Jump to content

Lithium Battery Best Practice


spark88
 Share

Recommended Posts

Calling on all the battery guru's out there. Does anyone have a best practice for Lithium Battery (I have Dyness) .... specifically if used with Victron.

Things I'm thinking of:

- What is the most damaging factor for your battery (would it be temperature ? If so what is a safe range ?)

- Would it be better to drain the battery slowly over a longer time than high load over a short time ?

- Would it be better to slowly charge it, vs pumping as much as you have into it ?

- Recommended DOD ?

- Should you charge the batteries to full 100 SOC at least X times a month ? Or is it ok to leave it at say 80%

 

Just some of my thoughts ... please feel free to add others.

Link to comment
Share on other sites

20 minutes ago, spark88 said:

- What is the most damaging factor for your battery (would it be temperature ? If so what is a safe range ?)

Most damaging is discharging a cell below 2.8V or charging a cell above 4.2V. That instantly destroys that cell. But your BMS will guard against that.

Second is charging too hard at cold temperatures (below 1°C, but that depends on battery specs, even better to keep it above 5°C). Again, the BMS should guard against that.

Third is running the batteries at high temperatures (due to hard charging/discharging). Most smart batteries will reduce charge/discharge power when they exceed 40°C. It also follows that it is better to store them in a room that remains under 40°C, for that same reason. This does not destroy them, but does cause them to degrade faster.

Fourth, I would say, is discharging too deep. Below 10% SoC reduces cycle life significantly. Pylontech (and I suspect Dyness by extension) will stop discharge at 10% SOC in order to reach the advertised cycle life (6000?).

Fifth, keeping the batteries fully charged all the time at a very high voltage. If you cycle the batteries daily, it is fine to charge them to 53.2V as the BMS requests. If you keep them permanently charged, 52.5V is better. Assuming a 15-series setup of course. Victron systems use a lower voltage for Pylontech already (but probably not for Dyness... I don't know what it does for that one). This is probably the least damaging factor and not one to worry about too much. But it does reduce the life a little bit, just like it does with all Lithium chemistries.

29 minutes ago, spark88 said:

- Would it be better to drain the battery slowly over a longer time than high load over a short time ?

Generally speaking (across all battery chemistries) slower discharge is better because it is more efficient. For lead acid batteries this is known as the Peukert effect. The harder they work, the more energy is lost in the process. BUT, with that said, Lithium batteries have almost no Peukert effect, to the extent that it is not even considered when designing. The calculated peukert constant is somewhere between 1 (ideal) and 1.05, whereas for lead acid batteries it's 1.1 to 1.2. That is to say, lithium batteries show almost no "fade" when you discharge them hard.

However... they heat up. And they don't like to get too hot. For this reason there will be healthy limits to how hard you can hit them and for how long. Generally you can discharge at C/2 for long periods with no adverse effects, but it is definitely better to remain below C/2.

The same applies to charging. You can charge at 1C or even 2C, but the battery gets hot and that can be detrimental, so generally stay below C/2.

38 minutes ago, spark88 said:

- Recommended DOD ?

80%. That's an opinion of course.

39 minutes ago, spark88 said:

- Should you charge the batteries to full 100 SOC at least X times a month ? Or is it ok to leave it at say 80%

Depends on how the battery does balancing. Most batteries does passive balancing at the top, so they have to be charged 100% regularly. At least twice a month, preferably weekly.

If the battery has active balancing and can also balance in the middle/bottom (this is rare... it is not easy to do, and only really works at low power levels or when the battery is idle), then it should not matter if you leave it at low levels for long periods of time. I know very few batteries that have active balancers. The only one I've seen in person was the Discover AES (which is a very expensive battery).

Finally... don't go outside the warranty parameters 🙂

 

Link to comment
Share on other sites

  • 1 month later...

I'm not familiar with the complete off the selve units, just worked with with old bare 18650's.  Between google and what I've experience, here is my perception of the best practices for lithium batteries.

As you might know, the battery (cell) is made from two plates and between them is the electrolyte, a graphene layer with a chemical composition - google is full of info!  Why does a cell getting weaker, deteriorate, what is it in the cell that fails? The electrolyte is the weak link. These cells got a nominal (average) voltage of 3.7v. This implies that 4.2v is max and 3.2v is min.

To me, the best practice is not to stress the electrolyte. If the cell charge can be kept 3.7v it will last ....forever I think!  So, the trick is to have the cycles as shallow as possible around 3.7v to ensure the minimum "stress" on the electrolyte!!  My cells get charged every day to 3.9v max and through the night (and 6kW later) they are discharged to 3.7v!  Yes, based on my argument for my consumption I should work between 3.8 and 3.6.  However, one have to be practical as well, I don't wanna wake up in the morning, cells on 3.6, its overcast and there is a power outage, I "save" the below 3.7v for true backup! (And yes my BMS is set to start balancing from 3.6v...)

There is enough proof in Google, new cells that work in full cycles (4.2 - 3.2), will at best make 600 cycles before it deteriorates to around 60-70% (which is seen as end-of-life)!!  Cycles between 3.9v (70%SOC) and 3.5v (70%DOD) will yield 6000+ cycles.

If I would bought these complete lithium packs, I will forsure want to know the number of series in it to be able to know the info per cell...!!

But @plonkster have good advice - stay in the warranty parameters!

Link to comment
Share on other sites

On 2020/04/18 at 12:28 PM, plonkster said:

Most damaging is discharging a cell below 2.8V or charging a cell above 4.2V. That instantly destroys that cell. But your BMS will guard against that.

Finally... don't go outside the warranty parameters 🙂

 

The 4.2V maximum voltage is accepted. The low voltage needs clarity for me. I've come across 3.0V  but have also seen 2.5V...

Has anyone got the warranty specs? My suspicion is that with different manufactures it varies judging by the performance of my collection of different 18650 cells..

Link to comment
Share on other sites

There are basic two types of lithium batteries, the li-ion and lifepo4 batteries and both are available in 18650 shape. Lifepo4 cells work at 3.2v noninal, that implies that their work range is between 3.7v and 2.7v... To charge lifepo4 cells to 4.2v is like charging li-ion cells to 4.7v - a practice that will garentee premature failure!

The low valtage limit from a practical perspective and why I guard against it...  Assume a li-ion pack of 14 batteries in series and 10 cells in parallel per battery fully charged, ie 4.2v/cell or 58.8v for the pack. P(watt)=V(volts)×I(amps). Assume a load of 500w, thus 500=58.8xamps =8.5A delivered by 10 parallel cells =0.85A/cell discharge rate...  At 3.2v for the same load the amps will be 500w=(3.2×14)×amps = 11.16A or 1.12A/cell.  This is all fine but do the maths now at 2.8v and you get that 1 cell has to deliver 500=(2.8×14)×amps÷10 = 1.28amps.  This practice will ensure more "stress" on the electrolyte as the C-rate is just getting higher. You will be able to feel the actual temperature increase on the cells!

Link to comment
Share on other sites

In terms of aging from charging and discharging, here is a nice more or less standard charge discharge curve that says it all:

http://www.hecobattery.com/photo/hecobattery/editor/20130708180128_98412.jpg

What you can learn from this is, don't go below 90% DoD or 3,1V and don't go 90% SoC or 3,45. But as mentioned already balancing takes places in higher regions like 3,55 or even 3,6V so take that into account as well. Now I don't know how and at what voltage these dyness batteries balance but it is important to give them a change to do so after each let's say 10 cycles.

And also very important: Do not charge below freezing temparatures at all!

Link to comment
Share on other sites

28 minutes ago, Richard Mackay said:

Many thanks.

Is this for Li-ion cells? (and how can you identify them??)

This is for LiFePO4. How to indentify... Well it's either NCM or LiFePO4. NCM is the most common for 18650 based batteries like laptops, powertools, Tesla cars and powerwalls etc.  LiFePO4 is less common because of their worse weight / capacity ratio. They are mostly seen as prismatic cells but also are there in a lot of other shapes. Identifying, well look at the label I guess... 😉

BTW, here is a nice article about how to get the longest life out of a lithium based battery.

Link to comment
Share on other sites

10 hours ago, RikH said:

NCM

NMC, Nickel Manganese Cobalt. There is also NCA (uses Aluminium, in 2015 Bosch acquired SEEO who is a giant in NCA... but so far not much more news on that front). Both of these are candidates for electrical vehicles as you pointed out, because of the higher energy density. But... LFP needs no Cobalt... and cobalt is the scarce component in the equation 🙂

 

Link to comment
Share on other sites

1 hour ago, plonkster said:

But... LFP needs no Cobalt... and cobalt is the scarce component in the equation

Besides that, the methods of mining cobalt, are to put it mildly, questionable...

Link to comment
Share on other sites

3 hours ago, Richard Mackay said:

Appears to be a Sony battery. I downloaded a doc on it but it still doesn't indicate what type of Li-Ion battery it is.. 

NCM because the nominal voltage is 3,7, low cut off 3,0 and max 4,2. That's typical for NCM.

Link to comment
Share on other sites

2 hours ago, RikH said:

Besides that, the methods of mining cobalt, are to put it mildly, questionable...

Yup. Cobalt is primarily a side-product of copper mining. Copper mining can be particularly destructive (at least, that is what the internet tells us!), but copper is also kinda sorta important for the stuff we do here.

There is no real way (yet?) to get cobalt out of recycled electronics, and we're talking of really small amounts in any case. To be fair, there is also no way to recover lithium, but in the case of lithium we still have oodles left, and the main sources are not hard or destructive to mine. That makes LiFePO4 a lot more friendly.

Link to comment
Share on other sites

On 2020/05/31 at 10:31 AM, Richard Mackay said:

Is this for Li-ion cells? (and how can you identify them??)

 

19 hours ago, Richard Mackay said:

Ok. I looked at the label: SF US18650GR

Appears to be a Sony battery. I downloaded a doc on it but it still doesn't indicate what type of Li-Ion battery it is.. 

 

Assuming there is no obvious marking/info on the cell I am not sure if the question relates to

1) a general differentiation between18650 Li-ion cells on one hand and LiFePO4 on the other or

2) whether it is about differentiating  the detailed exact chemistry for 18650 cells in general

18650_2.jpg.5c70f7c8b89caec4a5fba75da170ed51.jpg

If question option 1....To my mind currently Li-ion 18650 cells are firstly divided into cells with a nominal voltage of 3.6V/3.7V on one side and on the other side of the iron curtain (?) cells with 3.2V nominal (which if nothing else will at least affect whether your exsiting charger will be suitable to charge them). At the risk of having high school science teachers and battery purists choke on their corn flakes if they read this I separate the world into Lithium-ion cells (>3.6V) and Lithium-Iron cells (3.2V)....  So rule of thumb if the voltage is indicated on the cell as 3.2V it is LiFePO4 and any cells stating 3.6/3.7 are "non_LiFePO4". Other than the pink (Samsung) cells in the pic, all cells can be identified as "non-LiFePO4" because of the indicated voltage greater than 3.2V. 

The pink cells require some googeling (binging?) to get a spec sheet since no voltage is indicated and you should find multiple references of it being a 3.7V cell (again "non_liFePO4").

If the purple (?) cells (left most in pic) did not have the handy sticker I would be left ASSUMING they are 3.6/3.7V cells because there are no other info/identifying codes etc. on them and these are from a battery bank (the 3.6/3.7V cells are more common in consumer devices like battery banks and laptop batteries).

I have a very strong suspiscion that question option 2 (exact chemistry) is actually what you are after. The pink cells and the black cells in the pic at least give some clues in the product code. "INR" on the pink cells and "NCR" on the black cells help to a degree to identify the chemistry but only if you like trying to decipher non standardised industry alphabet soup. Have a look here  and here  

In the race to develop better cells, manufacturers use different chemistries in Lithium cells and the main unique chemistry components are then used when refering to different cells. A Nickel Manganese Cobalt (NMC) cell having different characteristics from a Lithium Cobalt Oxide (LCO) cell.....Easy... right? Well it turns out no one can decide what specific parts to single out when referring to the chemistry component. Some manufacturers/suppliers/users/forum gurus will single out different elements of cells consisting of the same base chemistry. So NMC cells can also be NCM, CMN, CNM, MNC, MCN..... remember the alphabet soup...

The pink Samsung cell is identified as INR18650 which is actually Lithium Nickel Manganese Cobalt cell... huh??? Turns out in the (samsung?) battery world the capital i ("I") should indicate "Lithium", "N" should indicate "NMC" as main chemistry and "R" should indicate Round/cylindrical (some sources claim the "R" as "Rechargeable"..). In the three letter "i-codes" the second letter should indicate:

C - LiCo  (ICR) ... also known as LCO or Li-Cobalt
M - LiMn  (IMR) ... also known as LMO or Li-manganese
N - NMC  (INR) ... also known as NMC (and the 128 variations of arranging the 3 letters)
F - LiFe  (IFR) ... also known as LiFePO4 or LFP

The black cell in the pic has a NCR18650 indication... where did the "i" go? Don't know but this is likely a panasonic made/inspired cell with a Lithium Nickel Cobalt Aluminum Oxide cell chemistry (also known as NCA.. a chemistry again different from LCO, NMC, etc.)

Getting to the sony cell it seems that sony do not like to make life easy since they have very little useful info on the cells and also seem to have scrubbed the internet of their official battery info documents. It is easy enough to determine that it is a 3.7V cell (i.e. "non-LiFePO4") but the chemistry seems harder to pin down. At the very least that sony US18650GR cells  appears to have started as a LCO type but the table here suggests that there are also NMC versions. Potentially the NMC versions will be newer cells?

If looking for a fairly comprehensive collection of 18650 info to identify cells have a look here and maybe more specific to your question this part

 

PS. apologies to OP for heading slightly off topic..

Edited by introverter
added off-topic disclaimer
Link to comment
Share on other sites

17 hours ago, introverter said:

Assuming there is no obvious marking/info on the cell I am not sure if the question relates to

1) a general differentiation between18650 Li-ion cells on one hand and LiFePO4 on the other or

2) whether it is about differentiating  the detailed exact chemistry for 18650 cells in general

For myself I'm only interested in the broad picture. The differentiation between LifePO4 and the rest is interesting and clearly the first decision to be made.

I basically need to know how to charge/discharge the battery to check if it's still good.

However that list of all the manufacturers is great to have as one gets to know them better..

 

  

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...