Battery Deep Discharge - Slightly Different Question

[LeafPeeper]

I know there are many differing opinions on whether it's a good idea to do an occasional deep discharge to recalibrate and rebalance the battery. There seem to be even more opinions on exactly how to do it. For those that do, I have a question. In all the tutorials, instructions and recommendations, they all talk about fully charging and then discharging to some low percentage. After a cool down, most say to recharge normally. However, none of them talk about that follow up charge. If the discharge helps set a low voltage point and 'reinitialize' the calibration, which charge counts when it comes to the full charge point? Is it the one before the discharge, or the one after the the battery recalibrates the low voltage level?

I tend to do it VERY occasionally, but it occurred to me yesterday that I might be doing it wrong. I fully charge and then discharge to 8-10%, but then after a cool down, I only charge it to 60-70% to put it away. If it's the charge AFTER the discharge that counts, I may not be doing myself any favors. Any thoughts or opinions?

LP

 

[PhantomFandom]

I know there are many differing opinions on whether it's a good idea to do an occasional deep discharge to recalibrate and rebalance the battery. There seem to be even more opinions on exactly how to do it. For those that do, I have a question. In all the tutorials, instructions and recommendations, they all talk about fully charging and then discharging to some low percentage. After a cool down, most say to recharge normally. However, none of them talk about that follow up charge. If the discharge helps set a low voltage point and 'reinitialize' the calibration, which charge counts when it comes to the full charge point? Is it the one before the discharge, or the one after the the battery recalibrates the low voltage level?

I tend to do it VERY occasionally, but it occurred to me yesterday that I might be doing it wrong. I fully charge and then discharge to 8-10%, but then after a cool down, I only charge it to 60-70% to put it away. If it's the charge AFTER the discharge that counts, I may not be doing myself any favors. Any thoughts or opinions?

LP

My "opinion", which is actually well established fact by labs and scientists, is don't do it. LiPo batteries do not suffer from the memory effect that NiCad batteries are susceptible to. Bringing a LiPo battery down to a very low voltage can and will damage it. So you are getting zero benefit from a deep discharge and are in fact potentially causing harm to the cells.

You mentioned both re-calibrate and re-balance as reasons for doing so. Due to no memory effect there is no need to re-calibrate. Balancing is done by the circuitry inside the battery each time you charge the battery. Once a cell starts to vary significantly in voltage from the other cells, it's time to throw away that battery.

 

[BigAl07]

We still do a Deep DisCharge (~8%) on or around the 20th charge cycle. We then followup with a full 100% charge. Just our process and it works well for us.

 

[LeafPeeper]

My "opinion", which is actually well established fact by labs and scientists, is don't do it. LiPo batteries do not suffer from the memory effect that NiCad batteries are susceptible to. Bringing a LiPo battery down to a very low voltage can and will damage it. So you are getting zero benefit from a deep discharge and are in fact potentially causing harm to the cells.

You mentioned both re-calibrate and re-balance as reasons for doing so. Due to no memory effect there is no need to re-calibrate. Balancing is done by the circuitry inside the battery each time you charge the battery. Once a cell starts to vary significantly in voltage from the other cells, it's time to throw away that battery.

Thanks guys. I do understand the argument both ways. No disrespect intended, but there are lots of threads arguing the point. Unfortunately, regardless of the 'no memory effect' claim, my own experience with LiPo batteries has been less positive. My typical mode of operation is to use my devices until they hit 40-50% and then either put them away or recharge them. Unfortunately, I have noticed that over time this does have a negative effect. After several cycles, they just seem to get wonky - ie: less linear in power loss. I'm not talking about a ton of cycles either - maybe 20 or 30. That's why I tend to do a full cycle every once in a while - but frankly, I don't really see much of a difference....which is why I was asking this question. For the record, even Apple recommends deep cycling once in a while to recalibrate the battery level. Given that their batteries are never user replaceable, it would seem odd that they would recommend that if it's actually bad for the battery (unless you want to get into conspiracy theories).

I didn't want this thread to be a 'to do or not to do' question. This thread was more about what people who believe the recalibration argument think with regard to which full charge is the most important. So for those who think deep cycling 'recalibrates' the battery level, does anyone know what this actually means, or are they just parroting what they've read on other sites? What does the algorithm look like? Is it reading a min and a max and interpolating? If so, which charge sets the max - the one before or the one after?

That's the kind of info I'm looking for...from the many threads, it seems like nobody knows.
Thanks,
LP

 

[LeafPeeper]

Well this seems to indicate both charge cycles are important....unless I'm misreading it.

BU-603: How to Calibrate a “Smart” Battery – Battery University

Learn why calibration is needed and how often it is required

batteryuniversity.com

LP

 

[bernek]

We still do a Deep DisCharge (~8%) on or around the 20th charge cycle. We then followup with a full 100% charge. Just our process and it works well for us.

I honestly believe this is a bad idea and you are actually using up another "cycle" in the battery by doing that ... It is not worth it imho. Like some said its just bad for li-po to go that low and for no reason. I think the idea of DJI telling you to do it (or was back in the P2 era) it is because if you discharge that low you might find problems with the battery for example once cell discharging way lower or some similar issues. It's more of a safety test instead of actually helping the battery live longer.

I never done it and will never want to use a cycle just to torture the battery but this is my opinion only.

 

[LeafPeeper]

So you believe the article linked above is wrong? Not arguing, and that's fine if you do....just wondering. It seems odd that there aren't clearer facts out there....

Honestly, with all my other electronics, I'd be fine with saying 'screw it, I'm not using up a cycle to do that'....but in the case of drones, I worry about the voltages going non-linear and unexpectedly dropping like a rock....followed by the drone dropping like a rock.

LP

 

[skyrex]

As comparison to testing or "torturing" a battery, I have an old mobile phone which I now-days use as a dedicated display for my controller, it's battery unquestionably shot, gets down to ~40-44% then drops to 0-2% and fails to keep the phone alive.
It spent most of it's life using the upper 2/3's SOC (30-100%), I rarely ever let it dip into the teens or single digits SOC.
Now on it's final legs of sorts, it seems the lower 1/3 SOC is useless.
Possibly because of the life it lived, or just because of age.

This is also probably worth noting for the above, it's a single cell Li-Ion vs 3-cell arrangement of DJI batteries.

 

[BigAl07]

I honestly believe this is a bad idea and you are actually using up another "cycle" in the battery by doing that ... It is not worth it imho. Like some said its just bad for li-po to go that low and for no reason. I think the idea of DJI telling you to do it (or was back in the P2 era) it is because if you discharge that low you might find problems with the battery for example once cell discharging way lower or some similar issues. It's more of a safety test instead of actually helping the battery live longer.

I never done it and will never want to use a cycle just to torture the battery but this is my opinion only.

I guess we'll just have to agree to disagree.

For the record I've got some Phantom3 batteries with a dizzying # of charges and flights on them and they are still going strong and being used in one way or another every week. I have a large # of LiPo (smart and not-smart) in my inventory and except for a few lemons I've gotten phenomenal service out of them.

I just ran 4 of my Mavic Platinum batteries down into "Deep Discharge" and 3 went to 9% but one of them stopped at 10%. It's worth noting the one that cut off just slightly sooner is also the oldest one in my MPP fleet. Just noting an observation.

 

[WAXMAN]

I guess we'll just have to agree to disagree.

For the record I've got some Phantom3 batteries with a dizzying # of charges and flights on them and they are still going strong and being used in one way or another every week. I have a large # of LiPo (smart and not-smart) in my inventory and except for a few lemons I've gotten phenomenal service out of them.

I just ran 4 of my Mavic Platinum batteries down into "Deep Discharge" and 3 went to 9% but one of them stopped at 10%. It's worth noting the one that cut off just slightly sooner is also the oldest one in my MPP fleet. Just noting an observation.

SOOOOOO, as of today we really don't have a 100% answer to this question. Do what you thing is best.

 

[PhantomFandom]

SOOOOOO, as of today we really don't have a 100% answer to this question. Do what you thing is best.

Well you could say that about climate change and global warming as well. However, the scientific facts are that we do contribute to global warming and that LiPo batteries do not need to be deep discharged. The rest is conjecture and opinion.

 

[LeafPeeper]

Well you could say that about climate change and global warming as well. However, the scientific facts are that we do contribute to global warming and that LiPo batteries do not need to be deep discharged. The rest is conjecture and opinion.

So maybe LiPo batteries don't need to be deep discharged - but they do age and lose capacity, right? If so, who's to say that the FIRMWARE doesn't need to see a full discharge in order to recalibrate it's interpolation algorithm for that reduced capacity?

My point? Maybe LiPo's don't need the cycle, but maybe the "smart battery" as a unit, does.

Unfortunately this thread has turned into the usual discussion over whether to discharge or not, and I stated from the beginning that I didn't want that. I want a better understanding of the smart battery capacity measuring algorithms, and it sounds like nobody knows.

LP

 

[BigAl07]

So maybe LiPo batteries don't need to be deep discharged - but they do age and lose capacity, right? If so, who's to say that the FIRMWARE doesn't need to see a full discharge in order to recalibrate it's interpolation algorithm for that reduced capacity?

My point? Maybe LiPo's don't need the cycle, but maybe the "smart battery" as a unit, does.

Unfortunately this thread has turned into the usual discussion over whether to discharge or not, and I stated from the beginning that I didn't want that. I want a better understanding of the smart battery capacity measuring algorithms, and it sounds like nobody knows.

LP

I have no "Technical" proof about the DJI batteries but from the link you provided and from the real-world evidence I've seen(in my own inventory) it stands that there is at leas some degree of merit to the need to "ReCalibrate" the "Digital Battery".

 

[bernek]

A happy lipo is one that lives its life between 50-80 but since the charge from 80% to 100% gives us important flight time we charge them to 100%.

I always full charge my batteries before flight and land at 40-50%. I have done this and my old p2 vision after hundreds of flights have batteries that offer very close flight times to brand new batteries.

I never charge immediatelly after flight and always store batteries at 45-55% charge.

By using this practice I've flown my P2 so many times without any issues at all.

I'm not trying to impose my way of using lipos I'm just sharing my personal way of using them.

Happy flying everyone ?

 

[Skott2]

Hi everyone,
I have about 6 flights on my original Mavic Pro battery and fly until I get the default warning indicator of a low battery. I land as quickly as I can and let the battery cool before charging it. Pretty standard use I think(?).
I do a full charge to store the battery and the top it off before my next flight. This might be a few weeks later so the discharge process starts and I want to say I think that is set at 10 days. I am not able to check right now).
The last time I charged up I got a single green light on cell 2, a bad cell.
Can I rescue this battery with one of the processes described in this thread?
Thank you.

 

[WithTheBirds]

The deep discharge routine is not likely to be of any significant benefit with a current DJI battery.

If you are interested in understanding why the answers are in the various Texas Instruments white papers and explanations for the operation of “Impedence Track” “gas gauge” (DJI uses TI battery management SOC).

Unlike earlier implementations of coulomb counting to estimate usable capacity and SOC the impedence track algorithms are able to reliably determine battery parameters without taking the cells to low SOC. Absent internal resistance being measured in real time and applied to informing the fuel gauging algorithms periodic deep discharge was required to Calibrate. This is simply because increasing cycle count and age leads too increased internal resistance which shifts the whole voltage/capacity map. The only way for the SOC determination to account for this change in performance was to measure the actual mah delivered- incorporating IR in the estimations provides for higher accuracy of usable capacity providing that SOC and run time to empty can be reliably calculated within 1% for the full service life of the battery .

To answer your question you may see a slight change in usable capacity and health % after a deep discharge and charge. I have 100’s of charge cycles on DJI battery without ever intentionally performing a deep discharge.

The fact it is still recommended might simply be a carry over from practices that were applicable to less advanced fuel gauging schemes.

 

[LeafPeeper]

The deep discharge routine is not likely to be of any significant benefit with a current DJI battery.

If you are interested in understanding why the answers are in the various Texas Instruments white papers and explanations for the operation of “Impedence Track” “gas gauge” (DJI uses TI battery management SOC).

Unlike earlier implementations of coulomb counting to estimate usable capacity and SOC the impedence track algorithms are able to reliably determine battery parameters without taking the cells to low SOC. Absent internal resistance being measured in real time and applied to informing the fuel gauging algorithms periodic deep discharge was required to Calibrate. This is simply because increasing cycle count and age leads too increased internal resistance which shifts the whole voltage/capacity map. The only way for the SOC determination to account for this change in performance was to measure the actual mah delivered- incorporating IR in the estimations provides for higher accuracy of usable capacity providing that SOC and run time to empty can be reliably calculated within 1% for the full service life of the battery .

To answer your question you may see a slight change in usable capacity and health % after a deep discharge and charge. I have 100’s of charge cycles on DJI battery without ever intentionally performing a deep discharge.

The fact it is still recommended might simply be a carry over from practices that were applicable to less advanced fuel gauging schemes.

Thanks WithTheBirds. Let me try to parse what you said to see if I captured the spirit....the TI battery management routine that DJI uses incorporates internal resistance measurements into the capacity measurement, so you don't need to do a discharge to recalibrate the state of charge scale as the battery ages. Did I get it right?

THAT'S the kind of info I was looking for.
LP

 

[WithTheBirds]

My "opinion", which is actually well established fact by labs and scientists, is don't do it. LiPo batteries do not suffer from the memory effect that NiCad batteries are susceptible to. Bringing a LiPo battery down to a very low voltage can and will damage it. So you are getting zero benefit from a deep discharge and are in fact potentially causing harm to the cells.

You mentioned both re-calibrate and re-balance as reasons for doing so. Due to no memory effect there is no need to re-calibrate. Balancing is done by the circuitry inside the battery each time you charge the battery. Once a cell starts to vary significantly in voltage from the other cells, it's time to throw away that battery.

Your opinion has no relevance to the Op's question.

What you say about LiION chemistry and memory effect is well established.

What you seem to have missed is that the OP is specifically interested in the behaviour and operation of the battery monitoring electronics.

Firstly the proposed deep discharge will not cause any significant harm- it can't. This is simply because the battery management SOC will turn off the B+ connection FET well before the cell voltages get anywhere near the minimum required for damage to occur. LVC is set at 3V (under load). Cell damage is around 2.7V, usually slightly under.

As to any benefits that might be realised by deep discharge? There aren't many. This isn't because of the LiPO chemistry- it is directly related to the advanced performance of the impedance track and fuel gauging algorithms implemented in the Texas Instruments SOC. While incorporating measured IR in the coulomb counting algorithm does and exceptional job of maintaining accuracy with cell ageing it can be improved by a deeper discharge. Packs that are routinely flown to higher remaining SOC at end of use in particular might see a slight improvement in life % (estimated capacity/manufactured spec) with a corresponding improvement in remaining % and time to empty estimations. You might also see an improvement in cell balance where that is a problem as the longer recharge cycle will allow more time for the balance bleed resistors to be applied to individual cells during the charge cycle.

 

[WithTheBirds]

Thanks WithTheBirds. Let me try to parse what you said to see if I captured the spirit....the TI battery management routine that DJI uses incorporates internal resistance measurements into the capacity measurement, so you don't need to do a discharge to recalibrate the state of charge scale as the battery ages. Did I get it right?

THAT'S the kind of info I was looking for.
LP

You certainly don't need to do it as frequently as you might with a less advanced management algorithm (no IR measurement). You can theoretically realise very slight improvements in some cases though (based on TI literature). TI claims within 1% accuracy over the service life of the battery so Im not sure it is worth doing on that basis alone.

 

[WithTheBirds]

Well you could say that about climate change and global warming as well. However, the scientific facts are that we do contribute to global warming and that LiPo batteries do not need to be deep discharged. The rest is conjecture and opinion.

Tou don't need to rely on conjecture/opinion. If you are interested I how the battery management implementations work there is some great information on the Texas Instruments website- simply search for "impedance track".