DIY storage battery from BEV cells

[nowty]

But you did not mention which Octopus Intelligent tariff you have so that could change economics. What are you on?

Just to clarify many of us including Joeboy are on Intelligent Octopus Go (IOG) which gives 7.5p / kWh for at least 6 hours a day.

It was "Intelligent Octopus" (IO) until about a couple of months ago and then the name was almost secretly changed to Intelligent Octopus Go (IOG).

[Stinsy]

Thank you so much for your kind feedback, this is really a strong encouragement for me to charge ahead. I have some experience in smaller scale solar, battery and electronics projects but the scale I am looking at now will come with new challenges. So let me share some more detail and I would appreciate the scrutiny and pointing out gaps in my plan would be welcome.

The inverter - Sunsynk ECCO 8.8KW - it can output up to 8000W from battery to grid and the same into the battery (up to 190A). So over 3 hour window it should be able to pump out 24KWh. The same goes for filling up with cheap power if forecast estimates a murky day ahead.

The cells - KIA NIRO EV 64KWh, there is an ebay advert for £3800 but I can get it for £3300 delivered. I only had bad luck with battery cells from China, I found 95% of time they lie about specs and quality. 200Ah cells I bought previously turned out to be 140-150Ah with quite high internal resistance. I suspect they are just old cells slipped into a new case and sold as new. I just don't feel lucky enough to try the same again. With second hand EV cells I know they were made to last and must have high C rating. Cycling 32KWh battery made from these cells at my anticipated 8KW rate is going to be a walk in the park.

BMS for batteries - Undecided and looking for advice. I am looking at some on Fogstar website. I imagine any programmable, common port, 16S, 100A, RS485/CAN with Pylon emulation. But actually just as a layer of protection, thinking to set Inverter treat battery as dumb and set manual voltage limits. BMS still can do its own thing just not talk to inverter. Hook them up in parallel like in my crude schematic.

You really don't need 3x BMSs. Your inverter can handle the charge-discharge Voltages and you need a single cheap balancer. Many of the £50 ones can shift 2A which is LOADS (the BMSs fitted to BEVs shift minuscule amounts).

[diyfan]

Hi @Stinsy,
I don't think I follow you at all. Do you mean I should build a single 14S20P 51V/1200Ah battery and use it with just a balancer in place of BMS? I must be misunderstanding you because it would be very unsafe to operate such battery Would you be able to elaborate on your thoughts?


Hi @Nowty
Ah, understood. Intelligent Octopus Go, very cheap night charge rate, that's great. But I cannot see export rates online. Do they allow energy export and pay a good price for it? Could you please share what the rates are?


Hi @Marcus
You mentioned G98,99,100 in your response. Do you happen to know the path needed to be taken in order to have a slightly more powerful inverter? The existing DNO permit is for 8kw solar generator. And the new inverter is technically capable outputting up to 8.8KW (solar + battery output). Would a new complete application be needed or is there an "upgrade" path? The regs is my weakest point in this project and the electrician I am getting to help me with this, I don't think he is familiar with regs for systems above the 4KW, so I could really use the guidance. Many thanks!

[nowty]

Hi @Nowty
Ah, understood. Intelligent Octopus Go, very cheap night charge rate, that's great. But I cannot see export rates online. Do they allow energy export and pay a good price for it? Could you please share what the rates are?

Fixed export (any time day or night) at 15p / kWh which I'm on, or Agile where you can get typically 20p / kWh during a few peak hours but much less than fixed most other times.

[Marcus]

Hi @Marcus
You mentioned G98,99,100 in your response. Do you happen to know the path needed to be taken in order to have a slightly more powerful inverter? The existing DNO permit is for 8kw solar generator. And the new inverter is technically capable outputting up to 8.8KW (solar + battery output). Would a new complete application be needed or is there an "upgrade" path? The regs is my weakest point in this project and the electrician I am getting to help me with this, I don't think he is familiar with regs for systems above the 4KW, so I could really use the guidance. Many thanks!

Ha! That'll teach me to stick my neck out
Actually I've not done it myself yet, but I think there are people on here who have. But my thought was that you can get inverters that will limit total export to an amount you specify, and if your already permitted 8kw then i would suppose you simply fill in the appropriate form with the new inverters details (can't remember which of g99 /g100 is relevant), and they ought be happy. I'd've though that was the simplest option.

If you want to be able to export 8.8 then yes you would have to apply for the increase. And they may say no, or ask for money to upgrade their equipment.

[robl]

@diyfan
Is your existing PV system a fit one ? If so, I recommend that you leave it well alone, and add this battery system completely independent to it - otherwise it gets very complex.
You’ll have to make a G99 application I think, most probably to be successful this must be with a G100 compliant inverter - that 8.8kW sunsynk is, i think - check the exact model on the ena database. G100 means it can monitor the overall home power, and ensure you don’t export over a given power, to be agreed with the dno. The limit is set by your installer I think, in a way that you can’t override. It’s likely they would accept 8kW max overall export - as you already have that!

[Stinsy]

Hi @Stinsy,
I don't think I follow you at all. Do you mean I should build a single 14S20P 51V/1200Ah battery and use it with just a balancer in place of BMS? I must be misunderstanding you because it would be very unsafe to operate such battery Would you be able to elaborate on your thoughts?

Why do you imagine that to be “unsafe”?

[Colin Deng]

Welcome to the forum
Some factory also using this kind of cells to build the battery pack
But not suggest that , cuase it can not replace easily when some cells go south
Also , the cycle life is less than the prismatic cells

For your information

[diyfan]

Hi @Stinsy,
I don't think I follow you at all. Do you mean I should build a single 14S20P 51V/1200Ah battery and use it with just a balancer in place of BMS? I must be misunderstanding you because it would be very unsafe to operate such battery Would you be able to elaborate on your thoughts?

Why do you imagine that to be “unsafe”?

Hi @Stinsy,

You surely must be testing my knowledge ! I am competent, I promise but okay.

Nothing lasts forever and every set up fails sooner or later (probably sooner for me as I plan to use second hand cells). And when that happens for my setup I want it to fail safely (or even have an advance warning so I know to look and prevent failure). As the BMS monitors overall battery temperature, voltages and currents, as well as each cell (or P block) individually, should it measure any abnormalities it will immediately isolate the battery from the circuit, thereby adding a layer of safety. The same level of monitoring cannot be achieved through the voltage limits on the inverter alone. Without BMS, such capacity battery has a potential to fail dramatically, without a warning.
Let's examine several modes of failure and potential for damage
1. Inverter fails to stop charging the battery past cut off voltage = severe overcharging of all cells, potentially resulting in a big crater where my house used to be.
2. One of the cells degrades much quicker than the rest = severe overcharing of a single P-block of cells (20 cells!), as above but hopefully less dramatic.
3. Inverter fails to stop discharging the battery past cut off voltage = all cells over discharged, entire battery pack ruined
4. One of the cells develops internal short = the charge from all parallel cells would discharge instantly, fire hazard due to amount of energy released, likelihood of physical damage to neighbour cells and risk of thermal runaway.

Having BMS would prevent first 3. And having fewer parallel cells in a pack would lower the potential bang for the 4). And with no parallel cells there is almost no risk if one of the cell shorts out. BMS could also have alerting capability, such as a buzzer or perhaps even cloud connection to warn of abnormal condition remotely.
I cannot think how balancer could prevent, mitigate or alert about any of the above situations. Once the battery is balanced, the job is done and there is hardly any benefit to keep the balancer plugged in. I may be missing something. Please let me know your thoughts.

[diyfan]

Welcome to the forum
Some factory also using this kind of cells to build the battery pack
But not suggest that , cuase it can not replace easily when some cells go south
Also , the cycle life is less than the prismatic cells

For your information

Hi Colin,

Thank you. Is there a web site I could go check out pricing for new cells at a budget price? I am after about 30kwh in total. I have seen your post about moving on from your old job. Congrats, and hope you settled in well!