270ah DIY LiFePO4 build

[KompetentKrew]

While I understand you are in the early stages of use for your pack I am a little bemused. If you are going to only charge to something around 50% capacity of your pack your available energy is only 50% of the maximum. Are you saying you designed a pack that has twice the required capacity in order not to need charging close to full capacity?

As I understand it, his current weekend sailing is only using a fraction of the bank's capacity, therefore he's keeping it at half charge to be kind to the batteries.

For longer term cruising he can charge to 100%, use them normally and run them down, enjoying the full capacity.

Good:

• Keeping the batteries around 30% - 70%
• Charging them to 100%, slipping lines and discharging as deep as you want. 100% -> 0% -> 100% -> 20% -> 100% -> 45%

Bad:

• Storing the batteries at 100%
• Staying on shore power for months and keeping them in the 98% - 100% range

 

[Poey50]

As I understand it, his current weekend sailing is only using a fraction of the bank's capacity, therefore he's keeping it at half charge to be kind to the batteries.

For longer term cruising he can charge to 100%, use them normally and run them down, enjoying the full capacity.

Good:

• Keeping the batteries around 30% - 70%
• Charging them to 100%, slipping lines and discharging as deep as you want. 100% -> 0% -> 100% -> 20% % -> 100% -> 45%

Bad:

• Storing the batteries at 100%
• Staying on shore power for months and keeping them in the 98% - 100% range

You got it.

I was reading that lithium batteries in satellites are used around mid-range because of the astonishing longevity this gives. I'm not going to stay at mid-range forever as said but if I did then they might well outlast not just me but even the next generation of Poey50s.

The longevity is, incidentally, also connected to the care in not messing them up. Sub-catastrophic exceeding of safe charging parameters will only shorten their life. Catastrophic breaches of parameters are a different matter and here the number of years also count. The cells don't care at all that everything went well the last 3,000 times you used them, you accrue no credit for that. Getting it wrong once is enough. This is very much in the spirit of the Nordkyn papers. He acknowledges that most systems work most of the time but that 'most of the time' is not enough and during a long life even a low frequency event comes onto the cards

 

[TernVI]

This has come up before. I can stay around mid-range for weekend and holiday use. I also have the option of charging fully for the longer offshore passages and cruising I eventually have in mind. LFP gives choices that are impossible with lead acid.

Likewise, in may use cases, you can use much more than 50% of lead-acid capacity for those infrequent events, without really impacting service life.
It's important to be clear what the design goals actually are.

 

[vas]

is it worth having two (or more) different useage profiles though? I mean wouldn't it be better to have one profile 90%-down to 15% and stick to that?
I understand that these are your first days with lithium, just wondering what/how you plan to use them in the future.
Based on your previous observation, what is the cost of a per-cell BMS vs the ones you have now?
and because I'm slightly confused, can a better BMS charge and balance cells in a dodgy/unbalanced battery or is it a job for another piece of s/w h/w which you have to manually fire up and monitor?

 

[yoda]

As I understand it, his current weekend sailing is only using a fraction of the bank's capacity, therefore he's keeping it at half charge to be kind to the batteries.

For longer term cruising he can charge to 100%, use them normally and run them down, enjoying the full capacity.

Good:

• Keeping the batteries around 30% - 70%
• Charging them to 100%, slipping lines and discharging as deep as you want. 100% -> 0% -> 100% -> 20% -> 100% -> 45%

Bad:

• Storing the batteries at 100%
• Staying on shore power for months and keeping them in the 98% - 100% range

The logic is fine but the reality of managing charge levels like this for most people may not be something they would want to do manually. For me I would need a bank that has a capacity at 70% to meet my needs and would charge to that level automatically at any time from alternator, mains, wind or solar. That does however seem a bit wasteful of 30% of the capacity. Seems to need an 'onboard' or 'away' switch that allows charge levels to be changed easily. The problem however would be what do you do to a 100% charged battery if you decide to leave the boat?

It's interesting stuff however and I wouldn't want anybody think I'm being critical, just observational!

Yoda

 

[Poey50]

is it worth having two (or more) different useage profiles though? I mean wouldn't it be better to have one profile 90%-down to 15% and stick to that?
I understand that these are your first days with lithium, just wondering what/how you plan to use them in the future.
Based on your previous observation, what is the cost of a per-cell BMS vs the ones you have now?
and because I'm slightly confused, can a better BMS charge and balance cells in a dodgy/unbalanced battery or is it a job for another piece of s/w h/w which you have to manually fire up and monitor?

Good questions.

Do bear in mind that my 270ah pack is roughly equivalent to around 500-600 ah of lead acid. So with my weekend usage if I charge to 90% S0C I might end the weekend on 85%. This would be fine if these were lead acid since my solar panels would bring the batteries back to float during the week. But for the LFP to stay at this state of charge gains me nothing and is actually a small disadvantage as mid-range favours longevity.

Future plans includes bluewater cruising. So for this I want a wide usage. Charging to 90 or 95% is fine because the pack won't be held there since loads will come on straight away. I don't have a big solar array but the Balmar alternator can efficiently put 80-90ah back in each hour.

So how to manage this wide variation? Actually it should be fairly easy. If the pack is balanced then I can set the charging limits to stop charging when the pack is around 90%. This will cover the biggest range and also the level of charge needed to balance the cells (more below). But for weekend use every charging source has a manual over-ride. The solar panels are disconnected between controller and panel and the alternator charging is ended by an isolator between external regulator and alternator as described in an earlier post. There is nothing to be gained from having a mid-range profile and impossible to achieve anyway since the voltage curves at mid-range are so flat. Voltage is no guide at all until the upper knee of the curve.

In summary, If I don't intervene then the pack should automatically go to 90% (I may need to tweak the absorption voltage to achieve this), For anything else I can stop charging manually at any point.

Most cell balancing is done by a BMS when the cells exceed 3.4 volts. Passive balancing (which the 123SmartBMS has) simply involves burning off power from the highest cell and allowing the others to catch up. It's not yet clear to me how often this should be done but if the cells go well out of balance then the difference may overwhelm the relatively small capability of the passive balancing and one cell may reach high voltage cut off before the others are charged. Some more expensive BMSs use active balancing which more powerfully can transfer power from a higher cell to a lower one.

Balancing is something that will be clearer in a few months time and, up to then, I will be monitoring the balance between cells.

 

[Poey50]

The logic is fine but the reality of managing charge levels like this for most people may not be something they would want to do manually. For me I would need a bank that has a capacity at 70% to meet my needs and would charge to that level automatically at any time from alternator, mains, wind or solar. That does however seem a bit wasteful of 30% of the capacity. Seems to need an 'onboard' or 'away' switch that allows charge levels to be changed easily. The problem however would be what do you do to a 100% charged battery if you decide to leave the boat?

It's interesting stuff however and I wouldn't want anybody think I'm being critical, just observational!

It is interesting stuff and good to think through these issues. As said in the reply to Vas, if using the boat on a daily basis then I don't think there would be much need for a manual end to the charging. Given the flat nature of the curve it may be difficult to use voltage alone to end charging at 70%, however. Others may know more about that.

The problem with a manual switch is that it relies on memory which is not my strong suit. So, although I'm not planning to use an inverter routinely I have set up a pair of cables ready to attach an inverter to which I can attach our 750 watt fan heater and drain off excess charge in this way.

Would be interesting to hear from other LFP users about how you are managing these issues.

 

[Zing]

I do it by voltage and amp counting and water temperature. If the DHW temp is getting low that is the main reason to run the genset, then I will gauge how much longer than an hour needed for water to run it to bring the battery up to 80% ish. Volts indicates state of charge well enough as a cross check with the amp count to say I’m there, it’s just a different calibration from lead. A phone timer tells me when to stop. I also set the charge voltages low, so very little chance of overcharging. I’ve worked for 8 years with no bms, but will get a GWL very soon. A mistake is inevitable at some point.

 

[Poey50]

I do it by voltage and amp counting and water temperature. If the DHW temp is getting low that is the main reason to run the genset, then I will gauge how much longer than an hour needed for water to run it to bring the battery up to 80% ish. Volts indicates state of charge well enough as a cross check with the amp count to say I’m there, it’s just a different calibration from lead. A phone timer tells me when to stop. I also set the charge voltages low, so very little chance of overcharging. I’ve worked for 8 years with no bms, but will get a GWL very soon. A mistake is inevitable at some point.

Sounds like you know how to work your system.

 

[Poey50]

Based on your previous observation, what is the cost of a per-cell BMS vs the ones you have now?
and because I'm slightly confused, can a better BMS charge and balance cells in a dodgy/unbalanced battery or is it a job for another piece of s/w h/w which you have to manually fire up and monitor?

Apologies - I missed this question. The two other BMSs I considered were the Orion Jnr and the REC Active. I think both of these have more functionality than my 123SmartBMS and can do more to control charging sources directly based on cell voltage rather than pack voltage. They may both have active balancing but I'm not completely sure. However the cost was much higher and the level of documentation that I saw was rather over my head so I didn't explore them seriously. The best place for good advice on BMSs is the Lithium Batteries on a Boat Facebook Group or perhaps the Cruisers Forum which has some specialist LFP threads and knowledgeable members.

 

[LoneHort]

The cells were imported direct from R J Energy in China. This is one of two or three companies that have established a reliable reputation on the large LFP forums, DIY Solar and the Lithium Batteries in Boats Facebook group. Apart from many successful sales recorded I noted two examples where R J Energy had replaced cells including three that had corrosion in the base (one of these in the photo above) despite there being a fair degree of user-error in the installation.

3.2 v 271ah, 3.2 rechargeable battery, 3.2 volt lithium ion battery, prismatic cell

The trader I dealt with, Carl Wu, has excellent English. I mentioned the forums I was a member of in order to harness substantial power to complain widely if things went wrong. These forums have thousands of members.

I got a satisfactory quote for delivered cost of £620 as follows (broken down in US dollars):

USD 620 cells
USD 8 connectors and screws
USD 112 shipping, taxes and import duty

Total USD 740
Plus 5% Paypal USD 37

Shipped cost is USD 777.

35 days later the tracking information via Poland was activated and within a couple of days they arrived, well packed with no damage en route.

If you visit the site don't assume this is a real factory or genuine certificates. I also don't necessarily assume that the cells are grade A or brand new. However, past experience is that the quoted capacity is real and that the cells generally come well matched. Nothing else matters as much as these two factors.

What a difference an island makes! I just got quoted a price of $649USD shipping to Ireland for 8 of the same batteries. I'm sure if I was in the UK it would be 1/2 that!

 

[Poey50]

What a difference an island makes! I just got quoted a price of $649USD shipping to Ireland for 8 of the same batteries. I'm sure if I was in the UK it would be 1/2 that!

That's a lot. Have you tried Xuba Electronics (Amy being the contact person)? Their 280ah cells are now popular. Good to get a quote for full delivered cost ... sometimes cells are priced low and transport is inflated so only delivered cost allows true comparison.

 

[Zing]

What a difference an island makes! I just got quoted a price of $649USD shipping to Ireland for 8 of the same batteries. I'm sure if I was in the UK it would be 1/2 that!

Get them shipped by sea. It’s a fraction of the air freight price. I’ve looked into the air shipment option and it appears to be not permitted. It is possible to get quotes, but most shipping companies refuse to quote because of the rules. I expect the shippers who do ship by air are mis-declaring them. It would be useful if this suspicion could be clarified. According to the stated IATA rules you cannot ship by air.

 

[LoneHort]

Get them shipped by sea. It’s a fraction of the air freight price. I’ve looked into the air shipment option and it appears to be not permitted. It is possible to get quotes, but most shipping companies refuse to quote because of the rules. I expect the shippers who do ship by air are mis-declaring them. It would be useful if this suspicion could be clarified. According to the stated IATA rules you cannot ship by air.

Yes I got the quote from Amy in XUBA, She told me it was due to the only shipper being UPS into Ireland and she said they do not ship by sea. At that price not sure if it is worth it but their batteries are very keenly priced, 4x270A here
4pcs Lifepo4 Rechargeable 3.2v 280ah Li-ion Lithium Battery For Solar System For Ev - Buy Battery Rechargeable,Li-ion Lithium Battery,Battery Lifepo4 Product on Alibaba.com

 

[Zing]

Yes I got the quote from Amy in XUBA, She told me it was due to the only shipper being UPS into Ireland and she said they do not ship by sea. At that price not sure if it is worth it but their batteries are very keenly priced, 4x270A here
4pcs Lifepo4 Rechargeable 3.2v 280ah Li-ion Lithium Battery For Solar System For Ev - Buy Battery Rechargeable,Li-ion Lithium Battery,Battery Lifepo4 Product on Alibaba.com

It also depends how fast you have to have them. Sea adds a month to six weeks. They ought to be able to arrange it for you, but if they won't and often they struggle because of transhippment issues, then an Irish freight forwarder will help and who will organise everything for you, including probably a transshipment in England or Holland/Belgium.

 

[Poey50]

Update: some post-installation thoughts on a weakness in my system.

For those following, you may remember that I am using a 4 stage monitoring and control system which escalates from occasional monitoring to catastrophic-level protection. To recap:

1. Occasional Bluetooth monitoring of state of charge, individual cell voltage, cell temperature, balance between cells and state of charge history via the 123SmartBMS Bluetooth app. (This is where the phone screen-shots I have been putting up come from.)

2. Alternator, mains and solar charging set to autonomously end when absorption voltage is reached (13.8v for alternator, 14.0v for solar). Alarm in 3 will sound if these figures are exceeded. In addition the external regulator has an additional automatic voltage disconnect at 14 volts in case the external regulator fails.

3. Monitoring and alarms for pack high voltage, low voltage, high temperature and low temperature using the relay of the Victron BMV712 plus external alarm siren. The BMV also gives another Coulomb counting display for state of charge. The parameters are slightly outside the voltage cut-off of the chargers and slightly inside the parameters for the BMS disconnects below.

4. BMS and relay high voltage disconnect of charge bus, low voltage disconnect of load bus, high temperature and low temperature disconnect. All of these are based on cell readings rather than pack reading. This is catastrophic-level protection designed to protect the cells but not leave me high and dry. If the charging is disconnected the loads are unaffected and vice versa. (There is also a Sterling Alternator Protect fitted to prevent an alternator fry-up in the unlikely case of a high voltage disconnect during alternator charging.)

The sharp eyed will have noticed that levels 2 and 3 depend entirely on the system responding to pack voltage. Only 1 and 4 are using individual cell data and this is where the weakness lies. Both the alarms and the autonomous chargers depend on the end of absorption voltage being safe for the LFP cells. So for an end voltage of 13.8 volts this is equivalent to 3.45 volts per cell - very safe and able to give a full charge. But if the cells voltages were as follows: cell one 3.3v, cell two 3.3v, cell three 3.3v, and cell four 3.8v then the pack voltage would be 13.7 volts and this would fail to trigger the alarm and the fail to stop the charger(s) which could damage cell four - any voltage over 3.65 volts being problematic. Of course the high voltage cut-off at level 4 would prevent this but I don't want to solely depend on that.

So long story, short ... the use of absorption voltage to trigger alarms and end charging is only functional when the cells are reasonably in balance. Therefore, some occasional monitoring of cells at level 1 is necessary to ensure this and some occasional full charge to allow the passive balancing of the BMS to work is also a good idea. More sophisticated BMSs are able to end charging and to sound alarms at individual cell level and this is the ultimate in a fully automatic system. Such BMSs also cost more and, it might be argued, there is less redundancy should the BMS fail.

Anyway, something to be aware of if you are considering this kind of system.

For those thinking about a BMS with greater functionality than the 123SmartBMS (including the ability to directly control charging sources, rather than rely on voltage alone to end charging) the Electrodacus SBMSO is well worth a look. Will Prowse has a video on it too. Cheaper than the 123Smart, although without bluetooth.

https://electrodacus.com/SBMS0/SBMS0.pdf

ElectroDacus

Good, if shirtless, explanation here ...

 

[Poey50]

This one covers installation. Direct control of charging allows a maximum state of charge to be programmed in. About 13 minutes in he shows how he set it to 80% SOC.

 

[Mikedefieslife]

I actually ordered a Dacus a few weeks ago. Just waiting for it and my shunts to arrive. It will be replacing a cheap chinese BMS.

I like the switched output/input of Dacus, and will be using it to remotely switch my Victron MPPT controllers, and on the low side it will switch a Victron Smart Battery protect.

Also going to have a play with the software and see if I can get json output for signalk via wifi.

 

[Poey50]

I actually ordered a Dacus a few weeks ago. Just waiting for it and my shunts to arrive. It will be replacing a cheap chinese BMS.

I like the switched output/input of Dacus, and will be using it to remotely switch my Victron MPPT controllers, and on the low side it will switch a Victron Smart Battery protect.

Also going to have a play with the software and see if I can get json output for signalk via wifi.

I hope you'll report back.

 

[gregcope]

I really struggle to understand the Dacus website. Ie what features it has / what parts you need, etc... Or is that just me?


@Mikedefieslife Would be interested in your design/Parts/features.