1400 Ah Lithium from cells

[CatNewBee]

Originally Posted by nebster

I've seen several folks very happy with their REC BMS. Personally I prefer to avoid active balancing and have taken advantage of that to set values such that we never enter the knee of the voltage curve. Do you know at what voltage your BMS begins active balancing?

Also, how big is your pack?



How did you determine that 3.0v is 40% SOC? Is that under a load? 3.0v at rest with LFP should be closer to 2% SOC.

Also, just out of curiosity, what is your peak load and how much voltage sag do you get at the inverter inputs, if you've had a chance to observe?


Thanks for sharing some of your system. Maybe you should start a thread so we can be sure to give you your due.

Well I am in the testing phase and play around with the parameters and the loads.

The REC BMS allows you to set every parameter to your needs.

The balancing process is quite complex so - generally speaking there are 3 strategies, that work together to keep your cells happy and you can set some thresholds when they apply and for how long.

The BMS uses an external Shunt to determine the current flow and uses this too for decision making, it measures the internal cell resistance of each cell and the cell voltages, additionally it has 2 external temp sensors which you want to place between cell1 and 2 as well as between 3 and 4, so you notice a temp increase and can react, also the bms has an internal sensor too.

Lets start with top balancing.

Top balancing occurs / starts when charging the pack, the criterea is, there is a current going into the pack over 0.2A, and the START balancing parameter is hit by the highest cell. Active balancing stops, if there is no current in this area. The balancing is done by a step-up converter, that uses the single cell as source and creates additional charging power to the pack, therefore very little energy is turned into heat., the balancing is done by impulses of a second or so, then the pack is measured again etc. Balancing current is most of the time around 1A - (up to 2.5A according to the spec).

This is the first stage.

After a while one or more cells reach the second threshold called END balancing - a little misleading - it does not mean that balancing eds, but that the 2nd stage is invoked and the system performs a balancing regardless of input current, as long as the voltage is above this value. This is used to finalize balancing in float conditions of the charger, where no more current goes into the pack, balancing ends when all cell voltages are within a set range (next parameter)

Finally the pack is full and in balance.

There are 2 more parameters: Charge End voltage - this can be used (by custom programming) to fire up the opto-coupler interface and force all chargers to float or to shut down. Also used to sync SOC reading.

The next one is the maximum cell voltage. If this is hit, the over voltage protection / stop charge relay is fired up and the system is physically disconnected from all charging sources, when the hysteresis is exided downwards and SOC is below 95%, charging relay is re-enabled. This is the safety-switch and will probably newer occur.

Most parameters have a second individual configurable offset called hysteresis. It is used to reset a set condition without frequent flapping of relays etc.

Now to discharging.

Usually there is no balancing if cell difference stays in a given larger balancing range (another parameter, around 250mV). If this is violated, the system starts balancing regardless of the situation and raises an audible alarm.

If everything is normal, the pack discharges to the minimum cell voltage

If this is hit, the deep discharge protection relay (UVP) is disconnected, charging remains enabled, if the other cells have enough power (voltage exceeds threshold), bottom balancing is invoked to top up the low cell. Eventually the hysteresis exceeds and discharging will be enabled again.

By custom programming you can have a separate voltage level and a hysteresis value to fire up the opto-coupler interface earlier to your needs and force disconnect of larger devices at any state, so you can disable inverter at SOC 40% or what ever you like and have enough remaining capacity for other stuff. This is based on Voltage of the lowest cell.

Alternatively you can use the alarm contact of your Victron BMV based on SOC to create a signal based on used Ah or SOC%.

So that's bottom balancing.

Voltages are measured and corrected by the measured internal cell resistance to have more accurate readings.

Another goodies are, you can set thresholds for charging and discharging based on temperatures (forbid charging and discharging at low and high temp - for Winston not relevant range is -45 to 80°C, good settings are -30 to 60), also for the BMS protection from overheat. In the SW UI you can modify 15 params directly and in total 32 standard parameters + programmed custom parameters via command line terminal access (also included in the software), there you can set / reset SOC, cycles count etc.

The best thing is, you can tune the system exactly to your needs.

My Pack is 12V 1000Ah by 4 yellow Winston 1000Ah cells, max discharge current I have tried up to now is 380A, charging 220A.

the 3.0V was a (good) guess, I must verify that at what point the system hits the 3.0V. I am currently testing the discharge, I have drawn now 900Ah and the cells still have 3.16V while discharging with 2000W (dyson heater).

[rom]

Originally Posted by CatNewBee

My Pack is 12V 1000Ah by 4 yellow Winston 1000Ah cells

You seem to have done your homework, but what about the warnings we can read here & there about cells bigger (heavier) than 200Ah for marine (small boats) use ? What's you take on this ?

[john61ct]

Link to the BMS tech docs, and sources please?

[senormechanico]

Originally Posted by kmacdonald

What is your algorithm for switching from absorption to float? What solar controller?

Blue Sky 2512 iX
13.8V no absorb time. Float set to 13.2 (in effect no currrent) 200 aH LiFePo4, 440 watts solar, Spectra, radar, refrigeration freezer with Aerogel, Ultrasonitec etc.
Fixed voltage alternator set to 13.6v. Not needed for charging,
In the PNW, it all works great 5-6 months of the year except when the wildfires got really bad last summer. Solar couldn't see the sun !

As we speak, the boat is at my dock. The only reason there's a shorepower cable attached is for the dehumidifier and ozone generator. Solar vent takes care of itself. No mold. In the Pacific NorthWET.
The solar takes care of the rest. Ultrasonitec is on, but fridge is off.

[nebster]

Originally Posted by CatNewBee

the 3.0V was a (good) guess, I must verify that at what point the system hits the 3.0V. I am currently testing the discharge, I have drawn now 900Ah and the cells still have 3.16V while discharging with 2000W (dyson heater).

So, if you've discharged 90% of your expected capacity and the voltage is 3.16, why do you expect SOC will be 40% at 3.00v?

Separately, at what voltage does the charge-shuttling balance begin? I realize you can configure it, but what will you configure it to?

[nebster]

Originally Posted by rom

You seem to have done your homework, but what about the warnings we can read here & there about cells bigger (heavier) than 200Ah for marine (small boats) use ? What's you take on this ?

I'll share another recent anecdote about large prismatics.

I recently learned of a new 400AH cell that came on the market. When I reach out to the factory to see about sourcing a few for testing, they told me that they'd stopped making them (already) because they couldn't get the yield/reliability up high enough!

It's definitely much harder to secure a lot more anode, cathode, and electrolyte into a larger enclosure. Those materials are pretty flexible (or liquid), and the mechanical stresses must really go up with larger unsupported spans. What I don't know is, can the factory find and filter out all of the weaker cells before shipping? Or do the cells fail under dynamic stress, such that only we customers can be the ones to figure out when a particular unit is faulty?

I certainly wouldn't want to be in the middle of no where on a boat with only four of these cells, because losing one would leave me in trouble. I would want a spare along for the ride.

[john61ct]

Voltage cannot be used to guesstimate SoC within the operational middle of the curve.

Only at the shoulders, where a careful owner rarely ventures.

Use an AH-counter to signal time to replenish, but don't charge until you actually need it, when actively cycling. Otherwise better to leave at low SoC.

Stop charging based on voltage, IMO zero Absorb, especially with low-current solar.

[john61ct]

Originally Posted by nebster

I would want a spare along for the ride.

A bank of 8 or 12 cells handles that nicely.

Actually maintaining a spare, rotating, re-balancing required, IMO too much PITA.

[CatNewBee]

Originally Posted by john61ct

Link to the BMS tech docs, and sources please?

REC d.o.o.
http://www.rec-bms.com/datasheet/UserManual_ABMS.pdf
http://www.rec-bms.com/datasheet/UserManualLCD.pdf
http://www.rec-bms.com/datasheet/Use...terControl.pdf

[CatNewBee]

Originally Posted by rom

You seem to have done your homework, but what about the warnings we can read here & there about cells bigger (heavier) than 200Ah for marine (small boats) use ? What's you take on this ?

while searching for my cat, i saw some liveaboard catamarans (Lagoon 440 and 450 fly bridge, 420), that had 1000Ah cells with 123BMS and 700Ah installs that have successfully crossed some oceans.

I don't think the 1000Ah cells have issues when moved around, as long as they are fixed properly - like any other battery pack. They are 35kg each, so also not heavier than GEL. I am pretty sure, they will work flawlessly for years - otherwise I wouldn't have made the purchase.

Smaller cells op to 400Ah are driving around daily in various RV's with no problems, the road is harder then the ocean btw.

So what is the point?

[CatNewBee]

Originally Posted by nebster

So, if you've discharged 90% of your expected capacity and the voltage is 3.16, why do you expect SOC will be 40% at 3.00v?

Separately, at what voltage does the charge-shuttling balance begin? I realize you can configure it, but what will you configure it to?

I had configured it as stated, however you are right, the measured values talk a different language. But easy fix.

So 1300Ah discharged, stopped at cell power at 3.05V without load, with load 3.0V (way above 2.8 as the lowest limit). I will drop discharge at 3.2V next time for the inverter (its around 20% remaining capacity - re-charged 300Ah and discharged 100Ah, then 3.2V fired up), and I will set the shutdown from 2.90V to more conservative 3.00V, I guess 1300Ah usable energy (over 15kWh) are more than enough.

The upper values will remain as tested during charging, Start balancing at 3.375 (around 13.4...13.5V pack voltage), volume balancing at 3.45V (13.7...13.8V), End of charging set to 3.60 for the highest cell, this will enforce immediately float, max. Voltage set to 3.65 (14.6V), this will disconnect all charging sources - but will not happen, because chargers are custom set to 14.0V.

Charger set to 13.5V float, 14.0V absorption for max. 30min. Repeat all 7 days for 30 min. ( I have to think about it a little more, maybe once a month, but it is irrelevant either, this is only if shore power is connected or the generator is running )

This settings should do the trick, because cells are now almost in balance. (that translates to cell voltage 3.375V in float and 3.500V for 30 minutes to allow the balancer to finalize equalization).

[kmacdonald]

Originally Posted by senormechanico

Blue Sky 2512 iX
13.8V no absorb time. Float set to 13.2 (in effect no currrent) 200 aH LiFePo4, 440 watts solar, Spectra, radar, refrigeration freezer with Aerogel, Ultrasonitec etc.
Fixed voltage alternator set to 13.6v. Not needed for charging,
In the PNW, it all works great 5-6 months of the year except when the wildfires got really bad last summer. Solar couldn't see the sun !

As we speak, the boat is at my dock. The only reason there's a shorepower cable attached is for the dehumidifier and ozone generator. Solar vent takes care of itself. No mold. In the Pacific NorthWET.
The solar takes care of the rest. Ultrasonitec is on, but fridge is off.

So you go to float, or turn off charging as soon as it reaches 13.8V. That's a good setup and nice to know the BS 2512 can do that.
How thick is the Aerogel and is it worth the cost?

[nebster]

Originally Posted by john61ct

A bank of 8 or 12 cells handles that nicely.

Actually maintaining a spare, rotating, re-balancing required, IMO too much PITA.

Absolutely. But a bank of 8 or 12 1000AH is likely impractical for the person who selects a 4s1p 1000AH pack for its simplicity and capacity.

This "carrying a spare" conundrum gets a lot easier with higher voltages, because you have wiggle room. I could easily run at 15s or 17s and be within the tolerances of my inverters, for example. So losing a single cell is not a showstopper.

[senormechanico]

Originally Posted by kmacdonald

So you go to float, or turn off charging as soon as it reaches 13.8V. That's a good setup and nice to know the BS 2512 can do that.
How thick is the Aerogel and is it worth the cost?

I have posted on CF about it.
Do a search for Aerogel. There are several threads.

[nebster]

Originally Posted by CatNewBee

I had configured it as stated, however you are right, the measured values talk a different language. But easy fix.

The actual values will be the same for all identical cell chemistries. At 3.00v open circuit, rested, your cell should be at well under 10% SOC. I'm not quite sure what you are saying above, but hopefully it makes sense to you!

So 1300Ah discharged, stopped at cell power at 3.05V without load, with load 3.0V (way above 2.8 as the lowest limit).

Are you saying that you're observing 1300Ah on a 1000Ah pack? That's pretty awesome and quite surprising. Maybe they are overbuilding them to account for slippage in production?

I will drop discharge at 3.2V next time for the inverter (its around 20% remaining capacity - re-charged 300Ah and discharged 100Ah, then 3.2V fired up), and I will set the shutdown from 2.90V to more conservative 3.00V, I guess 1300Ah usable energy (over 15kWh) are more than enough.

The reason I started asking about your discharge voltages is because my observation is that it is very hard to set a LVD disconnect that is high and, also, be able to apply a large load to an LFP bank that is at a low SOC. The IR creates a large voltage sag, and that fools the LVD cutoff detection -- unless it can compensate for the instantaneous load on-the-fly. I haven't seen a BMS that does that yet and I was curious if the REC one does.

The upper values will remain as tested during charging, Start balancing at 3.375 (around 13.4...13.5V pack voltage), volume balancing at 3.45V (13.7...13.8V), End of charging set to 3.60 for the highest cell, this will enforce immediately float, max. Voltage set to 3.65 (14.6V), this will disconnect all charging sources - but will not happen, because chargers are custom set to 14.0V.

Sounds reasonable, as long as you're okay charging to 100% SOC. The consensus best practice is to stop sooner to promote cell longevity, but you can certainly trade some of that for a bit more capacity.

You should start a thread for your build and keep us posted...