Welcome to our Community
Wanting to join the rest of our members? Feel free to sign up today.
Sign up

What you should know about Anafi batteries

Gipsz Jakab

Well-known member
Joined
Dec 22, 2019
Messages
739
Reaction score
344
Age
57
Location
Hungary
What you should know about Anafi batteries:
  • Type of LiHV is a high voltage Lithium polymer
  • 2 cells / battery = 2S (2 are connected in series)
  • Max. capacity 2700mAh
  • Max. cell voltage 4.35V / cell 8.7V / package (LiPo 4.2V 8.4V / package)
  • Nominall voltage 3.8V / cell (LiPo 3.7V / cell) 7.6V / pack (LiPo 7.4 / pack)
  • Min. voltage (below which it is severely damaged) 3.2V / cell 6.4V / package (this is also the voltage of the Anafi cut-off)
  • Recommended maximum discharge is 80% of the total capacity (2700-80% = 540mAh) then approx. 3.65V / cell 7.3V / pack remains in the battery.
  • So it is advisable to land by at least 20%, then the lifespan will not be significantly reduced. Under 3.5V / cell 7V / package, harmful processes are already accelerated.
  • If it is still below 20%, it is recommended to charge it to at least 60-65% 3.8V / cell (7.6V / pack) within 20 minutes (For example with a powerbank, on the spot)
  • Charging according to 1C (2700mAh x1) max. With 3A. As the internal temperature has a negative effect on the service life, it is recommended to charge less (2A mobile phone charger) or the USB socket of the computer.
  • Storage 3.8V / cell (7.6V / pack) at 62-65% in a cool dry place
  • It is not recommended to leave it in a fully charged state, even if the built-in smart circuit automatically discharges to a 65% charge level. This few days of 100% charge also has a detrimental effect on longevity.
  • Life life 200-150 cycles (but only if always done according to all specifications), shorter than 300 cycles of LiPo
  • According to some sources, it is recommended to discharge only up to 50% during the first 5 cycles (this initiation)
  • Its internal resistance (IR) increases continuously during aging (anode, cathode oxidation), therefore an increasing proportion of the transmitted and absorbed energy is converted into heat and therefore its capacity decreases. 6-7 mohm is still a good value, needs to be replaced above 20 miliohm.

In summary:
  • max. 4.35V (8.7V) 100%
  • nominal and storage: 3.8V (7.6V) 65%
  • min. safe: 3.6V (7.2V) 20%
  • absolute min. (shutdown) 3.2V (6.4V) 0-2%

% Values may vary for Anafi. These values are only valid for new and faultless batteries.
Voltage values, on the other hand, are important! These show its true state.
 
Last edited:
Thanks for the info Gipsz. WRT recharging, I've understood that it's important to allow the battery to cool before recharging. So not advised to recharge immediately following a flight and ideally allow 10-15 mins to cool-down. You sem to know a lot about the LiPo battery characteristics -- what is your understanding on this point?
 
  • Like
Reactions: Gipsz Jakab
Thanks for the info Gipsz. WRT recharging, I've understood that it's important to allow the battery to cool before recharging. So not advised to recharge immediately following a flight and ideally allow 10-15 mins to cool-down. You sem to know a lot about the LiPo battery characteristics -- what is your understanding on this point?
Thank you for your comment. I omitted this because it is contradictory to me. The Parrot factory description also mentions that it is recommended to charge within 20 minutes as it may be damaged. I've read the same thing elsewhere. However, if you wait until it cools down and it takes 10-15 minutes then you are already damaged ... I think you have to start the slow charge (around 0.5A, from an energy bank) while it can cool down slowly because this charging current is not really heating up. For me, the USB port didn't even get lukewarm while charging.
I don't dare say this with complete certainty, but it seems logical.
 
From the batteryuniversity.com/

"Simple Guidelines to Prolong Lithium-ion Batteries

Do not discharge Li-ion too low; charge more often. A random or partial charge is fine. Li-ion does not need to be fully full charged as with lead acid.
Heat the battery to room temperature before charging. Do not charge below freezing.
Limit the time the battery resides at 4.2V/cell (full charge), especially when warm.
Moderate the charge current to between 0.5C and 0.8C for cobalt-blended lithium-ion. Avoid ultra-fast charging and harsh loading.
When possible, lower the charge voltage limit to prolong battery life.
Keep the battery cool. Move it away from heat-generating environments. Avoid hot cars and windowsills that are exposed to the sun.
It is not necessary to unplug the laptop from the power grid when not in use. The charger stops charging when the battery is fully charged.
When the SoC fuel gauge becomes inaccurate, calibrate smart batteries by applying a deliberate full discharge and charge.
Add some charge before a long storage. The charge level is not as critical as cool storage."
 
One more piece of information:
I also have a DJI Spark drone. Its battery is also LiHV type (3 cell). I looked through my charged flights on the AirdataUav and it can be seen that one of the batteries I charged 61 times is still 94%. According to the logs, I landed around 20% most of the time, but there were 10% and 40% as well.
When I landed at 10%, the voltage of the cells did not go below 3.5V either.
So it can be concluded that the 3.7 V (43%) landing recommended by Solaris8x86 is too cautious an approach. I recommend a 3.6 V threshold for the landing voltage, approx. 18-22%
With similar logic, the Spark would be a 5-minute drone. Take-off and landing 1.5 + 1.5 minutes and 2 minutes left for video recording ...
Of course, the less we spend and the less we drain, the longer it works, but then we would have to start at 80% and land at 40% and then we could fly for 5-6 minutes ... That’s stupid. We want to use it, we didn't buy it as a room decoration ...

Gentle but useful use: possible between 100% (8.4 V) and 20% (7.2 V). Charging should be no more than 2A, recharge before flight to spend as short a time as possible by 100%. Charge the 20% batteries to 65% (7.6 V) that day and store them until you recharge.
I’ve always done this since the beginning (2017) and all items (3 Spark and 3 Anafi) are over 94% ...Some are 99%!
Spark akku.jpg
 
Last edited:
Isn’t there a mistake in you first message?
In the summary, you wrote that the max voltage is 4.2 (8.4v) 100%.
But for those Lipo HV, the max voltage is 4.35, so 8.7v for 100% of two cells.
Do i make a mistake myself?
 
Same question??
Are we talking about LiPo or HV LiPo?
Is the recommendation charge to only 8.4v for HV LiPo and how do we monitor for that specific voltage while charging the Anafi battery using USB C and/or USB C PD?
 
Isn’t there a mistake in you first message?
In the summary, you wrote that the max voltage is 4.2 (8.4v) 100%.
But for those Lipo HV, the max voltage is 4.35, so 8.7v for 100% of two cells.
Do i make a mistake myself?
Thanks! it was a spelling I corrected.
 
Same question??
Are we talking about LiPo or HV LiPo?
Is the recommendation charge to only 8.4v for HV LiPo and how do we monitor for that specific voltage while charging the Anafi battery using USB C and/or USB C PD?
For Anafi, 100% charge does not need to be monitored. The built-in smart circuit does this. It does not charge above the maximum of 8.7V.
And the 55-60% charge required for storage will last until the first two green LEDs are continuous, the third will start flashing. And with the modification made by our friend Anafipilot1, you can check the exact battery voltage, which should be 7.6V. If you need to charge more, if necessary you can discharge it with a little flight.
And you can measure directly with a multimeter (until are asleep)
 
Last edited:
I see you have corrected the first post, (yours) which caused me to question your values. I do understand how the smart circuit works, although how well is questionable from the stand point of charge AND (dis-charge).
 
I did a test flight this morning. I flew with a fully charged battery until the storage voltage dropped. I landed a little early because I didn’t expect the unloaded battery voltage to rise after landing.
As this battery still has a capacity of 97%, the voltage values and% values are still correct:
Battery monitor.jpg
 
It is strange that the After Landing FF6 % is lower than the landing one but the reported voltage is higher. In fact whilst flying the After Landing voltage of 7732mV would give a figure of greater than 60% but the final figure reported by FF6 is only 55%.
 
It is strange that the After Landing FF6 % is lower than the landing one but the reported voltage is higher. In fact whilst flying the After Landing voltage of 7732mV would give a figure of greater than 60% but the final figure reported by FF6 is only 55%.
Yes it was interesting! The percentage charge indicator stopped when landing. The voltage then rose, which he no longer followed.
I don't think that's a mistake. Anafi has finished flying and thinks the measurement is no longer needed ... :)
Logically: in this case, the battery can only be discharged, it cannot be charged, so it only counts downwards. The voltage often fluctuated (went up when the load was lower) but the charge% did not fluctuate. software balanced, smoothed as an indicator of the fuel level of cars. It doesn't fluctuate in the corners when the gasoline sways ...
 
Last edited:
It is strange that the After Landing FF6 % is lower than the landing one but the reported voltage is higher. In fact whilst flying the After Landing voltage of 7732mV would give a figure of greater than 60% but the final figure reported by FF6 is only 55%.

This is correct because the level is not based on voltage - the level is derived from the charge left in the battery. This is what makes these type of batteries smart, they count the charge going in to the battery when recharging, then count it leaving the battery as it is being used.

Just because off-load the voltage has recovered, extra charge has not magically appeared in the battery - there is still only 55% charge left to deliver to the load.
 
  • Like
Reactions: Dirk_ANAFI
I am having trouble getting my head around the charge in v charge out method of calculating the % battery left. I understand that the software can calculate the amount of charge put into the battery, provided that the USB-C connector is used and it is not charged via the battery connectors as some aftermarket chargers appear to do. But to calculate the actual charge in the battery it must know how full it was before charging started. Is that figure based on the resting voltage before charging or us their an internal table in the firmware that tracks each battery? I also assume that the Maximum Battery Capacity is also calculated from that figure.
 
I am having trouble getting my head around the charge in v charge out method of calculating the % battery left. I understand that the software can calculate the amount of charge put into the battery, provided that the USB-C connector is used and it is not charged via the battery connectors as some aftermarket chargers appear to do. But to calculate the actual charge in the battery it must know how full it was before charging started. Is that figure based on the resting voltage before charging or us their an internal table in the firmware that tracks each battery? I also assume that the Maximum Battery Capacity is also calculated from that figure.
Professional balance chargers display the battery charge after a short test.
I think they put a load on the battery for a moment and measure the voltage drop that occurs at that time. The voltage of a charged battery is high and falls less on load. A depleted battery has a lower voltage and this is greatly reduced by load.
So the voltage itself and the change under load provide information, but in this case under flight at a fairly constant load, monitoring the voltage tells the battery charge quite accurately.
 
@Gipsz Jakab
Could you open up ShowAnafiLog and then open up a bin file from "Show meta data from FDR log" under tools.
I would like to see what your old firmware numbers for battery versions are compared to mine.
Screen_Meta data_1.png
 
Here's another battery. I didn't just charge it twice, that's for sure ... Maybe I drained it twice so much that it counted as one cycle ...
Bin_2.jpg
 

New Posts

Members online

No members online now.

Forum statistics

Threads
5,290
Messages
45,023
Members
7,951
Latest member
Davecat44