[RE-wrenches] Automatic Dark Start of Depleted Lithium Batteries

[Steve Higgins]

Hello all...

The first issue is that inverter/charger parasitic loads have increased
exponentially in the past 20+ years. When the LBCO cuts out, the inverter
may shut off, but it does not remove itself or any other DC-connected
device from the battery.   These devices still draw a parasitic load. In
the 1990s, the Trace SW would pull about .3 to .4 amps of current from the
battery when connected to it. Today, many manufacturers use cheaper
transformers, and the high-frequency inverters draw a much higher current.
Some of these all-in-one inverters draw 1-2 amps of current from battery
banks, just connected and not even turned on.

What's important here is that the battery voltage is already very low when
you trigger an LBCO shutdown (it's not a disconnect). For a 48-volt system,
this is 44 to 47 volts, depending on where you set the LBCO. When a Lithium
battery is this low, the voltage dropoff is much higher. With a lead
battery, the voltage dropoff is much more linear, but with Lithium
chemistry, this voltage dropoff is more like a cliff.    This is why it's
important for many of these Lithium systems to set the Battery cutouts a
bit higher so people have more time to fix the situation before the BMS
shuts down.   Ideally, the customer should be educated not to
over-discharge the bank, which would help. Many of these customers want
turnkey systems that they don't want to think about but don't want to pay
for it or do the work that is required to maintain it.

Now, if the battery had gone into "Protect" mode and the BMS had shut down,
the battery is outputting very little votlage... the inverter/charger needs
voltage to run.  There used to be a line of inverters in the marine and RV
market that would do what we called "Dead Battery Restart".   This meant
there was a parallel circuit in the power supply so that when you supplied
AC to the input, a secondary power supply bypassed the regular battery
power supply and would power up the inverter and allow the charger to run.
 Most of the inverter manufacturers got rid of this circuitry because it
was not cheap, took up space on the boards, and was just another circuit
that could get damaged with generator/shore power surges.   I don't know of
an inverter today with this dead battery restarting circuit.

With this, you need to be very careful. If the customer has cratered the
battery voltage and drawn down the voltage so low that they have damaged
the cells, jumpstarting the battery can create a charging hazard, and that
could cause the cells to swell internally. If that happens, the battery
will get warmer and warmer under charge, and eventually, you could have a
cell rupture.  This can happen quickly with Li-ion, but with LFP, it's much
harder to create this problem. Usually, in LFP, cells will swell a bit, and
the current interrupter on the individual cell will open up and drop that
string, and you will lose capacity.

Like everything else, it's a race to the bottom on cost; this affects
quality and features... Everyone wants the "Bells & Whistles," but they
don't want to pay for it.

Steve Higgins


On Sat, Feb 22, 2025 at 5:08 AM Jason Szumlanski via RE-wrenches <
re-wrenches at lists.re-wrenches.org> wrote:

> I have been thinking a lot recently about the reasons off-grid systems can
> shut down, and working on strategies to prevent these nuisances that
> require manual intervention.
>
> Ideally, a BMS should never shut down due to low voltage/SOC because a
> properly programmed inverter should reach it's cut off well before the BMS
> decides it needs to protect the battery, especially where there is closed
> loop communication. But let's say that happens, where the BMS does make the
> DC battery output go to zero.
>
> It seems to me like the inverter should be able to start a generator, and
> then signal to the BMS that a charging source is available. But I'm not
> aware of any system that actually does this. The inverter should be able to
> wake up the battery. I can see this being particularly possible where one
> manufacturer is writing the code (I'm thinking Midnite AIO/Powerflo).
>
> Of course, the inverter would have to have power in order to do that, so
> if it's nighttime and there is no PV, the inverter power would need to come
> from somewhere. I have two thoughts. First, someone could manually start
> the generator, waking up the inverter, but they would not have to reset the
> BMS if the inverter told it to wake up. The second way would be for the
> inverter to somehow close the 2-wire start circuit upon inverter shutdown,
> restoring power to the inverter automatically.
>
> If those are not options, an external NO relay powered by the inverter
> output could be added to the 2-wire start circuit, perhaps with a time
> delay to return to the NO position to allow the generator to remain powered
> until the inverter does it's thing and starts charging the batteries.
>
> Anyway, my question is whether any inverter/battery combination out there
> works in a way that the inverter tells the battery there is a charging
> source available to wake up the BMS and reconnect DC power. And if not, why?
>
> Jason Szumlanski
> Florida Solar Design Group
> _______________________________________________
> List sponsored by Redwood Alliance
>
> Pay optional member dues here: http://re-wrenches.org
>
> List Address: RE-wrenches at lists.re-wrenches.org
>
> Change listserver email address & settings:
> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>
> There are two list archives for searching. When one doesn't work, try the
> other:
> https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
> http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>
> List rules & etiquette:
> http://www.re-wrenches.org/etiquette.htm
>
> Check out or update participant bios:
> http://www.members.re-wrenches.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20250222/4ef3a6bf/attachment.htm>