I expect your Seplos kit came with some guidance as regards maximum charge /discharge rates. (You might find a figure in the region of 5kW’s mentioned - this would be similar to that of my Tesla Powerwall charge rate). Now, if you wish to charge 29 kWh’s of battery from the grid in 7 hours, this suggests you will need a system capable of handing at least 4 kW’s. (4 x 7 kWh’s will provide ~ a full charge. You will need to check your Seplos data to ensure the batteries are capable of taken such a charge though. Do you have or are you planning on using solar PV to augment the system as this would need to be taken into account as regards to the type of inverter you use. If it is just to charge from the grid, then the choice may be simpler. Do you have ‘Islanding’ so that you can continue to use the battery power during a power cut? Presumably, you have some means of switching to battery power whilst using your Seplos kit? If you haven’t already done so, you will need to apply to your DNO to connect your kit to the grid by filling in a G98 or G99 application (Usually handled by your installer, but if you are self installing, then that is down to you.) Regards, Toodles.

Hi @homonid and welcome to this forum.

I have extensive experience of DIY build batteries and charging regimes.
Perhaps you could start by telling us if you've already bought inverters, cabling and any trips or bus-bars to connect up these Seplos units?

I note the comments from @toodles about G99 approval from your DNO.
That is required if your system has the capability to export to the grid.

If you only charge the batteries from the grid, then no consent is needed.

The charge rate for your Seplos units is limited by the BMS electronics it has inside.

We need to know which model you have.
Most Seplos BMS units have a maximum charge/discharge current of 200A,
but others are restricted to 100A when charging, and 200A when discharging.

100A at a nominal 50v comes to 5000 watts, which is a good match for the most common domestic inverters (5kW).

If your two Seplos battery units were connected in parallel to a single 5kW inverter, then it would take less than 6 hours to fully charge them from the grid.
There are 'system losses' in the inverter, which is why I'm not giving an exact figure.

The inverter would need to be supplied from a dedicated 32A-rated MCB or RCBO in your consumer unit because it would be drawing over 20A at 240v AC.
MCB is Miniature Circuit Breaker;
RCBO is Residual Current Breaker with Over-current protection.

Click on the image to enlarge it.
The diagram may not reflect what you're attempting to do.

It's intended as a starting point from which your system design can be discussed.

Thanks for the info. As yet no equipment has been purchased, things are only at the planning stage. I hope to add solar later so will go for a hybrid inverter, and do want islanding as I live in a rural location that can be prone to power cuts. I plan to get an electrician to do the final connection and commissioning of the system and to complete the G98/99 forms.
I've found a youtube channel where the chap has set up a battery system identical to what I have in mind so I might just follow in his footsteps.
https://www.youtube.com/@GaryMeatsLife

@transparent Thanks for your feedback. I have read many of your posts in this forum and copied various facts and figures to my notes, they have been very helpful in building an overall understanding of battery systems.

At present all is at the planning stage so nothing is set in stone.

One thing that confuses me is the C rating. I have read that lithium batteries have a C rating of 1 but the ones I have been looking at on the Fogstar website for building their Seplos battery pack, EVE 314Ah LiFePO4 (MB31), have a C rating of 0.5 for charging. Does that mean I have to double the charging time or the charging wattage?

Thanks again.

I've had a quick look at the videos published on that YouTube link @hominid

The presenter (in Ireland) appears to be using a Solis inverter and in a grid-tied configuration.
Because the inverter has export capability it will be G98 certified.
It is most likely in GB that the DNO will require a maximum inverter output of 3.6kW, which equates to 16A from a single-phase domestic feed.

That strategy will limit what you can do in future.
For example, you can't just add more inverters in order to have sufficient output to run your home.

Can I also suggest that we avoid using the term 'islanding'!

There is a term 'anti-islanding' within the requirements for G98 certification.
That means your inverter must cease its 240v main output in the event of a power outage.

Your house may not operate as an island... continuing to provide you with power.

So the term 'islanding' suggests a breach of that G98 concept.

The better terminology is 'off-grid operation'.

That allows the inverter to supply appliances with 240v AC using its own internal 50Hz oscillator.

There is currently only one company who have an inverter that is G99 approved for both grid-tied operation (with export) and off-grid operation.
That is SunSynk, not Solis.

You can run multiple SunSynk inverters in parallel, whilst the overall system abides by the anti-islanding regulations for export.
That strategy allows you the flexibility for future expansion.

Here's the wiring diagram for three inverters operating in parallel, taken from SunSynk's installation manual:

Finally, please keep asking the questions...

... You're heading in the right direction  😀 

Posted by: @homonid

↑

I have read that lithium batteries have a C rating of 1 but the ones I have been looking at on the Fogstar website for building their Seplos battery pack, EVE 314Ah LiFePO4 (MB31), have a C rating of 0.5 for charging. Does that mean I have to double the charging time or the charging wattage?

It's highly unlikely that anyone using LiFePO4 battery chemistry for domestic storage would ever want to be charging or discharging the cells at 'C'

The 14kW Seplos units you're considering will most likely be built using 280Ah cells.
The 'C' rating would therefore be 280A.
To facilitate 280A of current, the battery would require connecting with a copper cable of 50mm² at a cost of £15 per metre length.

The same 50mm² cross-sectional area would also apply to the busbars between cells of course...

... and you'd need an inverter specified at 15kW to deliver that 280A 🤨 
(280A x 52v ish)

I'm unsure why Fogstar have used the term "0.5C", but we can be pretty sure that the current limitation will be that specified by the choice of BMS, rather than the capacity of the cells.

@transparent Thanks for the feedback. I am confused about the time component for charging. 280A x 52 = 14,560w, the capacity of the battery pack, so would the 15kw inverter charge the pack on one hour? Would a 7.5kw inverter charge the pack in 2 hours?

@transparent Wow! So much useful information. I must go away and digest this all. I must also contact my DNO (Electricity North West) and see what their requirements are concerning the addition of batteries and inverters to the grid. My rural location might also limit connection capacity.

Posted by: @homonid

↑

would the 15kw inverter charge the pack on one hour? Would a 7.5kw inverter charge the pack in 2 hours?

Erm... yes, but that's bad practice.

As GB moves towards Net Zero, the topology of the grid is being adapted.
We can't all demand electricity when we want it, because the cost of upgrading the grid infrastructure is more than we can afford.

As you are in a rural location, it is the 11kV distribution grid which is most likely the main constraint.

The 11kV network can deliver all the electricity we require, but only if we time-slice its use.
We need to spread our demand thinly across as many half-hour usage-periods as possible.

When I hear you suggesting a fast-charge of a battery pack in one hour, it sets the alarm bells ringing in my mind.
That exactly what not to attempt!

You need to look forward to the time when you will also have access to input from solar panels.
Your 1st priority is to charge the batteries from solar.
When that solar input provides too little energy for the predicted consumption on the following day, you can pick up the deficit from the grid in the overnight cheap-rate slot.

But you should completely avoid using the grid during times of peak-demand, and spread your charging over as long a time as possible (subject to the terms of your tariff).

Everything you do must be 'grid friendly'.

I think I understand C rates now. The C rate is the rate at which a battery is fully charged or discharged in one hour. So

• a battery with a C rate of 1 will can discharge or charge fully within one hour.
• a battery with a C rate of 0.5 can discharge or charge fully within two hours.

So the C rate is only relevant if I am pushing the inverter/battery set up to it's limits. If, as you have advised, chargeing is set to match grid capabilities, for example charging over the 7 hour cheap rate period of economy 7 the system will not approach the constraints of the C rate, so it is not relevant.

Is that correct?

Thanks for your advice.

Yes, that's correct @homonid

The C rate applies to the amount of current which you pass through a cell (or a bunch of cells linked together as a battery).

That's why I was puzzled when you mentioned earlier that a fully-assembled battery with an integral BMS was being referred to as 0.5 C
If the limiting factor is the BMS, then it's not entirely correct to be using terminology related to the charge/discharge rate as a function of C

When the BMS itself is the item which constrains the charge-rate, then it would be usual to simply state the current in Amps.

Stick to Amps in this forum, and then everyone else will be able to follow what we're discussing.