304Ah LifePO Update - Year 2

[gregcope]

Hi All,

Thought I would give an update on year two of my DIY 304Ah Build.

Some data;



Some observations;

1. Battery One is 316Ah, Battery Two 307Ah. This is intentional to force the smaller capacity one to drop out first if we ever get very, very low SOC.
2. The FLA is charged by 110W of solar. The LifePo4 is charge by a 30W B2B from the FLA and during the season by 220W of solar. The Solar is removed when ashore (Nov-mid Mar). The batteries
3. Lowest temp onboard was 3C. In the graph the dip to -1C was when they were in my shed! I had taken the Batteries home in Oct 2022-Mar-2023
4. SOC is not accurate and should be treated as a guide
5. This system easily keeps up with domestic and sailing loads, even with a few days of gray/low solar days
6. Individual Cells voltage delta is around 1-2mV. There is a jump at 12mV when cells were being charged. The ballancer has not kicked in as the cells never reached 3.45V balancer voltage
7. The batteries themselves drift in terms of SOC/Voltage. I went to some lengths to ensure Batteries were equal in terms of resistance/cabling. I need to investigate this a little more, although its easily addressed by switching off a high/low battery and letting the other catch up - you can see a few examples of this over the years
8. I intentionally keep SOC between 30-70% Although this is hard when doing long motoring on sunny days as the solar and B2B easily charge the Batteries and drive up the SOC
9. Overwinter 2023-2024 the SOC/Voltage does drop. There was much more temperature variation in the (insulated) in land shed than onboard in the car park next to the sea.
10. Overwinter 2023-2024 the SOC/Voltage is flat. Remarkably low (nearly unmeasurable) self-discharge. The voltage and SOC are flat - cell voltage is hardly change, Battery voltage is unchanged as is SOC.
11. 40% SOC is around 130Ah (ish) which means I start the season with 260Ah nominal or around 130Ah usable (to 20% SOC). Which is still nuts as the previous FLAs I had combined had 130Ah usable if fully charged

Any thoughts or feedback? What have I missed or misunderstood?

I have battery heaters to install. Although they would not be used as I never seem to be in a position to charge them when below 5C as the boat is shore. Also I need to hook up the RS485 to a VictronOS monitoring solution and graph this with that data instead of me screenshoting the Bluetooth app and putting it into a google spreadsheet. This will also allow me to enable alarms on low or high SOC/Battery/Cell V etc.

 

[Neeves]

Really useful.

I confess, based on my location, that I am amused by the idea of the need to heat a battery - here the problem is heat and maybe? the need for cooling (computer fans). However I have noted that there have been comments, negative, from owners in the Northern hemisphere winter. I also receive mail shots from battery suppliers underlining they 'are equipped with a heater'.

Any chance you can define usage: what do you run off the arrangement? - primarily the domestic loads? (I can guess the yacht demands).

If I summarise you have 600ah of battery, 220w of solar and a 30w B2B

But you have a big FLA and it has 110w of solar - but the FLA is just, or only, engine start? and is 'also' charged from the alternator

Big question - where are you?

Jonathan

 

[vas]

Also I need to hook up the RS485 to a VictronOS monitoring solution and graph this with that data instead of me screenshoting the Bluetooth app and putting it into a google spreadsheet. This will also allow me to enable alarms on low or high SOC/Battery/Cell V etc.

nice Greg,

but unless you get 1min granularity on your data, you've done nothing
[where's GHA when you need him!]
as we've discussed, just get a rpi with a RS485 HAT, install VenusOS LARGE and you're ready to rock!

on a serious note though, if the system's been working flawlessly for (over!) a year, you shouldn't need it. I just have a look over the last 24h data whenever I remember just to make sure nothing's gone silly. Yes, email warnings will come on anything v.silly, but just in case!

I think after all these independent threads with different BMSs (cells are invariably v.similar - in our case EVE 280-304Ah!) running over 2-3yrs now we can 'safely' assume that LifePO4 is fine in PBO fashion in a boat assuming you understand what you're doing. Not plug and play (imho) though just yet, charging systems must be "updated" and that's not going to happen just like that...

cheers

V.

 

[gregcope]

Hi Jonathan

I am based in southern UK and do coastal sailing around the Solent area.

To be honest domestic loads are tiny compare to capacity - fridge/lighting/water pump/Internet 4G Router. I am fitting a large inverter (3000KVA) as I intend to spend more time on board. The main sailing loads are the Autohelm Ram (12V T2 B&G) and very (very) rarely radar. As LifePO4 becomes more prevalent, installs will increase in size (compared to present FLA) because you can relatively cheaply.

I have a 60Ah AGM (600A CCA) 12V engine start battery which is charged from 110W of Solar via MPPTs. There is another 220W of Solar charging the LifePO4 via MPPT and a 30W B2B which charges the LifePO4 from the AGM when the engine is running or when the solar output trips the B2B charge logic (basically voltage above a certain level for 120 secs).

I have two 300Ah LifePo4 Batteries. I would expect over 400Ah usable between them (ie not going below 20% or above 90% SOC). I could drop down below 20% in an emergency (120Ah! between the two).

Heaters are more or less essential for winter RV/Camper use as they tend to have their batteries outside the heated envelope and hence can easily see below Zero temps. LifePO4 does not like being charged below zero (you can trickle charge if brave). There have been issues with winter use with some "drop ins" that do not have heaters (or kept above 0C) hence the marketing. From all the datasheets I have read, LifePO4 is like a human. Loves 25C. One reason cars have complexity around battery thermal management to keep them warm/cool, especially when fast charging.

I think @geem was making/designing a cooled LifePO4 for warmer climes as in the tropics the batteries can see high 30's which impacts longevity.

 

[gregcope]

nice Greg,

but unless you get 1min granularity on your data, you've done nothing
[where's GHA when you need him!]
as we've discussed, just get a rpi with a RS485 HAT, install VenusOS LARGE and you're ready to rock!

on a serious note though, if the system's been working flawlessly for (over!) a year, you shouldn't need it. I just have a look over the last 24h data whenever I remember just to make sure nothing's gone silly. Yes, email warnings will come on anything v.silly, but just in case!

I think after all these independent threads with different BMSs (cells are invariably v.similar - in our case EVE 280-304Ah!) running over 2-3yrs now we can 'safely' assume that LifePO4 is fine in PBO fashion in a boat assuming you understand what you're doing. Not plug and play (imho) though just yet, charging systems must be "updated" and that's not going to happen just like that...

cheers

V.

I aspire to @GHA level of data logging... Too many projects, not enough time! Will get there with the PI! My RS485 cables are ready!

Agree that there are a number of these Batteries out there now based on these cells/BMS combos that seem on the whole reliable (without any real data to be fair!).

I am still thinking about what I might have missed/what might go wrong, unkown unkowns if you will.

 

[geem]

It would be interesting to see how many cycles you have done in that period.
As an observation, I would say you have too much capacity for your solar output. Thr beauty of lithium is being able to access the large capacity. Running them down to 30% is not a problem but you need the charging capacity to put it back.
I would add the solar you have charging the engine battery to the lithium bank. Then use a Victron DC/DC to keep the engine battery charged off the lithium. You are currently wasting that solar capacity.

 

[gregcope]

The BMS have registered no cycles this year. Not sure what they deem to be a "cycle"

@geem can you expand on "too much capacity for your solar output"?

I have though of this as Solar capacity (or better expected average yield) should match usage and storage is more a factor of space, budget and how long (days usage) you want to cover for poor yield days. I made two batteries as I needed 250A to cover the inverters max 3000VA as well as redundancy. Cells wise 304Ah was not much more than 280Ah.

I understand your point on the AGM Solar. Sadly, those MPPTs cannot be changed/configured (they are not Victron!). The boat lives on a mooring where I want the Engine start battery charged. This also runs the Bilge pump. When left, I isolate the LifePO4 to avoid the solar taking them to full in the time the boat is not used.

 

[geem]

The BMS have registered no cycles this year. Not sure what they deem to be a "cycle"

@geem can you expand on "too much capacity for your solar output"?

I have though of this as Solar capacity (or better expected average yield) should match usage and storage is more a factor of space, budget and how long (days usage) you want to cover for poor yield days. I made two batteries as I needed 250A to cover the inverters max 3000VA as well as redundancy. Cells wise 304Ah was not much more than 280Ah.

I understand your point on the AGM Solar. Sadly, those MPPTs cannot be changed/configured (they are not Victron!). The boat lives on a mooring where I want the Engine start battery charged. This also runs the Bilge pump. When left, I isolate the LifePO4 to avoid the solar taking them to full in the time the boat is not used.

If you were to discharge the lithium batteries you have, it would take a long time to charge them back up. This makes the prospect of discharging them, less than ideal. You arent leaving the solar to its charging whilst the boat is on the mooring so you really arent getting the best out of the lithium. If you set the maximum chaege voltage to 13.6v, you could leave the lithium connected to the solar whilst you were not there. You could then heavily discharge them whilst onboard for the weekend, knowing they would be charged back up by the following weekend. If you haven't registered a discharge cycle then you batteries are really not being used at all. They will die of old age before they ever get to even a 1/4 of their way through the design life cycles
By comparison, one of my batteries (280Ah@24v)has been in use 325 days and has done 82 full cycles. This is equivalent to approximately 610kWh. The other smaller battery (105Ah@24v) has done 24 cycles in 122 days. Equivalent to approximately 67kWh. We have 920w of solar connected, currently harvesting 3.5-4kw per day.

 

[gregcope]

Understand your points @geem

My intent was to have a little Covid project as I needed to replace my FLAs. I wanted a bank that would last a long while. Yes my bank could do allot more. It suits me well presently. Without any induction cooking or hotwater heating or watermaker our loads are low by most standards. In the summer the solar keeps up with that on most days, even in the UK, throw in any motoring and we easily keep up with loads. Your usage profile sounds allot different so I would expect you to be cycling more.

The only downside to my usage is that my bank might seem to be "too large" and I have over committed. Building a smaller one was not much cheaper as around half the cost is in the cases, BMS and lugs and not the cells themselves.

 

[vas]

If you were to discharge the lithium batteries you have, it would take a long time to charge them back up. This makes the prospect of discharging them, less than ideal. You arent leaving the solar to its charging whilst the boat is on the mooring so you really arent getting the best out of the lithium. If you set the maximum chaege voltage to 13.6v, you could leave the lithium connected to the solar whilst you were not there. You could then heavily discharge them whilst onboard for the weekend, knowing they would be charged back up by the following weekend. If you haven't registered a discharge cycle then you batteries are really not being used at all. They will die of old age before they ever get to even a 1/4 of their way through the design life cycles
By comparison, one of my batteries (280Ah@24v)has been in use 325 days and has done 82 full cycles. This is equivalent to approximately 610kWh. The other smaller battery (105Ah@24v) has done 24 cycles in 122 days. Equivalent to approximately 67kWh. We have 920w of solar connected, currently harvesting 3.5-4kw per day.

@geem, how are these cycles calculated? or are you just getting the amps in (or out in that matter...) from the battery BMS and divide by the 20-95% capacity of the bank?
Is there a formula to do that?
I mean off season mine does v.little (typical consumption of bms/nmea2k bus and devices/victron kit/battery bms is less than 50W) so even come winter 600W solar replenish that every few days at worse down here. Still after this off season, bms has clocked some Ah which was really getting the bank from 45-50% SoC or there abouts. In my mind this counts as bugger all cycles done...
OK when on board in the summer with watermaker/inverter/and whatnot I get more but rarely a 30-90% in a day to call it a cycle.
Haven't seen anywhere documenting real life cycling.

cheers

V.

 

[geem]

@geem, how are these cycles calculated? or are you just getting the amps in (or out in that matter...) from the battery BMS and divide by the 20-95% capacity of the bank?
Is there a formula to do that?
I mean off season mine does v.little (typical consumption of bms/nmea2k bus and devices/victron kit/battery bms is less than 50W) so even come winter 600W solar replenish that every few days at worse down here. Still after this off season, bms has clocked some Ah which was really getting the bank from 45-50% SoC or there abouts. In my mind this counts as bugger all cycles done...
OK when on board in the summer with watermaker/inverter/and whatnot I get more but rarely a 30-90% in a day to call it a cycle.
Haven't seen anywhere documenting real life cycling.

cheers

V.

My JK bms gives cycles. It is the 100% capacity of the battery, so in my case 280Ah @24v
It also tells me the amp hours used. This is 23,274Ah divide by 280 and you get 83 cycles

 

[vas]

My JK bms gives cycles. It is the 100% capacity of the battery, so in my case 280Ah @24v
It also tells me the amp hours used. This is 23,274Ah divide by 280 and you get 83 cycles

these are v.dodgy cycles then...
so my 12kAh over the last 12m on a 304Ah bank means almost 40cycles?
hope I'll live to see it dropping to 80% capacity!

 

[gregcope]

@vas 3500/40 = 87.5 years. That's at 1C.

You, like me are going to see calendar aging (is it around 1%/yr? or 20 years to 80%) before then...

 

[geem]

@vas 3500/40 = 87.5 years. That's at 1C.

You, like me are going to see calendar aging (is it around 1%/yr? or 20 years to 80%) before then...

Not much different for me.
I think the lesson to learn is don't over size your lithium battery bank. Make use of its capacity and save yourself money on lots of big cells.

 

[vas]

Life expectancy is increasing, I'm 60 in two weeks time, boat is 50... Lets see how it pans out, hope it's the batteries

on a serious note, in September with 3-4days overcast and need to run the watermaker a couple of times I can go from 90% to 20%, so I'd really not have liked a smaller bank. OK, I can fire up the generator but try to avoid it as much...

 

[geem]

Life expectancy is increasing, I'm 60 in two weeks time, boat is 50... Lets see how it pans out, hope it's the batteries

on a serious note, in September with 3-4days overcast and need to run the watermaker a couple of times I can go from 90% to 20%, so I'd really not have liked a smaller bank. OK, I can fire up the generator but try to avoid it as much...

I am 60 in July. Boat is 44. I am hoping my batteries also die first

 

[GHA]

Looks ike I popped in for a quick look at just the right time!
Thanx for the kind words, means a lot!

Down to business, I'm just over a year down the line as well & just got cell monitoring on the go. Using a 4 channel Opamp to differentially measure the voltage of each cell & then using a esp8266 send it to the trusty Pi. This bank has never been balanced, just had a first stab at calibrating the cell sensors against a fluke & looks pretty close, though don't think outside the voltage knees you can actually say much about a cell just looking at the voltage. So we wait the knee!


The blind optimistic hope is that watching where each cell knee starts will give a hint at relative SOC so then using a separate battery & CC CV power supply pump a little charge into any cells which look low. We'll see

Also, hurrah for Signalk!! Or Victron actually, now in their bluetooth app you can get the mac address & advertising key of a smartshunt or smartsolar & get a raspberry pi to read the data straight over bluetooth in a signalk app. This is nearly all bluetooth victron data, no wires, takes 5 minutes to set up. Pi3 struggled juyst getting data about once a minute but Pi4 in every second or 2.
In the wonderful Kip app.


Maybe if I keep quiet in here I can avoid arguing with the bitter forum dementors.... they'll suck the joy from your soul...

 

[GHA]

The only downside to my usage is that my bank might seem to be "too large" and I have over committed. Building a smaller one was not much cheaper as around half the cost is in the cases, BMS and lugs and not the cells themselves.

I don't see any problem going a bit big. As you say the money isn't that great and we're in unknown territory regarding lifespan, could be in the future people will sometimes take the batteries with them when selling a boat, put in some LA. Rod Collins was a very early adaptor and one of the very few who can actually do an accurate capacity test -
Drop-In LiFePo4- Be an Educated Consumer

It’s important to know that I’ve been using lithium iron phosphate batteries on my own vessel since early 2010 I built my battery bank back in 2009 well before any of these drop-in batteries even existed. My LFP bank will be 14 years old on May 10 of 2023. The bank has in excess of 2200 cycles on it & most every single cycle has been to at least 80% depth of discharge with many many, many cycles (at least 100+ cycles) going to 0%. That battery bank can still deliver 100% of its 400Ah rated capacity so LFP batteries can last, where lead-acid simply don’t.

There was great joy in the anchorage a few weeks ago onboard when I hit 100% for the first time since last October Certain death for lead acid that would be.

 

[geem]

@vas 3500/40 = 87.5 years. That's at 1C.

You, like me are going to see calendar aging (is it around 1%/yr? or 20 years to 80%) before then...

From data I have on calender aging. If you keep your batteries at 100% SOC, calender aging is worst. Expect to see 3.5% lose of capacity annually. If you mainly keep the batteries fully charged but with some use, so you then only experience 2.5% loss of capacity per year, you would see 13% loss over 15 years. Not massive. This assumes an average of 15degC. About right for the UK.
If you keep the batteries at 100% charged and average 35degC, you will likely see a drop of 10% in capacity annually. This is what we are battling with to some extent in the Caribbean. We don't leave the batteries at 100% and we run mechanical cooling to keep the temperature down, making use of cooler cabin temperatures at night to cool the batteries. We probably average about 28degC over the year in the Caribbean if we live aboard full time. The difficulty is when you leave the boat in a yard for the summer and batteries see an internal cabin temperature of 45degC for 3 months

 

[GHA]

From data I have on calender aging. If you keep your batteries at 100% SOC, calender aging is worst. Expect to see 3.5% lose of capacity annually. If you mainly keep the batteries fully charged but with some use, so you then only experience 2.5% loss of capacity per year, you would see 13% loss over 15 years. Not massive. This assumes an average of 15degC. About right for the UK.
If you keep the batteries at 100% charged and average 35degC, you will likely see a drop of 10% in capacity annually. This is what we are battling with to some extent in the Caribbean. We don't leave the batteries at 100% and we run mechanical cooling to keep the temperature down, making use of cooler cabin temperatures at night to cool the batteries. We probably average about 28degC over the year in the Caribbean if we live aboard full time. The difficulty is when you leave the boat in a yard for the summer and batteries see an internal cabin temperature of 45degC for 3 months

What's fully charged though? And how much difference does "almost" make to lifespan? Offgrid garage has loads of test data,

I go up to 13.7v then to 13.35v float with no absorption & from todays data & offgrid garage it's certainly not 100% but well up there. Will be interesting to see how repeatable this is & if pushing some charge into one cell at a time can balance them a bit closer without turning everything off. >





From forums I suspect a lot of people get freaked out with active balancing & keep the voltage high with timed absorption to let that kick in every day where it may be be better to keep the voltage lower as much as possible most of the time, just balance now & then. My cells were balanced before installing over a year ago. Bottom line on boats I reckon is nobody really knows. Far too early.