Memory proven effect on the Li-ion batteries, LiFePO4 ...

[Obamot]

The entire battery industry had sworn we did not!

That will be of interest to Citro, NLC, Remundo and you!

As has always been suggested here, I have always suspected Li-on of a memory effect:

[...] "we often say that we should not unload or load the Li-on thoroughly .... That there is also no interest in unloading them to their lowest level , since they have no memory effect, etc ...

It turns out that some of the most expensive Li-on packs I've ever used (Apple Minolta and Sony) are the ones I was unloading at the lowest recommended rate that lasted the longest (but I've never managed to overtake 4 years in situation of charges / litterous discharges).

So again, from theory to practice "[...]

https://www.econologie.com/forums/post238944.html#238944

Invalid personal experiences, because not supported by scientific evidence thought!
Yet repeated field experiments and doubts confirmed later by the informed opinion of our chemist. As a reminder:

But something bothers me. I have heard that the Li-on does not perform the same and will have almost a programmed lifespan no matter how often you use it. I have noticed it => ~ 2 1/2 years old.

You are surely referring to cell phone batteries. There are huge disparities depending on the brand and especially the modes of use ... Can you phone a lot with heavy landfills and deep cycling.

Some may be singled out by other users because they have found that relatively advanced but not complete discharges of Li-on and similarly with NiMh would make them last longer. I am part of it even if I know that the prescriptions say that it should not be done since it would have no more effect of memory effect (as with the NiCad) ... the facts are there ... Especially if they are used frequently. Here are some clues:

in principle the batteries "would love" to be charged / discharged all the time. Thus used, it seems to me that this would explain why they would last longer.

For what types of devices I tested this? PCs, razors, SLR cameras, tel. portable, home wireless phones, all in standard size battery (AA, AAA) or not ... and what do I know.

And I especially the example of the opposite: fully fusillées batteries because I had used them without taking into account any cycle of charge / discharge? isis this the reason (relative regularity of the charge / discharge cycles) that would make it last longer and that we could get confused? I do not know.

But this weak point of the batteries that one does not use for varying periods of time is the central point that would make that I would have been interested in an air propulsion vehicle, which theoretically would not suffer from this type of constraints / problems (it would eventually lose part of its air charge, but without affecting the "life" of the storage system, as is the case with conventional batteries). I say theoretically because for the moment nothing convincing is yet on sale (not before 2013 to the last news ...).

Even if it's just my experience and personal opinion ... I doubt that among the users rare are those who are totally delighted with the batteries they use whereas it should be the rule since time ... (I say that with a consistency in the duration of use) and I'm sure that except for those who do not care because they do not look at the expense or those who know from professional experience what type it should or should not buy, users lambdas are a little condemned to change over and over with more or less happiness, but ultimately always with the same disappointment in the end.

Side performance difference from one model to another I have tried everything [...] (and I say that the brands most famous brands lambda, it would hardly be any difference) I admit that you 'In general I would never believe what I will be told about the longevity of batteries / accumulators unless I have tried. So let's stop the salads about the manufacturers, who defend their meadow, most of them are subject to very severe market laws that must constrain them from all sides to some things not very clear ...

This is what our chemist has revealed to us as common business practices:
Marketers argue "new types" and R&D at the forefront with a view to constant improvements. Each time, it is true, we see that the latest versions are a little less bad than the previous ones ... But since innovation would not advance at the same speed that the renewal of the product cycle would not really allow it do (chemistry does not work miracles especially before the summer period or during the Christmas holidays ... lol) at the beginning we would see an improvement in performance, then the manufacturers would change the internal formula and / or the components of the batteries for them. artificially making less efficient ... To redo the “novelty” stroke for us in good times, again with the phrase “who is better”. All the more so since it would allow them to put less expensive components inside during a “break” period ...

In the meantime we would be bloused by believing in the consistency of the products [...]

https://www.econologie.com/forums/post185879.html#185879

And now comes the announcement that slams like a thunderclap ....

Memory effect in a lithium-ion battery

Nickel-cadmium and nickel-metal-hydride batteries. If these batteries are recharged repeatedly after being discharged, they are easily used. Lithium-ion batteries, in contrast, are considered to have no memory effect. LiFePO4-one of the materials used for the positive electrode in Li-ion batteries-which appears already after only one cycle of partial charge and discharge. We characterize this memory effect of LiFePO4 and explain its connection to the particle-by-particle charge / discharge model. This effect is important for the battery, it can be used to reduce the energy consumption of batteries.

Source: http://dx.doi.org/10.1038/NMAT3623
Diagrams in small, but we can buy the article.

Validated and taken over by the Paul Scherrer Institute:
http://www.psi.ch/
And placed on the official website of the Helvetic Confederation:
http://www.admin.ch/aktuell/00089/?lang=fr&msg-id=48489

A memory effect also discovered in Li-ion batteries

Lithium-ion batteries are power batteries used for storing the energy of many commercially available electronic devices. They can store a significant amount of energy for a relatively small volume and weight. Moreover, and until now, they have had the reputation of not being sensitive to the memory effect. This is how the experts refer to a potential deviation of the battery, the latter is caused when the battery is not fully charged or discharged. The result is that the stored energy is only partially available and it is no longer possible to make a reliable estimate of the state of charge of the battery. Researchers at the Paul Scherrer Institute (PSI) and their colleagues at Toyota's research laboratory in Japan have now identified a memory effect in a widespread type of lithium-ion battery. This discovery is of particular importance in view of the imminent arrival of lithium-ion batteries in the electric vehicle market. Their work appears today in the journal Nature Materials.
Even if they are not as "perfect" as the ad would like us to believe, many of the devices we use every day, and which draw their energy from a battery, often have a kind of "memory" . The user, who routinely and cautiously reloads his razor or electric toothbrush before the battery is completely empty, risks a bad surprise after the fact. The battery seems indeed to notice that only a part of its specific capacity has been taken - so that it stops one day to remember that it can deliver more energy. The specialists then speak of "memory effect"; this occurs when the cycling potential of the battery decreases over time due to incomplete charge / discharge cycles. In other words, even if the battery still has the available charge, the potential it provides is at a given moment too low to operate the device. The memory effect therefore has two negative consequences: on the one hand, it reduces the available storage capacity of the battery; and, on the other hand, the correlation between cycling potential and state of charge is shifted, so the state of charge can no longer be determined reliably. The memory effect is well known nickel-cadmium and nickel-metal hydride batteries. For lithium-ion batteries which began to be marketed in the early 1990 years, the existence of such an effect was, however, ruled out so far. Wrongly, as this new study shows.

Consequences of the memory effect for the hybrid and electric vehicle

The memory effect accompanied by its abnormal deviation of the potential in cycling has been identified in one of the most commonly used materials as a positive electrode of lithium-ion batteries: lithium iron phosphate (LiFePO4). In the case of lithium iron phosphate, the potential remains unchanged over a large part of the charge / discharge cycle. The smallest difference in battery potential could therefore be misinterpreted as a significant change in the state of charge. However, in the present case, since the state of charge of the battery is determined by the potential in cycling, a very small deviation of the potential can lead to a significant estimation error of the state of charge. The existence of this memory effect is especially important in view of the imminent arrival of lithium-ion batteries on the electric vehicle market. This effect would particularly affect hybrid vehicles, since under normal conditions of use, these vehicles experience many charge / discharge partial cycles. The engine, in these vehicles, is transformed into a generator, and charges the battery with each braking. The latter normally discharges only partially and assists the engine during the acceleration phases. The many successive charge / discharge cycles that follow lead to isolated memory effects that accumulate to create a significant memory effect, as shown in this new study. This induces a poor estimation of the state of charge of the battery, in the case where the state of charge is estimated by software that is based on the current value of the potential.

The causes of the memory effect

Research into the causes of memory effect, such as charge and discharge of batteries, has been studied at the microscopic level. The electrode material - in this case the lithium iron phosphate (LiFePO4) - is composed of a multitude of particles of the size of a few microns, which are loaded and unloaded one after the other. Researchers refer to this model of charge / discharge called "multi-particle model". The charge therefore proceeds particle by particle and involves delithiation. A fully charged particle therefore no longer contains lithium and is composed only of iron phosphate (FePO4). Conversely, the discharge consists of the reverse reaction, the lithium atoms react again with the electrode material so that the iron phosphate (FePO4) becomes lithium iron phosphate (LiFePO4). Modifications of the lithium content, which are associated with charge / discharge states, result in a change in the chemical potential of each particle, thus changing the potential of the battery. However, charging and discharging are not linear processes. Thus, during charging, the chemical potential increases as delithiation progresses. But then, the particle reaches a critical value of its lithium level (and therefore of its chemical potential). A steep transition then occurs at this point: the particles very quickly lose the remaining lithium ions, but their chemical potential does not change. It is precisely this transition that explains the fact that the potential of the battery remains practically unchanged over a long part of the cycling (plateau of potential).

"Rich" or "poor" lithium particles

The existence of this potential barrier is crucial for the appearance of the memory effect. Once the first particles have crossed it and no longer contain lithium, the particles that make up the electrode are divided into two groups. In other words, there is a clear separation between lithium-rich particles and lithium-poor particles (see illustration). If the battery is not fully charged, there remains a number of lithium-rich particles, which failed to cross the potential barrier. But these particles do not stay long at this barrier because their state is unstable under these conditions; they "slide backwards" thus "along the slope of the curve of charge / discharge", which means that their chemical potential decreases. Even when the battery is discharged again and all particles return to the potential barrier, this division into two groups remains. And here lies the crucial point of the memory effect: during the next charging process, it is first the first group (that of low lithium particles) which crosses the barrier, while the second group (particles rich in lithium) lithium) remains "lagging behind". For this "laggard" group to cross this barrier, it must imperatively increase its chemical potential, and this is precisely what causes the characteristic overvoltage of the memory effect ("hump" visible in the illustration). The memory effect is therefore the consequence of the division of the population of particles into two groups, with significantly different lithium contents. These particles must then cross the potential barrier one after the other. The overvoltage, by which the effect becomes visible, corresponds to the additional work that must be provided by the latent particles which have been blocked by the potential barrier after an incomplete charge.

Wait until the memory fades

The time between charging and discharging the battery plays an important role in determining the state of the battery at the end of these processes. Charge and discharge are processes that affect the thermodynamic equilibrium of the battery; but this equilibrium can be restored after a certain lapse of time. The researchers discovered that when the latter was long enough, the memory effect was canceled out. But according to the "multi-particle model", this cancellation occurs only in certain circumstances. The memory effect would disappear only if one waits a long time after a cycle composed of a partial charge, followed by a complete discharge. In this case, the two groups of particles are certainly always separated after the complete discharge, but they are all on the same side of the potential barrier. The division then disappears, because the particles tend towards a state of equilibrium, where they all have the same lithium content. On the other hand, the memory effect is maintained after a partial load and before the incomplete discharge. In this case, the particles are found on both sides of the potential barrier, and this prevents a return to the division of "low lithium" and "lithium-rich" particles.

According to Petr Novák, director of the Electrochemical Energy Storage section at PSI and co-author of the publication, this study sweeps away a long-held misconception: "To our knowledge, no study has looked for targeted way a memory effect in lithium-ion batteries, "he says. "Until now, it was simply assumed that such an effect did not occur. This conclusion, which researchers have come to, he says, is due to a mix of speculation and diligence, which is often fruitful in research: "Our discovery results from a combination of critical questioning and detailed observation. "Continues the researcher. "The effect is tiny: the difference relative to the level of potential is only a few thousandths. But the decisive idea was to look for this effect. In conventional battery tests, complete charge / discharge cycles are performed, not incomplete cycles. To ask the question of the consequences of a partial load, such was the necessary genius. This all-new discovery, however, does not sign a halt to the use of lithium-ion batteries. It is perfectly possible that an intelligent adaptation of the software, within the battery management system, is enough to detect this effect and to take it into consideration in time, underlines Petr Novák. If such an adaptation can work, the memory effect would not hinder the safe use of lithium-ion batteries in electric cars. The ball is now in the engineers 'camp: it's up to them to find the right way to manage the batteries' memory.

According to the "multi-particle model" described here, when charging and discharging the battery, the particles advance one after the other. By particle, we mean here a species of "grain". In other words, the material (LiFePO4) does not present itself in one piece: it is composed of a multitude of grains, in which the crystalline structure is nominally always the same; but these grains have minute differences in size, shape, or crystallographic orientation. This is simply the description of the appearance of a powder. In specialized language, we speak of "crystallites". The whole thing can be described as an alignment of small cubes of roughly the same size, where each is slightly oriented according to its neighbors, which means that the cubes do not all have the same orientation, while they all have the same orientation. same crystalline structure (their cube shape).

Enjoy ...!

See the beginning of the discussion here too: https://www.econologie.com/forums/electrique ... 10540.html

[I Citro]

Obviously the news reached me.
It is not surprising.
I remember my surprise by visiting the Laboratory of Condensed Chemistry at the IMCB Bordeaux.
Researchers who look at the heart of matter see the considerable structural changes that occur over each load / discharge.
They see above all, that cycle after cycle the matter does not return exactly in its initial state.
So, for scientists, the degradation is "fast".

When I asked them what they thought about NiCd, they were even more pessimistic about the longevity of this electrochemical couple ...
I replied that this mediocre longevity seemed satisfactory to us because we drove daily with for years on second-hand vehicles very badly used by their first users and which allowed us a suitable use.

The batteries of my 106 are 2000, I use it daily for more than 4 years and more than 45.000km (the car at more than 70.000km) and they have started to give up the ghost for 6 months, or 12 years of service , this is correct for "unreliable" technology.

Yes, it will cycle the batteries to maintain them at their best, it is not a problem in itself.
Just set it up in the vehicle design as there are battery chargers that have the unloader function before starting a charge.

For many uses, the cycling will occur naturally if the capacity of the batteries corresponds to the distance traveled daily.

If we only use a third or a quarter of the battery capacity on a daily basis, it is no longer an econological use, it means that we have "invested" for 60 to 75% of unnecessary capacity . This is called waste, and it is counterproductive because it increases the cost of purchasing and using the vehicle.

In short, there is no problem, there are only solutions.

[Obamot]

Bein it already wrings the neck definitely to a received idea!

If we only use a third or a quarter of the battery capacity on a daily basis, it is no longer an econological use, it means that we have "invested" for 60 to 75% of unnecessary capacity . This is called waste, and it is counterproductive because it increases the cost of purchasing and using the vehicle.

In short, there is no problem, there are only solutions.

This is called knowing how to carve the share of Li-ion

Is it necessary to deduce that this is valid also for (electric) hybrids?

[Remundo]

Thank you Obamot for this compilation,

it's a lot of things to read ... but interesting.

In reality, we are all in the "empirical" on Lithium batteries. We do things and we wait for the return in a succession of trial / error without really knowing what is going on in the secret of battery electrochemistry.

I use my 8 kWh of LiFePO4 by mixing the 2 methods:
1) about 4 charges / 5 at partial cycling
2) the other load on 5 in complete cycling (maximum discharge / maximum load)

Up to now, I have never noticed a drop in the autonomy of the "hardware" (that is to say that my batteries work like new, apart from a block which gave up the ghost shortly after l 'purchase).

On the other hand, in Li-On, the decreases in autonomy are mainly due to a bad balancing of the batteries ; and there it is the BMS (batteries managing system) that hurts his job.

I am forced periodically (once a month) to disconnect and do the balancing myself: it consists in clipping the battery that taps up the first one.

The clipping takes place by a gentle discharge of 9h in a 60V bulb, in 12V, that draws about 3 A, the 1 hours remove 9 Ah to the imbalance block.

This "trifle" allows other thirsty batteries to booze again and the autonomy can jump by + 25% without problem ...

[I Citro]

Bein it already wrings the neck definitely to a received idea!
... This is called knowing how to carve the share of Li-ion

Is it necessary to deduce that this is valid also for (electric) hybrids?

Yes, even though most hybrids use Nimh chemistry (only the Prius + uses Lithium for its hybrid batteries and the Prius PHEV for its traction batteries).
In the case of hybrids, battery management already incorporates sophisticated management of charge / discharge cycles.
If some parameters must evolve because of new knowledge on the batteries, they are easily implementable during a maintenance of the vehicle by updating the computer.

[Obamot]

Aaaah, but to know it finally clear, it also changes his appreciation of the merits of choosing a type of battery rather than another.

Something that was not so clear until then! And it will go back to R&D, which will be able to better support the arguments of their decisions with their superiors. And of course, this will now have an impact on the choice of the end user, therefore consumers ...! It’s not nothing.

Now you guys will finally have something to argue about, studies in support! (that lithium etc ....)

[I Citro]

Aaaah, but to know it finally clear, it also changes his appreciation of the merits of choosing a type of battery rather than another.

Yes and no.
We are still in the infancy, we do with what we have and laugh when we no longer use batteries, replaced by super capa or other storage systems ...

What we call memory effect may be called otherwise when we have perfectly identified the phenomenon, it is far from the case.
I was talking this morning with a researcher specializing in LiFePo batteries and BMS that I wanted to put in touch with our friend nlc. He explained to me that to date, we are unable to say if one LifePo element will be more reliable than another one based on measurement tests.
Only an ultra-rigorous manufacturing process can optimize the results. The post-production tests we are trying to set up to build battery packs are just one element to minimize later problems.

This does not preclude prospective organizations (to bait investors) to publish this week that the longevity of Lithium batteries would be greater than the forecast with a average service time between 5 and 20 years... but without mentioning a manufacturer ...

[Obamot]

All the same to the manufacturers, to review the "User manuals", which until then claimed that a Li-ion battery could be recharged at any time - regardless of its level of discharge - and supposedly without any negative consequences on its lifespan!

[Obamot]

On Wiki-thing:

Lead-acid batteries or Lithium-based (like Li-ion batteries) are not sensitive to the memory effect; "we can therefore recharge these batteries without having previously discharged".

http://fr.wikipedia.org/wiki/Effet_m%C3%A9moire

Will you have to correct?

Another interesting hypothesis - and quite different - but which curiously can eventually lead to the same conclusion, here:

The "memory effect" is a phenomenon that characterizes a battery that refuses to deliver all the energy it has in store, because it has been recharged before being completely exhausted and has memorized the threshold reached at the time of the recharge as a depletion threshold.

This phenomenon was initially discovered by NASA. Its satellites, placed in orbit around the earth, charged during the periods of days and discharged during the periods of night. Attention, the durations of charges / discharges remained perfectly identical. After several charges / discharges, NASA noticed that the battery was no longer energizing beyond the point where it was used to being discharged. The battery had kept "in memory" the usual threshold of discharge and refused to go further.

In everyday life, it is almost impossible to suffer the effect of memory on Ni / Cd batteries. Charge and discharge times are rarely (if ever) scrupulously identical over successive cycles. The consumer should seek to reproduce the effect of memory which is a school hypothesis. In fact, even in the laboratory the reproduction of the memory effect is difficult and the average user does not fear anything.

http://www.presence-pc.com/tests/Li-ion ... n-22678/6/

Except that, if we start from this logic ...: by unloading a battery to his "lowest level not harmful"is not a "memory effect" created intentionally at this threshold precisely, which would in fact prevent it from occurring at another level of discharge? Prolonging at the same time (if at all possible) the life of the batteries? And so of all accu? This is my hypothesis.

PS: if NASA said ... ( )

[Remundo]

I took the time to read a bit more about the work of this Novak

Charge and discharge are processes that affect the thermodynamic equilibrium of the battery; but this equilibrium can be restored after a certain lapse of time. The researchers discovered that when the latter was long enough, the memory effect was canceled.
[]
The memory effect would disappear only if we wait long enough after a cycle consisting of a partial charge, followed by a complete discharge. In this case, the two groups of particles are certainly always separated after the complete discharge, but they are all on the same side of the potential barrier. The division then disappears, because the particles tend towards a state of equilibrium, where they all have the same lithium content.

However, the memory effect is maintained after a partial charge and before the incomplete discharge. In this case, the particles are found on both sides of the potential barrier, and this prevents a return to the division of "low lithium" and "lithium-rich" particles.

My choices of use LiFePO4 would be the good ones ...

Full discharge 1 times all 5 charges, long waiting period between battery use, and refills always complete ... Alternating partial cycling and total cycling.

More this goes for my LiFePO4 pack, not for each individual battery of this pack ...

Because in practice, the discharge stops as soon as only one of them "taps down," and the charge stops as soon as only one of them "taps up"; the majority of batteries are never completely discharged or fully charged (even if in the case of a very good balance, it is not far from it).

I personally think that the estimate of the remaining energy in the Lithium batteries by the voltage will never be reliable. It would take much more muscular BMS to balance the Accus and count the Ahs rather than observe the tension of the pack ...

In any case, personally, I have never seen a "measurable" memory effect on my LiFePO4. I especially noticed imperfect refills by unbalancing the pack, which I have to correct manually.