All the business on Lithium life-span

[Biker44]

The author of this article doesn't have a lot of respect for his suppliers - but he also makes some very bold and potentially very useful statements about prolonging battery life - should we trust what he says?

How to Prolong Lithium-based Batteries - Battery University

... Lithium-ion suffers stress when exposed to heat and kept at a high charge voltage. ... anything that dwells above 30°C (86°F), and ... higher than 4.10V/cell. ... Exposing the battery to high temperature and being at full state-of-charge for an extended time can be more damaging than cycling. Manufacturers do not like to talk about these environmental conditions and release information only in confidence when so requested.

... Most packs last three to five years, less if exposed to high heat and if kept at a full charge. ... The worst condition is keeping a fully charged battery at elevated temperatures, which is the case when running a laptop on the power grid. Under these circumstances the battery will typically last for about two years, whether cycled or not. ... Satellites and electric vehicles are examples where longevity is important.

... At a charge to 4.10V/cell, the battery holds a capacity that is about 10 percent less than going all the way to 4.20V/cell. In terms of optimal longevity, a charge voltage limit of 3.92V/cell works best but the capacity would be low. Besides selecting the best-suited voltage thresholds, it is also important that the battery does not stay in the high-voltage stage for a long time and is allowed to drop after full charge has been reached.

The voltage threshold of commercial chargers cannot be changed, and making it adjustable would have advantages, especially for laptops as a means of prolonging battery life. When running on extended AC mode, the user would select the "long life" mode and the battery would charge to only, say, 4.05V/cell. This would get a capacity of about 80 percent. Before traveling the user would apply the "full charge mode" to bring the charge to 4.20V/cell. ... A slightly larger pack compensates for the reduced runtime. Another option to extend battery life is removing the pack from the laptop when running on the power grid.

If the above is correct, then it should be worthwhile (for people like me who don't really do long journeys) to partially charge the battery.

If I've got it all correct, how would I arrange 80% charge? Use a timer, only one or two hours at a time? Charge only until the Green indicator light on top of the battery has started to appear?

 

[flecc]

This often crops up. That article is reliable, but the gains for such treatments are often very small. It's not possible to reliably charge to 80%, simply because battery meters are not precisely accurate so the start point of charge isn't known. Also, your e-bike manufacturer has taken the battery's needs into account, the BMS will not charge the cells well above 4.1 volts and there's a low voltage cutoff to prevent stress at the low end of the charge.

Best advice is just use your battery and charge it frequently to full.

The Battery University advice is for all lithium batteries and doesn't take into account the sophisticated electronic battery management that's in our e-bike batteries. Have a look at all the electronics in this one for example.
.

 

[Biker44]

This often crops up. That article is reliable, but the gains for such treatments are often very small. ... your e-bike manufacturer has taken the battery's needs into account, the BMS will not charge the cells well above 4.1 volts and there's a low voltage cutoff to prevent stress at the low end of the charge. Best advice is just use your battery and charge it frequently to full.

The charger to my nominal 24V battery claims to deliver 29.4v (ie 4.2v/cell, the figure used by this BU boffin who thinks the manufacturers are parsimonious with information). My ancient Fluke almost agrees, it says that the BMS delivers 29.45 volts.

So does the sophisticated BMS electronics recognise when it's been charging a battery for a reasonable length of time and, and drop the voltage towards the end of the charge?

The Battery University advice is for all lithium batteries and doesn't take into account the sophisticated electronic battery management that's in our e-bike batteries. Have a look at all the electronics in this one for example.

Hmmmm ... so what if I was determined to be contrary, could I fit a (switchable?) shunt to the BMS and force it to deliver a lower voltage?

Or would I be better to start afresh and lash-up a charger to use when tootling around town, keeping the original charger for those times when I plan a longer trip?

Or would it all be a waste of time ....

 

[flecc]

You say the charger delivers that voltage, but the battery management system (BMS) inside the battery doesn't just throw charge at it in that way. It monitors the cells individually, cutting off each one when it reaches the safe level while continuing to top up the others to equalise them. None of them necessarily see the charger voltage which has headroom to ensure full charging can be attained.

I think it would be a waste of time to attempt short charging or lower voltage charging, and I doubt it would make any measurable difference to life, and certainly not enough to justify any expenditure of time or cash on it. The Battery University argument though accurate is highly theoretical, and much more is needed to get real gain.

For example, lithium batteries in satellites last many years, but to achieve that isn't practical on e-bikes. They do it by using batteries that are 7 times the capacity that's necessary, the one seventh daily discharge during the out of sunlight phase being replaced with that one seventh each time the sun is in view of the photocell array again.

So to imitate that you need a battery 7 times it's present size and weight!

 

[Scottyf]

The BMS in most bike batteries off the shelf should be balancing all the time. They never stop.
So you could only part charge your battery to prolong a little bit extra life. But that would mean knowing what capacity you've used and then charging on a timer to get it about 80%

It simply is alot of hassle for a battery that in 2 years time your going to be able to buy for £150 (Based on 36v 10ah).

Lipo dedicated chargers have various settings to short charge (To say 4.15v) or again I can watch the cell voltage and stop it. I just balance charge up to 4.15v and then use it. I don't go past much 3.8v when using the bike as I have a large battery.

Again its pain in the bum and Lipo isn't that badly priced that its a nightmare to replace them.

I'd say if you can be bothered to watch a timer then do it. But for the extra life span should it work I don't think its worth it.

The temperture and age will eventually kill the battery down to 2/3 years anyway...

 

[Biker44]

You say the charger delivers that voltage, but the battery management system (BMS) inside the battery doesn't just throw charge at it in that way. It monitors the cells individually, cutting off each one when it reaches the safe level while continuing to top up the others to equalise them. None of them necessarily see the charger voltage which has headroom to ensure full charging can be attained.

Ah, I'm obviously confused with which does what.

The Battery University argument though accurate is highly theoretical, and much more is needed to get real gain.

This is what the Battery University is claiming:

From the article at How to Prolong Lithium-based Batteries - Battery University

... or rather, they would be claiming a much better life from charging at 4.05 volts if their graph results were any good, and actually showed such a result! Maybe 4.05 is really not so different from 4.20.

For example, lithium batteries in satellites last many years ... by using batteries that are 7 times the capacity that's necessary

Run that past me again - satellites are even more concious of extra weight than we are. How are they getting their extra life? Is it from a very low discharge rate? (1/7th C divided by 45 mins perhaps) Do they charge at a low voltage suggested by the BU and used by Scotty? (3.92V/cell BU or 4.15volts Scotty)

It might simply be they develop a very high internal resistance in the bitter cold! Though I believe satellites are carefully insulated.

 

[flecc]

Ah, I'm obviously confused with which does what.

Our chargers are not smart chargers, the whole control of charging is with the battery's internal BMS.

That chart is historic, today's lithium batteries are rated for at least 500 charges and many of the li-polymer types for 800 to 1100 charges. The e-bike manufacturers together with their battery manufacturers design the BMS to suit, so I think it's unlikely that you'd guess better than they do.

Run that past me again - satellites are even more conscious of extra weight than we are. How are they getting their extra life? Is it from a very low discharge rate? (1/7th C divided by 45 mins perhaps) Do they charge at a low voltage suggested by the BU and used by Scotty? (3.92V/cell BU or 4.15volts Scotty)

It might simply be they develop a very high internal resistance in the bitter cold! Though I believe satellites are carefully insulated.

Weight is critical in satellite design, but the impossibility of popping in new batteries is even more so! Hence the excess weight wins. The battery isn't huge anyway since the current demand is low. Their battery life is entirely due to that cyclic part charging of a very large capacity, this minimising the chemical stress. They are indeed very well insulated, and the battery certainly doesn't suffer from high cell resistance since that would shorten life.
.

 

[Scottyf]

Having such a large battery and lower charge and dischage effects would lengthen life.

Even in Ebikes if you carried a 10ah battery vs a 20ah battery then not only do you have more range should you want it but the stress on each cell when a current is required is also halved. This in itself actually helps the life cycles. It also allows the cells to stay more in balance too.

Its always gonig to be a hot topic as it matters more and more as we progress through alternative fuels.

 

[Biker44]

Our chargers are not smart chargers, the whole control of charging is with the battery's internal BMS.

Got it. The charger that came with this machine is a precision instrument, but only as regards producing a dependably constant voltage wouldn't can't damage the cells even if the BMS went to open circuit. (Not a desirable situation however, the cells tend to damage each other).

The e-bike manufacturers together with their battery manufacturers design the BMS to suit, so I think it's unlikely that you'd guess better than they do.

They might not appreciate the degree of my tight-fistedness! Or previous generations of e-cyclists may never have much been from my demographic, not travelling very far.

Their battery life is entirely due to that cyclic part charging of a very large capacity, this minimising the chemical stress.

So they might be 4/7th charged at the end of 45 minutes in the sun and 3/7th charged at the end of 45 minutes in the dark? That cycling causing minimum stress.

 

[flecc]

The battery and bike manufacturers design the BMS for best life since that has been the market's demand and an obvious need. Any greed they portray is exhibited in the prices they charge!

I've no doubt the satellite batteries operate efficiently fairly close to fully charged. The life gain is from the huge disparity of supply to demand and the slow and very small proportion of daily charge input. The solar powered charger will see a very low input resistance due to those factors, hence the generation of minimal chemical stress.

I'm confident that playing with small variations in charge voltage will gain little and could result in no gain if the usage approached the lower charge content levels. The worthwhile life gain comes mainly from an excess of capacity to need. That's because the high demand to supply ratio of our motor/battery combinations severely chemically stresses the cells, that more than anything else limiting life.
.

 

[NRG]

I think you'll find most chargers are CC / CV types. The BMS provides over and under voltage protection, discharge current protection, short cct protection and cell balancing. The BMS wont hold the charge current constant thats the role of the charger and towards the end of the charge will revert to CV to allow the BMS to balance out the cells.

 

[Biker44]

I've no doubt the satellite batteries operate efficiently fairly close to fully charged.

I'm still not sure I understand. When you speak of "fully charged", that's not some kind of fixed point. It depends entirely on the voltage applied. For the purpose of e-bikes, it's been declared (pretty much arbitarily) that "100% fully charged" is based on the storage achieved by 4.20volts.

The Battery University seems to be saying that putting a Li-ion on 4.10 volts will charge it to 90% of "full" capacity and 4.05 volts charges to 80% of "full" capacity.

Then the BU claims that 3.92 volts gets the very longest life (but doesn't tell us how much charge is then stored). Separately. the BU claims that 40% charge is the best level for a battery left in storage.

So, if I put those two nuggets of information together, and add your kindly provided satellite information, it seems as if the maximum possible life is attained by charging at 3.92 volts and cycling between about 34% and 40% of what we arbitarily call "full charge".

In other words, I have a battery which might push me along about 25 miles - that's if I drain it from 100% to 0%. For maximum life I should restrict myself and only drain it from 40% to 34%. I think that's about 1 mile.

It gets even worse, because, judging again by the satellite useage, stress on the battery is minimised by taking about 45 minutes for that discharge.

The worthwhile life gain comes mainly from an excess of capacity to need.

Yeah, I think I can see what you mean. I would need a battery 25 times bigger, and have to travel at 1/10th of my usual speed to really maximise battery life!

Despite your very good advice, I still think a switch on the charger that tops off the battery at 80% charge is a must-have feature on future e-bikes.

 

[flecc]

Despite your very good advice, I still think a switch on the charger that tops off the battery at 80% charge is a must-have feature on future e-bikes.

But still not worth it I believe:

1) The chemical stress of the last 20% is more likely to occur if starting with this low content point, and that equally shortens battery life. Avoiding that means only 60% of the content is available, and that unacceptably shortens range, necessitating very large batteries as you've said.

2) Life depends on discharge rate, and the high discharges of our e-bikes do sufficient damage to prevent long life, regardless of charging methods. Again this is only overcome by using oversize batteries, the more the oversize, the longer the life.

The battery that deals with both these factors to give apparent worthwhile gain will be far too large and heavy for e-bike practicality. Even if it were practical, it's much greater size would carry a much larger price to match, and that extra price could cancel the life gain saving, why I used "apparent" gain.

On scooters to some extent and to a much larger degree on cars, these tactics can be slightly more practical, but mainly on so-called hybrids like the Toyota Prius. That operates it's part time battery cyclically between 20% and 80% of charge, and the battery has an 8 year warranty. In this case it's an NiMh battery, illustrating that being lithium is irrelevant, these are rules for most battery chemistries since it's solely a chemical stress issue.
.

 

[Biker44]

But still not worth it I believe: The chemical stress of the last 20% is more likely to occur if starting with this low content point, and that equally shortens battery life. Avoiding that means only 60% of the content is available, and that unacceptably shortens range, necessitating very large batteries as you've said.

Maybe it's that most e-bike users take longer trips than me. I've still not been further than about 2 miles in any direction, so 80% charge with frequent top-up is entirely practical to me. Similarly, a restriction of "use only middle 60% of charge" is perfectly tolerable.

Is it possible that the weight restriction of e-bikes is such that the fast discharge rate is pretty much the sole enemy of a long life? Even for someone like me, it might not make a jot of difference to keep the battery in the middle 60% of it's charge.

... like the Toyota Prius. That operates it's part time battery cyclically between 20% and 80% of charge, and the battery has an 8 year warranty. In this case it's an NiMh battery, illustrating that being lithium is irrelevant, these are rules for most battery chemistries since it's solely a chemical stress issue.

I get the impression that lithium is the "best" technology for ebikes. Despite it's relatively high cost, it handles the stress of discharge much better than Lead-acid and perhaps better than NiMh.

So is lithium handicapped by life-span, making it the least suitable for life-extending charging routines? Laptop batteries don't last very long - but you tell me that Li-ion is favourite on satellites. Is the lithium chemistry in satellites subtly different (eg better materials) giving it the same or better life than other chemistries? Or could it be that the cycle life of Lithium beats all other chemistries - but only at very slow discharges?

Some of the answers to these questions should be out there somewhere, but I don't detect a lot of interest. Our nearest relatives are laptop and camera and phone users, for whom spares and replacements are relatively cheap.

 

[flecc]

I've still not been further than about 2 miles in any direction,

That's very unusual. Many e-bikers use their bikes for leisure runs, others commute up to 30 or even more miles a day. There's also the fact that people often don't like charging hassle. Many in your position still want enough capacity so they only have to charge once a week for example.

Is it possible that the weight restriction of e-bikes is such that the fast discharge rate is pretty much the sole enemy of a long life? Even for someone like me, it might not make a jot of difference to keep the battery in the middle 60% of it's charge.

Not the sole, but together with bulk, by far the largest elements. If they could have a battery four times the capacity with battery management to suit, and six times the size to give lower cell content density, it could do wonders for life.

So is lithium handicapped by life-span, making it the least suitable for life-extending charging routines?

Not particularly, as mentioned it's mainly about chemical stress, and most battery types can suffer that to a damaging degree.

 

[Biker44]

Not the sole, but together with bulk, by far the largest elements. If they could have a battery four times the capacity with battery management to suit, and six times the size to give lower cell content density, it could do wonders for life.

Right, I think I understand at last.

If I fit a battery around 6 times bigger and 6 times more expensive, I can benefit from cycle life up to 5 times longer, perhaps 10 years instead of 2 years!

 

[flecc]

That's it, the classic catch 22.