270ah DIY LiFePO4 build

[vas]

I'll keep the bulky one for a future solar installation at home. Got around 20sqm to fill with panels

on a more serious note, I guess all of you fit the fuse on the positive side and close to the bank, right?
Was hoping to bolt the fuse housing on the positive side compression endplate but it's too bulky tbh!

hm, and another Q, you do power the BMS from the bank before the fuse, right?

V.

 

[Poey50]

on a more serious note, I guess all of you fit the fuse on the positive side and close to the bank, right?

hm, and another Q, you do power the BMS from the bank before the fuse, right?

Yes and yes.

 

[Poey50]

This is the latest article from Rod Collins of MarineHowTo, thankfully just back from a horrifying stroke as he explains. It is, of course, written within an US context but should count as a must-read for anyone considering drop-in LFP.

Drop-In LiFePO4 Batteries - Be an Educated Consumer - Marine How To

 

[gregcope]

Ah. Great to see Rob Collins posting again.

Also his pack stats are interesting; nearly 13 years old, 2200 cycles and still thier 400Ah capacity. I do like those winston cells.

 

[vas]

Ah. Great to see Rob Collins posting again.

Also his pack stats are interesting; nearly 13 years old, 2200 cycles and still thier 400Ah capacity. I do like those winston cells.

you wouldn't have two 12V 304Ah packs if you chose to go Winston though Greg
hope our EVEs can do half of that, would be good for a decade for me!

 

[vas]

hm,

I think we should seriously consider the resources needed for all the paraphernalia related to a safe LiFePO4 installation.
In the following pic you see the Class-T fuse, the shunt for the BMS and a last resort disconnect relay by Blue Sea.
I'm not talking about the combined cost of almost 400euro, I'm talking about the sheer space all these will take.
Looks like I'll set it up so that LiFePO4 positive is coming from the left to the fuse, then to the shunt and finally to the relay before going to the boat circuit.



in the space over the (unfinished) shunt I'll probably stick the BMS board (that's an un-finished board I have spare, fully functioning one is currently monitoring the bank on the floor next to me)



easily accessible, clean, cool, dry, 300X250mm area with almost 100mm clearance next to the battery bank is not always easy to find on a boat!

I understand that atm I'm alone in this forum going overboard with the diyBMS and shunt (already have a Victron shunt and will keep it for overal Victron kit measuring consumptions) but I doubt the other setups are going to be much leaner in terms of space needed! What I loose in space I gain in intelligence and integration of the system so that the BMS will be controlling charging and discharging from the Victron Inverter/charger and mainly the MPPT where the 95% of my charging comes.
Still one tiny chip to arrive from the depths of CN to complete the shunt and I'll be ready to move the lot to the boat! Snowing outside atm so not too worried with the delay

cheers
V

 

[vas]

oh well,

make that 350X250mm



wonder if having 100+A going all around the PCB of the BMS are going to cause havok to comms and measurements. Will ask the designer!

 

[Zing]

Do we really need to fuse the battery?

I don’t have one. All consumers downstream are fused. All cables from the battery are very big and routed separately to the main bus with no feasible possibility of a short as the neg wires are remotely located. In effect the battery can be thought of as directly connected to the bus bars.

I have some ridiculous peak currents in my setup. A big electric bow thruster, electric furlers, winches and windlass. Unlikely to be on at once, but if they were to be, no fuse could cope, so another reason to not do it.

I understand this is a common practice, as it is to not fuse the wires to the starter.

A bigger risk is an battery internal short. Maybe that should be addressed first.

 

[vas]

Do we really need to fuse the battery?

from my understanding, you sort of need it for the case of a short between the battery bank and the connection to the rest of the fused consumptions.
the way you describe it, you probably don't need it. I'm on the better safe than sorry (for 120odd euro, if it was 500 I'd definitely rethink about it!)

 

[Poey50]

hm,

I think we should seriously consider the resources needed for all the paraphernalia related to a safe LiFePO4 installation.
In the following pic you see the Class-T fuse, the shunt for the BMS and a last resort disconnect relay by Blue Sea.
I'm not talking about the combined cost of almost 400euro, I'm talking about the sheer space all these will take.
Looks like I'll set it up so that LiFePO4 positive is coming from the left to the fuse, then to the shunt and finally to the relay before going to the boat circuit.

View attachment 131609

in the space over the (unfinished) shunt I'll probably stick the BMS board (that's an un-finished board I have spare, fully functioning one is currently monitoring the bank on the floor next to me)

View attachment 131610

easily accessible, clean, cool, dry, 300X250mm area with almost 100mm clearance next to the battery bank is not always easy to find on a boat!

I understand that atm I'm alone in this forum going overboard with the diyBMS and shunt (already have a Victron shunt and will keep it for overal Victron kit measuring consumptions) but I doubt the other setups are going to be much leaner in terms of space needed! What I loose in space I gain in intelligence and integration of the system so that the BMS will be controlling charging and discharging from the Victron Inverter/charger and mainly the MPPT where the 95% of my charging comes.
Still one tiny chip to arrive from the depths of CN to complete the shunt and I'll be ready to move the lot to the boat! Snowing outside atm so not too worried with the delay

cheers
V

It all depends on what space saving you have from taking out the lead. I've been able to use the spare space in my moulded battery box for all the supplementaries. But my 270ah pack takes up the space of a typical 80ah lead acid.

 

[Poey50]

Do we really need to fuse the battery?

I don’t have one. All consumers downstream are fused. All cables from the battery are very big and routed separately to the main bus with no feasible possibility of a short as the neg wires are remotely located. In effect the battery can be thought of as directly connected to the bus bars.

I have some ridiculous peak currents in my setup. A big electric bow thruster, electric furlers, winches and windlass. Unlikely to be on at once, but if they were to be, no fuse could cope, so another reason to not do it.

I understand this is a common practice, as it is to not fuse the wires to the starter.

A bigger risk is an battery internal short. Maybe that should be addressed first.

I don't have big draw items so haven't looked into that aspect but the only pictures I've seen of burnt-out installations have been shorts where the massive amperage delivered into a short seems to be the culprit. Internal shorts within individual cells are certainly possible because of dendrite formation but don't seem to present those kind of problems because the job of the BMS is to prevent those conditions.

At some point insurers will catch up with the US market and insist on compliance with emerging standards all of which look likely to require Class-T main fuses.

 

[vas]

It all depends on what space saving you have from taking out the lead. I've been able to use the spare space in my moulded battery box for all the supplementaries. But my 270ah pack takes up the space of a typical 80ah lead acid.

well, not at all as the 4 T105RE Trojans are in the e/r and they are staying there as a back up in case for any odd reason bms decides to activate the bluesea relay and get the LiFePO4 offline - haven't worked on that at all right now.
Don't want extreme temps so LiFePO4 bank goes in the salon taking some of my easy to access tool chest under a two seater sofa.
for the record I was told in the meantime that there are likely to be comms issues with the fields created around these thick cables, so BMS will have to move closer to the batteries and circa 200mm from the cables.

 

[gregcope]

you wouldn't have two 12V 304Ah packs if you chose to go Winston though Greg
hope our EVEs can do half of that, would be good for a decade for me!

I did seriously look at them. Were quite expensive. Lots of youtubers have them (Ran/Sail Life) however there costs may not be such an issue.

 

[Zing]

I did seriously look at them. Were quite expensive. Lots of youtubers have them (Ran/Sail Life) however there costs may not be such an issue.

Compared to lead they are cheap into their cost being trivial no matter what you choose and secondary to their other advantages. Trojans lasted 4 - 5 years only. Replaced twice. Winstons, I just replaced with CALB, lasted for 12/13 years. They would have lasted longer if I hadn‘t left them stored fully charged and had I not cycled them so deep. I know better now. Maybe I also will deal with the summer heat in the tropics too, though I can see that running into cost/benefit issues.

 

[geem]

Compared to lead they are cheap into their cost being trivial no matter what you choose and secondary to their other advantages. Trojans lasted 4 - 5 years only. Replaced twice. Winstons, I just replaced with CALB, lasted for 12/13 years. They would have lasted longer if I hadn‘t left them stored fully charged and had I not cycled them so deep. I know better now. Maybe I also will deal with the summer heat in the tropics too, though I can see that running into cost/benefit issues.

If my Trojans last four or five years I will be super happy. Heat kills batteries whether they are lithium or lead. Most friends I know running Trojans out here get circa 8 years out of them but they are using 8 batteries and I suspect a vastly reduced capacity at 8 years. I reduced my bank size to just 4 T105s since we rarely see them drop below 80% charged. At £600 for four it's a no brainer on cost.
We will be installing a liquid cooled lithium prototype battery later this year but I will keep the T105s as a back up/engine start arrangement. We have plenty of space for both

 

[Zing]

If my Trojans last four or five years I will be super happy. Heat kills batteries whether they are lithium or lead. Most friends I know running Trojans out here get circa 8 years out of them but they are using 8 batteries and I suspect a vastly reduced capacity at 8 years. I reduced my bank size to just 4 T105s since we rarely see them drop below 80% charged. At £600 for four it's a no brainer on cost.
We will be installing a liquid cooled lithium prototype battery later this year but I will keep the T105s as a back up/engine start arrangement. We have plenty of space for both

Just four Trojans! Impressive. I think you must have an interesting story to tell about energy efficiency?

I got 6 or 7 years out of my engine/genset AGM Lifelines for what it’s worth. A better life, but a higher net cost. When they die again I will probably get lithium and a battery to battery charger.

 

[geem]

Just four Trojans! Impressive. I think you must have an interesting story to tell about energy efficiency?

I got 6 or 7 years out of my engine/genset AGM Lifelines for what it’s worth. A better life, but a higher net cost. When they die again I will probably get lithium and a battery to battery charger.

I read an excellent article that suggested the most economical battery bank was the one that reached its design life cycles at the same time as the heat in the tropics killed them. The idea being to use a relatively small battery bank, cycle them to 50% everyday then replace them. This is why we didn't go for a large bank of Trojans. In reality, we don't use much power. 1.5-2 kw of solar a day plus some input from the wind turbine runs our boat including making a tank full of hot water. We haven't used the battery charger in months.

 

[gregcope]

Compared to lead they are cheap into their cost being trivial no matter what you choose and secondary to their other advantages. Trojans lasted 4 - 5 years only. Replaced twice. Winstons, I just replaced with CALB, lasted for 12/13 years. They would have lasted longer if I hadn‘t left them stored fully charged and had I not cycled them so deep. I know better now. Maybe I also will deal with the summer heat in the tropics too, though I can see that running into cost/benefit issues.

@Zing interesting you have replaced some Lifepo4.

What capacity did your Winstons fall to (if that was a reason to replace?)?
How many cycles did you get?
Do you have high average temps?
Also what do you call deep cycling and what would you aim for now?

I am intrigued by your real world experience as there are few people who have this real world longer experience.

 

[Zing]

@Zing interesting you have replaced some Lifepo4.

What capacity did your Winstons fall to (if that was a reason to replace?)?
How many cycles did you get?
Do you have high average temps?
Also what do you call deep cycling and what would you aim for now?

I am intrigued by your real world experience as there are few people who have this real world longer experience.

It was a saga of slow death. On cell failed, replaced, then another, replaced, then another including a used one I bought to replace a failed cell. Then I binned the lot. Cell swelling was the problem with a failure to hold a charge.

Cycles were not so high. These cells have a limited cycle life and a limited age. Mine are big and I will never use up the cycles. I ran up against the age limit I believe. It's still an impressive life. I expect it to be a bit longer now I am looking after them better.

I used to run them 95% to zero. Not too clever I recently learnt. I now run them mostly in the middle, so 40% to 70% with the occasional run to full or empty to reset the coulomb counter, which wanders off a good bit.

 

[Poey50]

I've just seen these extracts from the ABYC technical guidance. I've never seen a similar list - only bits.

ABYC recommendations to be considered when designing and/or implementing a LFP battery system. These are taken from ABYC TE-13 - "Lithium Ion Batteries" Technical Information Report (soon expected to become a standard).

"13.7.7 No electrical connections should be made directly to lithium ion battery that would bypass a BMS or the protection relays. Exception: connections made specifically per battery manufacturer instructions.
...
13.5.3 Considerations should be given to providing alternative power for critical systems (e.g. engine starting, navigation lights, etc.) if a BMS shuts down the battery. The secondary power source can be a second lithium ion battery bank.
...
13.5.6 If a shutdown condition is approaching, a BMS should notify the operator with a visual and/or audible alarm prior to disconnecting the battery from the DC system.
...
13.7.8 All battery output circuits should have overcurrent protection as per ABYC E-11, AC&DC Electrical Systems on Boats.
...
13.7.8. 1 If necessary, a battery bank should be subdivided into units such as the ampere interrupting capacity (AIC) of the overcurrent protection device is not exceeded.
NOTE: Generally, fast acting current limiting fuses such as Class T fuses in an approved Class T fuse holder have an AIC of 20.000 amps at 12VDC and will be adequate for a subdivided bank.
...
13.9.2 Redundant Protection - A BMS should be equipped with high/low voltage cutout actions in response to high/low voltage events when the programmed functions in the charging sources, inverters, or inverter/chargers, etc. fail to do so.
...
13.9.4. 1 Means of protection should not disconnect critical loads without prior warning and should not stop the charging source in a manner that causes damage to the charging device. (e.g. alternator disconnect under load).
...
13.9.5 The BMS should respond to low or high temperature situations in a manner that ensures the battery cannot be driven into an unsafe condition."