Victron LiFEPO4

[jbinbi]

https://www.emarineinc.com/categorie...hium-Batteries

Maybe I missed it, but I don't see anyone talking about these. A complete battery solution from Victron. Is this pricier than other solutions?

Interesting that the 150 Ah weighs 44 lbs, the 300 is 110 lbs. 2x the capacity at 3x the wt?

If there is another thread , my apologies.

[mikedefieslife]

I saw some of the Victron LiFePO4 batteries last time I was in the boatyard. Their 90ah unit was as large as my DIY 271ah bank.

Yes they are pricier than custom solutions but then you get the support and benefit that everything just work together.

[john61ct]

Originally Posted by jbinbi

Maybe I missed it, but I don't see anyone talking about these. A complete battery solution from Victron

Along with Lithionics from OceanPlanet, and Mastervolt to a lesser extent, these are mentioned in every thread looking for recommendations.

Victron is the leading top "packaged system" vendors worldwide, although some great systems are also sold from local vendors in Australia.

But yes, **lots** more expensive than buying quality bare cells and adding the BMS functionality yourself.

Then there are "drop-ins", sealed up inaccessible units, I consider unsuitable for most use cases, but the lines are getting greyer as new products come out on the market.

[s/v Jedi]

Actually, I recently watched a video where a Battleborn battery was cut open and what came out was a pleasant surprise. They are twice the price of Firefly carbon AGM but can do 3,000 cycles at 100% DOD

Watch the video on YouTube.

[john61ct]

At what **they call** 100% DoD.

The internal BMS isolates the battery (without warning) at the point any condition is not conducive to the battery's health, including low voltage.

That BMS is opaque, parameters are not under user control, can't be adjusted or disabled, no communications to the outside, and they can suddenly isolate the battery with no warning.

Special steps must be taken to ensure this can't damage charge sources or sensitive load devices.

And when (not if) the BMS fails. you will need to cut open the bank to replace it, if the warranty/maker won't be round to swap it out.

I am very skeptical that drop-ins have the same quality of manufacturing, purity of chemicals etc.

Who is the actual manufacturer of the cells internally?

It is a very dicey market, thousands of small Chinese companies competing in sometimes shady ways, very little transparency no decent word of mouth.

I think very unlikely they will have the lifespan of Winston/Thundersky/Voltronix, CALB, GBS, Sinopoly. And so little collective experience has accumulated wrt drop-ins, this factor won't be know for decades

Terminating charging when full, or when temps get too low or too high, should be able to signal other gear, or controlling generic relays.

As opposed to simply taking a batt offline without warning, which load dump can cause damaging spikes/surges.

The BMS can't be bypassed, even to check on the cells' voltages, and often drastically limits charge / discharge C-rates to a tiny fraction of LFP's capabilities.

LFP is usually capable of accepting very high charge rates, no problem. To the point that 2-3 day's worth of energy can be pumped back into the bank in an hour or two if you have a high-current charge source available. But the cheap included BMS in drop-ins prevents that, restricts you to a slow charge (and discharge) rate.

Check to see if they even allow serial or parallel connection into a bigger bank?

Allowing the "Full" or "Too Low" setpoints to be adjusted by the user to extend lifespan. The balancing function may require a charge voltage that is higher than what you want for longevity.

Allowing the user to even **see** the state of cell balance, verify that BMS function is working.

Allowing a cell to be swapped out if one proves faulty.

All impossible with drop-ins.

You have to understand, the whole "drop-in concept" is a lie if you want longevity, you need a ​good system, all the infrastructure designed for LFP from the ground up

not just BMS
over-current protection
LVD and HVD*safety*cut-offs
temperature protection

but also
100% PROGRAMMABLE charge sources
IMO a separate charge and loads bus

Finally for a large bank, they're rarely cheaper, might as well get a packaged system from OPE-Li3 (Lithionics/Ocean Planet), Victron or MasterVolt.

Super B looks worth checking out also.

If you aren't willing / able to go the DIY route from bare cells.

[s/v Jedi]

I was thinking like you, then I watched that video. I bet you didn’t watch it yet, you should.

These guys are in Lithium Valley in Arizona right next to lithium factories from Tesla etc. I also believe you get a 10 year warranty.

Yes, the BMS cuts off without telling you. But you have a SOC indication, right? And a whole bunch of these 100Ah batteries in parallel.

Originally Posted by john61ct

At what **they call** 100% DoD.

The internal BMS isolates the battery (without warning) at the point any condition is not conducive to the battery's health, including low voltage.

That BMS is opaque, parameters are not under user control, can't be adjusted or disabled, no communications to the outside, and they can suddenly isolate the battery with no warning.

Special steps must be taken to ensure this can't damage charge sources or sensitive load devices.

And when (not if) the BMS fails. you will need to cut open the bank to replace it, if the warranty/maker won't be round to swap it out.

I am very skeptical that drop-ins have the same quality of manufacturing, purity of chemicals etc.

Who is the actual manufacturer of the cells internally?

It is a very dicey market, thousands of small Chinese companies competing in sometimes shady ways, very little transparency no decent word of mouth.

I think very unlikely they will have the lifespan of Winston/Thundersky/Voltronix, CALB, GBS, Sinopoly. And so little collective experience has accumulated wrt drop-ins, this factor won't be know for decades

Terminating charging when full, or when temps get too low or too high, should be able to signal other gear, or controlling generic relays.

As opposed to simply taking a batt offline without warning, which load dump can cause damaging spikes/surges.

The BMS can't be bypassed, even to check on the cells' voltages, and often drastically limits charge / discharge C-rates to a tiny fraction of LFP's capabilities.

LFP is usually capable of accepting very high charge rates, no problem. To the point that 2-3 day's worth of energy can be pumped back into the bank in an hour or two if you have a high-current charge source available. But the cheap included BMS in drop-ins prevents that, restricts you to a slow charge (and discharge) rate.

Check to see if they even allow serial or parallel connection into a bigger bank?

Allowing the "Full" or "Too Low" setpoints to be adjusted by the user to extend lifespan. The balancing function may require a charge voltage that is higher than what you want for longevity.

Allowing the user to even **see** the state of cell balance, verify that BMS function is working.

Allowing a cell to be swapped out if one proves faulty.

All impossible with drop-ins.

You have to understand, the whole "drop-in concept" is a lie if you want longevity, you need a ​good system, all the infrastructure designed for LFP from the ground up

not just BMS
over-current protection
LVD and HVD*safety*cut-offs
temperature protection

but also
100% PROGRAMMABLE charge sources
IMO a separate charge and loads bus

Finally for a large bank, they're rarely cheaper, might as well get a packaged system from OPE-Li3 (Lithionics/Ocean Planet), Victron or MasterVolt.

Super B looks worth checking out also.

If you aren't willing / able to go the DIY route from bare cells.

[s/v Jedi]

I went over to find the video. Here it is:

https://www.youtube.com/watch?v=G5E30u-66VI

[smac999]

The 150 is a newer model and weighs way less then the older ones. The 200-a just came out and is way lighter then the previous 200 as well.

[john61ct]

Putting more than a few strings in parallel to build capacity is not a good idea, imbalance issues will cause uneven wear and reduced longevity.

If you can't buy cells where Ah are big enough for your bank size, then they should be paralleled at the bottom layer, where they will act as a single cell.

Will has learned a fair bit in the last 18mo's, but he's not an expert by any means.

I'm not at all knocking Battle Borne, out of that category they are respectable, may well honor their warranty for 10 years if the company lasts that long

and for some use cases I'd be happy to recommend their product.

But the whole category of "drop in" is flawed and very strongly do not reco that for use in a large House bank.

Compared to any of the alternatives.

If you can't/won't build up from bare cells, and can't afford the quality packaged systems, better off sticking with lead.

[nebster]

Originally Posted by john61ct

Putting more than a few strings in parallel to build capacity is not a good idea, imbalance issues will cause uneven wear and reduced longevity.

Objectively false, in my experience:

It can be a good idea.

There is no detectable imbalance.

The wear is even, to the extent it is even measurable.

Longevity is tracking great.


The typical user is paralleling these (or one of many other) drop-in packs. I have personally seen dozens of field installs of these specific batteries at this point.

I am not an advocate of drop-in packs for other reasons, including some that you cite later in your post, but you are still leading off with this incorrect statement.

[john61ct]

Sorry, just wrong, or certainly at least not optimal.

I concede that if all other conditions are optimised, and intra-bank wiring connections are done properly

http://www.smartgauge.co.uk/batt_con.html

imbalance problems may well not be noticed to the untrained owner for many years, even past half way through the bank lifespan.

But they will arise, even the best cells will over time have increasingly differing internal impedance/resistances, therefore the same voltage comes to mean different SoC.

So, the guideline remains, if your battery bank requires more than 2-3 paralleled strings, it is better to choose higher capacity cells in order to reduce the parallel connections.*

The chances of different resistance paths in just a pair of a parallel wired strings is high.

More strings increases those odds. Also means many more cells to service, more connections that can go loose or fail, even if not, higher resistance overall.

More parallel strings will mask problems in adjacent strings

much higher chance that you will have a cell failure and not catch failure before the whole string is destroyed.

A lithium bank is best built by doing all the paralleling down at the lowest, first 1S cell level to get the required Ah, and then put those groups in balanced series, to get the required voltage.

[nebster]

Originally Posted by john61ct

Sorry, just wrong, or certainly at least not optimal.

Sorry, you are wrong. "Optimal" is in the eye of the beholder/designer. There is no one true answer for battery topology.

I concede that if all other conditions are optimised, and intra-bank wiring connections are done properly

No one hand-building a complex battery should be cross-strapping thousands of dollars of LFP (drop-in or not) using that old document. Bus all the batteries up to a central point.

imbalance problems may well not be noticed to the untrained owner for many years, even past half way through the bank lifespan.

Maybe, but you'll have to show your work, John. I'm not observing any in my battery.

So, the guideline remains, if your battery bank requires more than 2-3 paralleled strings, it is better to choose higher capacity cells in order to reduce the parallel connections.*

Let's be clear: that's the latest version of John's guideline.

There is absolutely nothing that suggests that n > 3 strings introduces some magical new problem. In real world batteries being used for years, like mine.

The chances of different resistance paths in just a pair of a parallel wired strings is high.

It's not only high, it's guaranteed, and that would be true even if you had ideal cells with identical internal resistances. Have you ever measured interconnect resistance variation? I suspect you haven't. It dominates in a parallel-string scenario with healthy cells, and it is a big part of why the theory you advocate doesn't hold in real life.

More strings increases those odds.

I question the premise.

Also means many more cells to service, more connections that can go loose or fail,

The number of cells in a battery is independent of topology. If you're making an unrelated point, namely that fewer connections is simpler and better, then I completely agree. But it has nothing to do with serial- or parallel-first.

even if not, higher resistance overall.

More connections definitely produce higher resistance. You got a clause correct!

More parallel strings will mask problems in adjacent strings

A single string will mask problems in individual cells, which is what you must measure no matter the topology. This sentence is incorrect.

much higher chance that you will have a cell failure and not catch failure before the whole string is destroyed.

I dispute the premise.

A lithium bank is best built by doing all the paralleling down at the lowest, first 1S cell level to get the required Ah, and then put those groups in balanced series, to get the required voltage.

There is no one right answer. Stop spoon-feeding people half the story.

[evm1024]

Originally Posted by john61ct

Looking for sources for **quality** LiFePO4 prismatic cells. Ideally 20-25Ah, but 15-40Ah is OK. I'm in North America, but realize a local supplier is unlikely or would be too expensive.

Ideally would sell on eBay or at least take PayPal, unless they have a stellar reputation here for integrity & customer service.

I'm looking for the bigger block style "prismatics", where one cell is like a brick size, 20-40Ah each, so only need say 6P4S to make a 240Ah 12V pack, worth maybe a grand USD

CALB, Winston, GBS, Sinopoly are well-regarded examples.

I'd really rather not just ship import myself, if they turn out to test out as sub-par, any warranty doesn't do much good if shipping costs back to China costs more than the cells are worth.

John61ct, I know you were going to buy some LiFePO4 cells back in Feb 2019 and assemble a bank to get some experience on. Did you ever do that? You do hold some very strong opinions on LiFePO4 and it is good that you get some real world experience.

I don't think that you could go wrong with a Victron LiFePO4 battery either by the way.

Also, did you ever find those A123 type cells you were looking for back in May 2019?

Originally Posted by john61ct

Looking for good LFP cylindrical batteries, buying new.

So far it seems A123 is the gold standard.

It is really good to see you build up your experience by buying cells and such. Rather than assembling a few A123 cells you could just buy a Battleborn and disassemble it to get access to the cells for measurement. This might be a great platform to learn on.

Lastly, back in April 2019 you made some recommendations that appear to contradict other posts of yours:

Originally Posted by john61ct

Get bare LIFEPO4 cells - maybe LiPo if only to be used occasionally, choose the ports / electronics to suit your needs, and buy a Pelican knockoff box to enclose it.

Care to comment?

[smac999]

Testing a Victron setup today. That’s the 150 batt. Pretty easy wiring. Bms, charge bus, load bus. And I made the (non victron) inverter / charger Shut off via either bus. Still missing the dc to dc charger To change bank from engine bank. leaving the int reg alt and limited amps via dc - dc charger.

Acr is engine batt to thrust batt. Not part of victron system.

Actually one wiring error in that pic I fixed after :/

[CatNewBee]

Originally Posted by smac999

Testing a Victron setup today. That’s the 150 batt. Pretty easy wiring. Bms, charge bus, load bus. And I made the (non victron) inverter / charger Shut off via either bus. Still missing the dc to dc charger To change bank from engine bank. leaving the int reg alt and limited amps via dc - dc charger.

Acr is engine batt to thrust batt. Not part of victron system.

Actually one wiring error in that pic I fixed after :/

You mean the shunt and the solenoid on the right, and the very thin wires to the battery, and the missing fuse on plus upstream?