Epoch Made All What We Want

[SOLAR SUPPORT]

It is important to top up LFP batteries quickly using alternator charging. It allows you to save time and fuel. Especially if the primary charging source is the alternator. But why are advanced regulators needed to balance LFP battery cells with the alternator?

You can charge LFP batteries up to 90-95% Soc by using the alternator at full capacity. By controlling the alternator's rotor excitation with a simple thermostat, you should pause its operation to help the overheating alternator cool itself when necessary. You do not need to wait for the battery and BMS manufacturers' model containing a programmable relay for 4s. Safe alternator charging is possible using a simple, cheap, programmable, low capacity 10-20A VSR and thermostat.

Postpone the cell balancing phase of the LFP battery beyond the 90-95%Soc level and occasionally perform cell balancing without using the alternator. There are advanced BMSs that perform precise cell balancing using small currents and an active balancer, or smart active balancers that work separately and work together with the BMS.

At this stage, BMS does not need as much energy as the alternator produces. Therefore, performing cell balancing using alternators and smart regulators results in waste of fuel and time. Instead of cell balancing using the alternator, use a small solar panel that produces 3-5A and a simple charge controller. For LFP batteries containing standard BMS with passive cell balancing feature, effective cell balancing cannot be done with the alternator.

[rslifkin]

Originally Posted by SOLAR SUPPORT

At this stage, BMS does not need as much energy as the alternator produces. Therefore, performing cell balancing using alternators and smart regulators results in waste of fuel and time. Instead of cell balancing using the alternator, use a small solar panel that produces 3-5A and a simple charge controller. For LFP batteries containing standard BMS with passive cell balancing feature, effective cell balancing cannot be done with the alternator.

If you're running the engine just for charging, it's a waste. But if you're going to be motoring beyond the point where the batteries are full anyway, then nothing is wasted.

[MerMike]

Originally Posted by rgleason

Opinion: We need a small Epoch 100a (similar to the Epoch 300ah and 460ah) with Canbus and Victron Communications (using DVCC proocal). Also all of these batteries should be confirmed to work with the Zeus and Wakespeed WS500 regulators, directing their charge routine. This 100ah-120ah battery can we paralleled to create a resilient 300 ah battery that should be adequate for many smaller cruising sailboats.

The Epoch 120AH Dual Purpose has Victron comms. I doubt the integration with Zeus or Wakespeed is any different than for the 460, but that's speculation as I don't have one to test. In the Zeus' case, Arco will implement DVCC, but it's not released yet.

I would like to test the WS500 with the Epoch batteries, but I don't have one to play with.

[MerMike]

Originally Posted by MV Intrigue

The scenario where the Arco Zeus shuts down if it loses the ATC from the Cerbo GX via Epoch 460 might not be desirable. If the battery malfunctioned and it was not allowed to charge and not allowed to discharge and the Arco made use of this signal to also shutdown... then the boat is dark. Also simply putting it on a switch like you did may be a good option.

At the moment the Zeus doesn't participate in DVCC, so there's no impact on the alternator when the ChargeMOS on the Epoch shuts down. The Zeus loops through a soft ramp-bulk-fault condition until there is a power demand, then raises the field current to meet it. DVCC should address this with a little more elegance.

I use the Zeus ATC input on my test bed so I can reliably drop the field on the alternator before shutting off the motor. No reason you couldn't do the same on an installation.

I would love to be able to manipulate the Epoch Charge and Discharge MOSFETS in the app but they're password protected (hint, hint). I would like to see if both MOSFETs were off then would the boat indeed go dark? Would the regulator be able to power itself from the alternator? It might depend on whether it was generating power at the moment. Inquiring minds want to know...

[MV Intrigue]

Originally Posted by MerMike

At the moment the Zeus doesn't participate in DVCC, so there's no impact on the alternator when the ChargeMOS on the Epoch shuts down. The Zeus loops through a soft ramp-bulk-fault condition until there is a power demand, then raises the field current to meet it. DVCC should address this with a little more elegance.

I use the Zeus ATC input on my test bed so I can reliably drop the field on the alternator before shutting off the motor. No reason you couldn't do the same on an installation.

I would love to be able to manipulate the Epoch Charge and Discharge MOSFETS in the app but they're password protected (hint, hint). I would like to see if both MOSFETs were off then would the boat indeed go dark? Would the regulator be able to power itself from the alternator? It might depend on whether it was generating power at the moment. Inquiring minds want to know...

Really we are still in the infancy here. Ill wait till the kinks are worked out. I put the 2 Epoch 460s in place in the boat yesterday. Need to mount the Multiplus2 and Lynx Power in as well as Lynx distributer. Then I can start making cables. For now the MC618 will have to do.

[CaptainRivet]

Originally Posted by rslifkin

If you're running the engine just for charging, it's a waste. But if you're going to be motoring beyond the point where the batteries are full anyway, then nothing is wasted.

Sure, if you turn your alternator off you need less fuel. And you can easly charge up to 14.0V and cut the charge which is approx 98%.
But in real world i always have use for energy to run watermaker, washing machine , dishwasher, dehumifer/aircleaner, breadmaker to burn excess energy. In Reality if i reach 90% i put on the above stuff...
Solar support is right you simply just need bulk charge phase and done with alternator. I do balancing with solar or shorepower charger (seldomly).

[CaptainRivet]

Originally Posted by MV Intrigue

Really we are still in the infancy here. Ill wait till the kinks are worked out. I put the 2 Epoch 460s in place in the boat yesterday. Need to mount the Multiplus2 and Lynx Power in as well as Lynx distributer. Then I can start making cables. For now the MC618 will have to do.

The MC618 will do it all the time, thats already a luxery one and more then you actually need.

[rgleason]

Solar Support wrote:

"For LFP batteries containing standard BMS with passive cell balancing feature, effective cell balancing cannot be done with the alternator.

Once the 300ah LFP are charged to 96%-99% using the 160ah Alternator power managed down to (80-100ah charging).

Just to understand top balancing:

How long will it take 100watts of Solar regulated with Victron MPPT Contollers to complete the balancing, assuming the internal BMS has 2amp active balancing?

How often should the LFP normally be top balanced? Weekly? Monthly? Every season?
My understanding is that for good matched cells and a good BMS the need to do this is reduced.

If your cells are weak, bad or not matched you will certainly want and need to do this more often.

I am listening to the experienced LFP users here.

[SOLAR SUPPORT]

Originally Posted by rgleason

Solar Support wrote:

"For LFP batteries containing standard BMS with passive cell balancing feature, effective cell balancing cannot be done with the alternator.

Once the 300ah LFP are charged to 96%-99% using the 160ah Alternator power managed down to (80-100ah charging).

Just to understand top balancing:

How long will it take 100watts of Solar regulated with Victron MPPT Contollers to complete the balancing, assuming the internal BMS has 2amp active balancing?

How often should the LFP normally be top balanced? Weekly? Monthly? Every season?
My understanding is that for good matched cells and a good BMS the need to do this is reduced.

If your cells are weak, bad or not matched you will certainly want and need to do this more often.

I am listening to the experienced LFP users here.

Let's assume that the cells of an LFP battery that is charged and discharged for the first time are initially in balance. Serial connections between these cells have very small resistance differences. In each discharge and charge cycle, the energy input-output between these cells, which are connected to each other in series, will not be exactly equal. Even though this difference is very small at first, the energy gap between cells tends to accumulate in small amounts with each discharge-charge cycle.

The task of the passive balancer, which comes into play when Soc values ​​rise to 95% and above, is to equalize the cell voltages. Cells with higher voltage than others are discharged at milliampere level using small resistors. This energy is released to the external environment as heat. The active balancer works differently. It transfers the energy it receives from the cells with high voltage to the cell with the lowest voltage through its own capacitor. Available in capacities from 1-5A, active balancers can initiate cell balancing much faster than passive balancers and above 3.3Vpc.

During the cell balancing phase, the energy entering and exiting the LFP battery will be low and in a non-fluctuating profile, allowing the BMS to do its job so that the cell voltages can be measured precisely and the necessary correction can be made. The lower the resistance differences between the serial connections between LFP battery cells, the lower the cell balance will be at the same rate if the cells are of exactly identical structure. However, cells are not perfectly identical. Therefore, correcting the intercellular balance problem at certain intervals is important in terms of solving this small problem while it is small.

An LFP battery that has not been balanced for a long time for some reasons and whose balance difference is increasing between its cells may only have a BMS with passive balancing feature. At this stage, the HVC level of the battery BMS tends to decrease gradually. When the balance gap between cells exceeds a certain level, the battery capacity gradually decreases. If the total voltage of the LFP battery cells falls below the passive balancing activation voltage of the BMS, the battery capacity continues to decrease irreversibly. Therefore, the advantage of an LFP battery with an active balancer is that this level is much more flexible and the battery capacity can be recovered by the active cell balancing strategy of the BMS.

An LFP battery with an active balancer does not need to balance the cells every day. A decrease in the HVC level of the LFP battery BMS indicates the need for inter-cell balancing. It is not enough to monitor LFP bms HVC for the cell balance only, but for an alternator with a smart regulator, the charge cut-off voltage initially recorded as constant may remain above HVC. You wish you had a desposible APD connected to your charging infrastructure or had a LA battery connected to dump the alternator's excessive energy created at this moment. If the cell balancing task is ignored and neglected from the beginning, the LFP battery cannot make the number of discharge-charge cycles promised by its manufacturer.

[rgleason]

Thanks Solar, interesting, much of which I know. It is a partial answer.


Just to understand top balancing:

How long will it take 100watts of Solar regulated with Victron MPPT Contollers to complete the balancing, [of the 300ah LFP charged to above 95%] assuming the internal BMS has 2amp active balancing?

How often should the LFP normally be top balanced? Weekly? Monthly? Every season?

[SOLAR SUPPORT]

rgleason,

The answer I wrote was not specific to you, but general. We cannot talk about a standard balancing interval time for LFP batteries.

[CaptainRivet]

Originally Posted by rgleason

Thanks Solar, interesting, much of which I know. It is a partial answer.


Just to understand top balancing:

How long will it take 100watts of Solar regulated with Victron MPPT Contollers to complete the balancing, [of the 300ah LFP charged to above 95%] assuming the internal BMS has 2amp active balancing?

How often should the LFP normally be top balanced? Weekly? Monthly? Every season?

How long it takes to balance a 300AH pack depends how big the devation is and how much current the balancer at this deviation really uses.
The typical 5A Heltec Balancer has around 1A at 100mV deviation. 100mV at 3,55V a 300AH cell is actually around 1-1.5AH means the 5A heltec would need 10h-15h.
My longterm experience here is the following:
Factory resistance Match cells are typically in the 5-10mV range (if you run them 3 or 4p till 15mV)and won't differ much, hitting the balance area once a month is ok and giving them once a season really a top balance means shorepower charger and keep them in the balance area for 12h, most likely not necessary but doesn't hurt either means optional.
Vendor matched cells (cells failed but matched by capacity) are typically in the 10-30mV range, hitting balance area all 2 weeks keeps them in balance. Once a season a real top balance is needed or cells more often in balance area.
Real grade B or C means no matching are 30 till 100mV, balancing area min once a week better twice. And a real top balance once a month (60-100mV area) or all 3 month if 30-60mV.
This when charge/discharged with max 0.3C.
More C-rate the factory match balance don't care as long within spec, the other 2 double balancing need.
This is just an overall summed up experience and can be used as rule of thumb.
I would always get the biggest active balancer the BMS has to offer as even perfect factory cells age and start to differ over time and as higher the balance current is the less time you need to achieve a balance and the weakest cell is less stressed means wears less. And if you have a cell that ages faster a bigger balancer can cope with that.
The areas given are deviation at 3,55V immediatly after charge cut off.

[CaptainRivet]

Originally Posted by rgleason

Thanks Solar, interesting, much of which I know. It is a partial answer.


Just to understand top balancing:

How long will it take 100watts of Solar regulated with Victron MPPT Contollers to complete the balancing, [of the 300ah LFP charged to above 95%] assuming the internal BMS has 2amp active balancing?

How often should the LFP normally be top balanced? Weekly? Monthly? Every season?

How long it takes to balance a 300AH pack depends how big the devation is and how much current the balancer at this deviation really uses.
The typical 5A Heltec Balancer has around 1A at 100mV deviation. 100mV at 3,55V a 300AH cell is actually around 1-1.5AH means the 5A heltec would need 10h-15h.
My longterm experience here is the following:
Factory resistance Match cells are typically in the 5-10mV range (if you run them 3 or 4p till 15mV)and won't differ much, hitting the balance area once a month is ok and giving them once a season really a top balance means shorepower charger and keep them in the balance area for 12h, most likely not necessary but doesn't hurt either means optional.
Vendor matched cells (cells failed but matched by capacity) are typically in the 10-30mV range, hitting balance area all 2 weeks keeps them in balance. Once a season a real top balance is needed or cells more often in balance area.
Real grade B or C means no matching are 30 till 100mV, balancing area min once a week better twice. And a real top balance once a month (60-100mV area) or all 3 month if 30-60mV.
This when charge/discharged with max 0.3C.
More C-rate the factory balance don't care as long within spec, the other 2 double balancing need.
This is just an overall summed up experience.
I would always get the biggest active balancer the BMS has to offer as even perfect factory cells age and start to differ over time and as higher the balance current is the less time you need to achieve a balance and the weakest cell is less stressed means wears less. And if you have a cell that ages faster a bigger balancer can cope with that.
How long a pack of cell stay within factory matched area depends on quality of cells, I know 12 years old winston cycled hard and with up to 2C that are still within 10mV.
Fist gen EVE/Lishen/Calb the same if used below 0.3C, the higher the C the more they start to differ more after around 5 years.
The new gen should be quite close to winston if used below 1.5C.

[rgleason]

Thank you Capt Rivet, this is something I think I can understand, and I am going to try to get into the weeds here. Sorry for all the questions but this is very informative. I just fixed my 30v 30a DC Switching Power supply after having it sit in the shop for 2 years. It was a pretty dumb thing, just had to change the fuse. But this will be useful when I get LFP for seasonal balancing certainly :

How long it takes to balance a 300AH pack depends how big the deviation is and how much current the balancer at this deviation really uses.

The typical 5A Heltec Balancer has around 1A at 100mV deviation [between cells]. 100mV deviation at 3.55V a 300AH, cell is actually around 1-1.5AH [which] means the 5A heltec would need 10 hours -15 hours.

I don't know what a Heltec Balancer is. The Epoch BMS I am considering, generally offers 2a active balancing. The Heltec is about 2.5x's more powerful.

Wouldn't a 2a active balancer with the 300ah battery at 3.55V have a similar 100mV cell deviation and around 1.-1.5ah and that would mean 10-15 hours of charging?

What is the math behind this?

My longterm experience here is the following:

1. Factory resistance Match cells are typically in the 5-10mV range (if you run them 3 or 4p till 15mV) and won't differ much, hitting the balance area once a month is ok and giving them once a season really a top balance means shorepower charger and keep them in the balance area for 12h, most likely not necessary but doesn't hurt either means optional. -What does "3 or 4p" mean?
2. Vendor matched cells (cells failed but matched by capacity) are typically in the 10-30mV range, hitting balance area [every] 2 weeks keeps them in balance. Once a season a real top balance is needed or cells more often in balance area.
3. Real grade B or C means no matching are 30 till 100mV, balancing area min once a week better twice. And a real top balance once a month (60-100mV area) or all 3 month if 30-60mV.

This is when they are charge/discharged with max 0.3C.

• If more C-rate the factory balance cells don't care as long within spec,
• If more C-rate the Vendor matched cellls and Real Grade B or C cells need 2x the balancing.

This is just an overall summed up experience.

What does "3 or 4p" mean?
What does "...are 30 until 100mV" mean?

I was hoping to charge via alternator at .3C most of the time and up to 95-95% when needed. 300ah battery /100a alternator but sometimes charge at a faster rate occasionally when needed.

I was hoping the 100 watt solar which can put out 4-5 amps during the main part of the day (5 hours) would be adequate for balancing, particularly when we are not on the boat and it is at a mooring, however it appears that at least 10-12 hours of charging will be needed. I wonder if that can be done over 4-5 days while to boat is just sitting on the mooring with just the bilge pump occasionally being used (maybe that and the nav lights should be on the starter battery). What do you think about this approach?

What level of cells do you think Epoch Batteries have (I believe they are Evo)? 1,2 or 3 above?

I would always get the biggest active balancer the BMS has to offer as even perfect factory cells age and start to differ over time and as higher the balance current is the less time you need to achieve a balance and the weakest cell is less stressed means wears less. And if you have a cell that ages faster a bigger balancer can cope with that.

How long a pack of cells stay within factory matched area depends on quality of cells,

1. I know 12 years old winston cycled hard and with up to 2C that are still within 10mV.
2. First gen EVE/Lishen/Calb the same if used below 0.3C, the higher the C the more they start to differ more after around 5 years. The new gen should be quite close to wWnston if used below 1.5C

Thank you, this is hugely helpful, as it helps to refine my expectations and LFP Battery maintenance practices.

[CaptainRivet]

Don't overcomplicate things. You don't harm the cells when balancing longer then you need

Heltec Balancers are the most used once, these are the capacitor based balancer means they take current from the highest cell, charge a capacitorn up and distribute it to the other 3 lowest (in 4S or 7 when 8S).
How much they can balance dependens on the devation, the higher it is the more the more they can shift.
Other active balancer like eg the one in the JK BMS with 2A balances the highest cell to the lowest cell with 2A, so it balances with higher current but only one cell to another.
And there is a 3rd one that takes the 3 highest and discharge into lowest.
Question is what epoch has, i assume the heltec capacitor version as most common.
But in the end that really doesn't matter as all these active balancer need to shift that capacity difference around and similar spec means roughly the same time.

The ranges i wrote are the cell deviations at 3.55V end of charge value directly after charger shuts off, so 14.2V. Best is when the bank was really discharge to as low SOC as possible and then fully charge. So discharge to 10%SOC then charge to 100% at 3.55V and note the deviation thats shown by BMS right after charge cutoff.


3p or 4p means cells in parallel, eg i have factory matched 272AH Lishen in 4p4S so 1088AH means the balancer needs to be bigger as 100mV difference would be not 1.5AH but 6AH so 4 times longer to balance.
Having cells parallel means also that devation rises a bit as 4 cells in parallel versus another 4 in parallel is not as close as single cells.
Being factory matched i have 4-6mV devation in the flat area and being discharged to 10%SOC and then charged to 100%SOC the deviation is 12mV at 3.55V just after cutoff. This without active balancer, just the 200mA passive one from Electrodacus. Doing the same with an active balancer 5A Heltec switch on at 95% SOC is 8mV deviation at 3.55V directly after cutoff.
Discharging to 2.5V then reveals 30mV deviation at cutoff means i have a difference in capacity of around 1AH between the cellpacks so eg highest capacity is 1088AH, the lowest pack is then 1087AH.
So i have 1087Ah capacity as the lowest cell defines capacity of the whole bank. Top balance now means the lowest pack is 100% full while the other packs get charged to 99.9% and also having 1087AH.
Having 12mV means all is perfect as this is basically identical and you always have very small differences between the cells that a balancer can balance but thats then fake balancing as the next cycle that balance is gone.
I pull sometimes 450A whi is 0.45C of the bank so after a month rarely in the area where balancing happens i am at around 15 to 20mV devation as with that load the resistance is slightly different which discharges the cells differently.
Lets take 20mV to 12mV natural deviation so i have an imbalance of 8mV means around 0.2AH that the active balancer needs to shift. At that deviation the 5A heltec does 30mA means its 6.5h that the balancer needs to shift around to get back to the natural 12mV deviation.
I configured my balancer/BMS so active balancing is on above 95%SOC so when i maybe hit this 95% for a short time all 2 weeks its only 2mV or maybe 0.05AH and the balancer needs 30min to balance that so after a month i still have my 12mV deviation at cutoff.

Now take unmatched cells with 70mV deviation...you need to hit balance area every 2nd day that these don't get quickly out of balance as even the active balancer have a limited capacity they can shift.