Lithium Lipo + LifePo4 batteries in parallel

[Christophe]

I would also take some photos with an IR camera: we can see that when charged (4A therefore not even 1/2 C) the elements do not have the same temperatures at all, signs that differential wear (IR precision at 0,1, XNUMX ° C).

These LifePo4 elements were not bought new but used, I don't know their number of cycles. I think they are between 5 and 10 years old. The guy had provided me with elements in RAB which are HS (0.0V). Good for the bucket or the grinder to see how was done (if that interests you?)?

[izentrop]

Hello,

these are LifePo4 10Ah elements in 32650

The datasheets found correspond to a capacity of 5 Ah http://www.devon-company.com/spec/32650 ... 20Spec.pdf https://www.batteryspace.com/prod-specs ... 20Spec.pdf

The same curves as given above and an internal resistance per element ≤ 20 mΩ.

In this document is clearly explained the role of the BMS https://www.victronenergy.fr/upload/doc ... 200-FR.pdf

Basically, no more current flows through a fully charged LiFe cell. The BMS will divert the current to finish charging those which, connected in series, are not yet connected, while protecting against overloads.

It also explains that the fact of putting it in parallel with the lipo, it will not take any more current and will start to discharge only after the lipo is it completely (under 3.2 V by element), unless your circuit is cut before that this does not happen.

They are now replaced by 32700 http://budgetlightforum.com/node/65598

[Christophe]

The container does not have a unique capacity!

It's like AA LR6 rechargeable batteries, there are 800 mAh, 1200, 1500, 2300 ...
ditto for lithium 18650 ...

Mine are indeed 10Ah ... I will make you a photo if you want ...

Please note that charging is done in series with my BMS. the balance wires do not allow a current of several amps. The balance is done in 50 mA with this BMS

[Christophe]

Mea Culpa, these are 38120 (I was not at home when I posted the photo!)

Manufacturer reference: B38120 LID 05

Unfortunately no redundancy on Google: "No document matches the specified search terms (B38120 LID05)."

[izentrop]

Please note that charging is done in series with my BMS. the balance wires do not allow a current of several amps. The balance is done in 50 mA with this BMS

It's enough. The difference is often small if it is caught up with each refill. Lead-acid batteries are balanced by a current of C / 30 at the end of the charge, which results in a slight release of gas, not possible for lithium, I believe.

[Christophe]

So there are rare BMS with strong balance current (up to 5A), the other BMS have a balance current between 50 and 250 mA ... whatever the capacity of the battery on which it is mounted . There is still a slight correspondence since a BMS is given for a maximum current. in my case 50A ... which therefore corresponds to a certain range of battery: we will not mount a 1 Ah battery on a BMS of 50A ...

With the Lifepo4 battery I therefore have a balance current of 50/10 = C / 000 ... it therefore takes a maximum of 200 hours to balance 200 completely unbalanced cells.

I think that a slightly stronger current would be more interesting all the same, because if you want to do 1 cycle per day, the balance does not have time to be done. A balance current at C / 10 would be more efficient AMHA (for electro mobility)

The balance can be static (all the time except during charging) or only during charging (not very useful given the weak balance currents). 99% of BMS are configured as static.

On the Lipo I have a BMS with bluethooth module, here is what it looks like, it is also open now for the annual review

Lipo_BMS_Bluetooth.JPG (287.85 KB) Viewed 3656 times

No no no it's not a bomb! The 4 black wires are temperature sensors (3 are placed between the Lipo elements, 1 on the BMS).

There are 20 12Ah lipos in series. Nominal voltage 20 * 3.7 = 74V. Theoretical capacity *: 74 * 12 = 888 Wh. Weight all inclusive: 5,3 kg. Actual mass capacity: 888 / 5,3 = 170 Wh / kg.

The battery uses cells which date from the end of 2016, 3 years already, I used it widely to make propeller engine tests on my paramotor project.

It took a continuous discharge current (= more than 10 seconds in a row) of 140A so 140/12 = 11,7C.

While it took 2 boxes to make a Lifepo4 of 768 Wh, it only takes one (and not even filled) to make a Lipo of 888 Wh.

The advantage of Lipo is also by weight! LifePo4 = 9,8 kg ... actual mass capacity = 80 Wh / kg, half the Lipo.

At the BMS Bluetooth XiaoXiang level, here are 2 screenshots:

BMS_Lipo.png (91.8 KB) Viewed 3656 times

Balancing is frankly not bad for a Lipo who is 3 years old! Note that it can be done to 0.001V near! (configurable value). In approximation at 0,01V all the elements would have been at 3,82V

BMS_courbe_lipo.png (96.52 KB) Viewed 3656 times

You have seen: we have current / voltage curves for each cell!

The only criticism that can be made of this BMS is that the module does not have its own memory: the curves are available only when the phone is connected. Obviously it is not 24/24!

A small memory of a few KB and a few thousand measurements would have been welcome with a download and reset on each connection ...

Otherwise wiring a BMS is really quite painful but good if you want durable batteries ... no choice!

* the mass capacities are theoretical and if obtained it is with low discharge currents (<1C), in practice I noticed in my current ranges (peaks at 50A max therefore) that we can remove a good 20 % in both cases ...

[Christophe]

if it is caught on each recharge.

Yes if ... and if not a stronger current BMS could resolve in a few hours ... something not possible with low currents if the charge / discharge cycles are fast ... and especially deep!

It all depends on the application of the battery!