. Anyone among us who flew LiPO powered RC models quickly became accustomed to the necessity to monitor cell voltages, particularly given it proved to be the only easily implemented means to track pack capacity and health.

The reality is the currently implemented algorithms do a significantly better job than we might hope to, regardless of what practice we may choose to employ.

If you consider simple observations of voltages in flight may prove to allow for a more accurate and reliable indication of SOC and or remaining flight time than a system that also includes temperature, current (instantaneous and average) and cell condition (cumulative age and use related degradation) in the equation by all means continue. We need to be ignorant of the tech to suggest to others it is unreliable however.

The only occasions where it seems the battery smarts fail is in circumstances where a partially charged pack is flown and this seems to be in most instances where the self discharge function has commenced.

If you are interested in how the current battery monitoring schemes operate the impedance track and fuel gauging white papers published by Texas Instruments are a great read.