Marine battery compartment: don’t place lead acid and lifepo4 in the same box…

[Substrate]

aGMs are completely unneccessary and poor choices for boats. The PSOC issues common on boats kills AGMS faster then any other batteries. If it’s a military jet use AGM , if it’s a boat use ordinary sealed wet lead acid , this is especially true for starters where the AGM is ridiculous overkill

That's salesman kind of talk. Well, perhaps you are right for those who are just slap-n-go, don't need no knowledge. I'll just replace my fla's while abusing them all the same.

Heh, how many know the proper placement of fla's in a marine environment to not expose the plates?

Installation & Orientation of Flooded Batteries on Boats - Marine How To

Port to Starboard Orientation This photo is simulating a sailboat heeling with flooded batteries. For years I’ve wanted to write this article, but until my friend Dave let me shoot this polycarbonate cased UPS battery, it was going to be tough to explain. As

marinehowto.com

I'll stop, because my point is, it is KNOWLEDGE, not chemistry that decides what is a good or poor choice. And we'll never come to agreement. And that's ok.

I'm just saying sweeping generalizations are a red-flag regardless of chemistry.

 

[Goboatingnow]

That's salesman kind of talk. Well, perhaps you are right for those who are just slap-n-go, don't need no knowledge. I'll just replace my fla's while abusing them all the same.

Heh, how many know the proper placement of fla's in a marine environment to not expose the plates?

Installation & Orientation of Flooded Batteries on Boats - Marine How To

Port to Starboard Orientation This photo is simulating a sailboat heeling with flooded batteries. For years I’ve wanted to write this article, but until my friend Dave let me shoot this polycarbonate cased UPS battery, it was going to be tough to explain. As

marinehowto.com

I'll stop, because my point is, it is KNOWLEDGE, not chemistry that decides what is a good or poor choice. And we'll never come to agreement. And that's ok.

I'm just saying sweeping generalizations are a red-flag regardless of chemistry.

AGMs have beneficial characteristics that are hard to exploit on a boat and have deficiencies ( especially psoc ) that are all too easy to generate on a boat.

Hence it makes no sense in the vast vast majority of cases to deploy them and the marine sales people are guilty of over selling

A good quality sealed wet acid is by far the best compromise of features against price.

I have all the tech arguments if you want to go here But I’ve seen more AGMs ruined on boats by PSOC issues much to the dismay of the owners who paid big bucks thinking they were buying the best.

AGMs are great if your flying an F15 mind you.

 

[Substrate]

I have all the tech arguments if you want to go here But I’ve seen more AGMs ruined on boats by PSOC issues much to the dismay of the owners who paid big bucks thinking they were buying the best.

It is a people problem, not technical. The average boater / renter, doesn't want to KNOW how to avoid PSOC. Salesman just sell them 4 batteries that aren't fully charged, and slap em together, blowing their tops. Dang AGM! Obviously it's the tool, not the owner!

It's too much maintenance for the new agm owner to actually charge each one individually before putting them into service in series / parallel, and then set up the solar panel *to the proper voltage* to get that last 1% in.

The real reason for fla is that you can abuse them - just pour more water in as the cells become unbalanced with improper charging and high-resistance infrastructure.

I mean c'mon - how many boaters with fla have a hydrometer on board to measure the SG?

 

[Skypower]

It is a people problem, not technical. The average boater / renter, doesn't want to KNOW how to avoid PSOC. Salesman just sell them 4 batteries that aren't fully charged, and slap em together, blowing their tops. Dang AGM! Obviously it's the tool, not the owner!

It's too much maintenance for the new agm owner to actually charge each one individually before putting them into service in series / parallel, and then set up the solar panel *to the proper voltage* to get that last 1% in.

The real reason for fla is that you can abuse them - just pour more water in as the cells become unbalanced with improper charging and high-resistance infrastructure.

I mean c'mon - how many boaters with fla have a hydrometer on board to measure the SG?

 

[Substrate]

Sorry I'll stop on a lighter note. LiFePO4 for experimental aircraft. With redundant bms and vents for indoor-cabin use too!

Lithium Batteries For Experimental Aircrafts | EarthX

Our powerful and lightweight batteries are designed specifically for your aircraft starter or backup battery needs. Shop today.

earthxbatteries.com

lead-acid (fla or agm) bores me now.

 

[Skypower]

It is a people problem, not technical. The average boater / renter, doesn't want to KNOW how to avoid PSOC. Salesman just sell them 4 batteries that aren't fully charged, and slap em together, blowing their tops. Dang AGM! Obviously it's the tool, not the owner!

It's too much maintenance for the new agm owner to actually charge each one individually before putting them into service in series / parallel, and then set up the solar panel *to the proper voltage* to get that last 1% in.

The real reason for fla is that you can abuse them - just pour more water in as the cells become unbalanced with improper charging and high-resistance infrastructure.

I mean c'mon - how many boaters with fla have a hydrometer on board to measure the SG?

You know how many times I’ve responded to a “My engine’s won’t start” and upon investigation of the wet cells, I pull the caps and see dry plates? No maintenance and/or bad charger. The funny thing is, the more effluent the owner, the less they care or don’t want to spend the labor to have it done. I try to show them how keep things seaworthy but they just want to do something else or tune me out. Two years ago a large yacht had a very bad “drip less” shaft seal. I informed the owner to not operate that engine and bring it immediately to the nearby boatyard. I could tell he wasn’t going to do anything about it or thought I was over reacting. I wrote a report and gave him a copy and emailed a photo of the report to myself. A few weeks later his attorney informed me his boat sunk as he was leaving harbor. He said I inspected it. The salvage company said it was the shaft seal. I showed the email to the attorney and he said have a nice day. Always CYA with the public!

 

[brb58]

When I got call backs within a few years with LA, sealed or not and all the AGM’s I’ve installed are still going. The results, at least to me speaks volumes. I’ll stick with results. You do you & I’ll do me. Have a nice day

I have an AGM on my pontoon that I now use as a house battery and it was used as a starting battery for 5 years on another boat starting a 7.4L engine. It has a total of 10 years on it now. It is an Odyssey. Expensive battery.

I will not use any other type of battery for a car or boat.

 

[Luthj]

There are quite a few marine applications where a modest sized lead acid battery in parallel with the LFP (after the disconnect BMS), can provide significant protection from unexpected disconnect, without the cost of a redundant LFP pack. They don't carry any of the load generally though.

Some common industrial lead acid flooded batteries have optional vent caps, which vent to a common tube, and some even include automatic watering. Thought I wouldn't ever put an LFP pack (or any sensitive electronics) in the same battery box as a flooded battery.

 

[Goboatingnow]

There are quite a few marine applications where a modest sized lead acid battery in parallel with the LFP (after the disconnect BMS), can provide significant protection from unexpected disconnect, without the cost of a redundant LFP pack. They don't carry any of the load generally though.

Some common industrial lead acid flooded batteries have optional vent caps, which vent to a common tube, and some even include automatic watering. Thought I wouldn't ever put an LFP pack (or any sensitive electronics) in the same battery box as a flooded battery.

It’s not a good idea to parallel different chemistries and can have dangerous consequences.

The new ISO standard on lithiums in boats specifically forbids it.

“4.9 In normal operation, different battery chemistries should not be connected in parallel or in series. “

Most boat Li systems are “ hodge podges “ of components often wired together by people with “ a little knowledge is a dangerous thing “. Standards hopefully will slowly eliminate poor technical solutions

 

[Luthj]

It’s not a good idea to parallel different chemistries and can have dangerous consequences.

The new ISO standard on lithiums in boats specifically forbids it.

“4.9 In normal operation, different battery chemistries should not be connected in parallel or in series. “

Most boat Li systems are “ hodge podges “ of components often wired together by people with “ a little knowledge is a dangerous thing “. Standards hopefully will slowly eliminate poor technical solutions

I am sure there is a reason for that spec, but its not the be-all or end-all of design considerations. Several authoritative sources I have encountered over the years break down the details of such a configuration, and even recommend it for systems where redundant battery packs are not feasible.

Here is one example.

Electrical Design For a Marine Lithium Battery Bank | Nordkyn Design

Electrical design considerations for the integration of lithium battery systems on board marine installations.

nordkyndesign.com

As far as dangerous consequences, as long as normal precautions are taken, and the system is appropriately labeled, I am not aware of any unique issues. If you have a specific concern, please detail it for us. I suspect the reason for the generic recommendation in that spec, is referring to different Lithium chemistries, which are much more sensitive, especially NMC, than lead acid and LFP.

Typically this method uses a ~50AH lead acid battery to provide voltage surge suppression, and allow for a graceful shutdown of the copious sensitive gear in many small marine applications. For example you are night sailing, and the BMS on the LFP pack drops out due to some random nuisance fault. The lead acid battery which is still connected to the bus gives you a few minutes to figure out if you can safely bypass the BMS with the override switch, and suppresses any voltage spikes which may result, preventing damage to equipment.

Under almost all normal operating conditions the LFP pack will have a higher voltage then the lead pack, and will supply nearly all load current. The lead battery will have a small parasitic draw from the LFP pack due to cross charging, but that will typically be around 500ma per 100AH of lead capacity, sometimes less.

 

[Goboatingnow]

I am sure there is a reason for that spec, but its not the be-all or end-all of design considerations. Several authoritative sources I have encountered over the years break down the details of such a configuration, and even recommend it for systems where redundant battery packs are not feasible.

Here is one example.

Electrical Design For a Marine Lithium Battery Bank | Nordkyn Design

Electrical design considerations for the integration of lithium battery systems on board marine installations.

nordkyndesign.com

As far as dangerous consequences, as long as normal precautions are taken, and the system is appropriately labeled, I am not aware of any unique issues. If you have a specific concern, please detail it for us. I suspect the reason for the generic recommendation in that spec, is referring to different Lithium chemistries, which are much more sensitive, especially NMC, than lead acid and LFP.

Typically this method uses a ~50AH lead acid battery to provide voltage surge suppression, and allow for a graceful shutdown of the copious sensitive gear in many small marine applications. For example you are night sailing, and the BMS on the LFP pack drops out due to some random nuisance fault. The lead acid battery which is still connected to the bus gives you a few minutes to figure out if you can safely bypass the BMS with the override switch, and suppresses any voltage spikes which may result, preventing damage to equipment.

Under almost all normal operating conditions the LFP pack will have a higher voltage then the lead pack, and will supply nearly all load current. The lead battery will have a small parasitic draw from the LFP pack due to cross charging, but that will typically be around 500ma per 100AH of lead capacity, sometimes less.

Sorry for a start boats subject to standards often via insurance conditions cannot flout such rules

Secondly leaving a lead acid in parallel constantly in my view is vey dangerous and I would question the whole validity of that link


A 90-100 % LFP will have a typical voltage of 13.8 ( or even slightly less) this is very similar to SLA terminal voltage.

If the SLA Soc gets out of sync large and dangerous currents can flow from it or to the Li.

Secondly the charging regime set for the li is then wrong for the LA.

ISO and many lithium supplier forbid paralleling chemistries.

The whole “ nonsense “ stems from. Complete misbelief that your li pack is unreliable. It’s way way MORE reliable then your SLA.


From battle born batteries

“For battery safety, we do not recommend combining different types of lithium batteries and lead-acid batteries. This is because the load characteristics and capabilities of these batteries are drastically different which can lead to safety issues.”

 

[Luthj]

If the SLA Soc gets out of sync large and dangerous currents can flow from it or to the Li.

I can do the math for you, but I don't have the time or motivation right now. We are talking about a small lead acid battery, which by their basic characteristics, are current limited. Even if one of was totally dead, internal resistance, and wiring resistance, will easily limit current. Now two large battery banks is another story, and that is not what we are talking about. Again, I am not suggesting using two large banks to share loads. This doesn't make sense for numerous reasons, and there are some moderate safety concerns.

Secondly the charging regime set for the li is then wrong for the LA.

The charging voltage of LFP is totally fine for a small standby lead acid battery. Float around 13-13.3V, and absorb to 14-14.4V. The lead acid battery won't normally be cycled, so it doesn't need an extended absorb.

The whole “ nonsense “ stems from. Complete misbelief that your li pack is unreliable. It’s way way MORE reliable then your SLA.

This is a gross misrepresentation of reasoning and purpose of the suggestion. LFP packs have numerous failure modes which can result in a complete disconnect of the battery from the bus. This results in zero power. A lead acid battery will almost never fail this way. A main fuse might blow, but thats to protect against a dead short in a high current circuit anyways.

Lead acid batteries almost always fail gradually. The operator has some warning, as system voltage will sag, and a warning buzzer/light can easily be configured. Similar things can be done with LFP, but not all conditions will provide warning before the pack drops out.

Well built LFP packs are reliable, and provide superior service in many applications. They are not foolproof, and anything from corrosion to a BMS internal failure, can cause them to drop out. Heck even a single event upset caused by cosmic rays flipping a memory bit, can cause the system to drop out.

 

[Substrate]

I am sure there is a reason for that spec, but its not the be-all or end-all of design considerations. Several authoritative sources I have encountered over the years break down the details of such a configuration, and even recommend it for systems where redundant battery packs are not feasible.

I'll tell you the reason. And why perhaps you should carefully evaluate your "authoritative sources". Most are missing real-world exceptions and rely solely on mind-logic. YOU might get away with it, but there is a reason from a safety standpoint in the real world.

You have claimed that lead-acid is self-limiting. Not really, but yes, if you accidentally charge a conventional AGM by slapping it to an LFP, which provides current well beyond the agm's max rush current rating, what happens is that it reaches the CV point real real fast. And it sure *looks* like it is self-limiting as it immediately goes into absorb and a lowering of current. The chemical reaction is so inefficient though, that the result if you do a capacity test you realize that most of that quick charge is a superficial surface charge. But that's not the concern - safety is:

One factor that is overlooked is that by charging an AGM (accidentally or otherwise) well beyond it's max current rating, is that by nature, AGM's use very-thin plates spaced closely together. This is what give them their superior lower internal resistance. However it reduces their true deep-cycle capability by being constructed this way.

So what happens is that for that brief moment when the the AGM goes to the CV terminal voltage lickety split due to over-current, it is just long enough for those thin plates to short. It only takes a minor touch. Pure-lead agm's are another story.

And now you have an agm linked to your LFP, that has an internal short. Given time, this shorted-cell batt goes into THERMAL RUNAWAY.

This is the kind of stuff that those authoritative resources never spell out because they are only using mind-logic, or it has never happened to *them*. But it could easily happen to you.

The problem with nearly all lead-acid discussions is that it is far too easy to get a paper education by reading the interwebz, and never testing or encountering the hazards yourself.

Be safe and follow the safety guidelines. They actually do know what they are talking about and are trying to save yes, even the kludge guys from themselves.

 

[Hedges]

The funny thing is, the more effluent the owner, the less they care or don’t want to spend the labor to have it done.

Freudian slip again?
Or is that just what you think of (certain) customers?

 

[Goboatingnow]

I can do the math for you, but I don't have the time or motivation right now. We are talking about a small lead acid battery, which by their basic characteristics, are current limited. Even if one of was totally dead, internal resistance, and wiring resistance, will easily limit current. Now two large battery banks is another story, and that is not what we are talking about. Again, I am not suggesting using two large banks to share loads. This doesn't make sense for numerous reasons, and there are some moderate safety concerns.



The charging voltage of LFP is totally fine for a small standby lead acid battery. Float around 13-13.3V, and absorb to 14-14.4V. The lead acid battery won't normally be cycled, so it doesn't need an extended absorb.



This is a gross misrepresentation of reasoning and purpose of the suggestion. LFP packs have numerous failure modes which can result in a complete disconnect of the battery from the bus. This results in zero power. A lead acid battery will almost never fail this way. A main fuse might blow, but thats to protect against a dead short in a high current circuit anyways.

Lead acid batteries almost always fail gradually. The operator has some warning, as system voltage will sag, and a warning buzzer/light can easily be configured. Similar things can be done with LFP, but not all conditions will provide warning before the pack drops out.

Well built LFP packs are reliable, and provide superior service in many applications. They are not foolproof, and anything from corrosion to a BMS internal failure, can cause them to drop out. Heck even a single event upset caused by cosmic rays flipping a memory bit, can cause the system to drop out.

An LFP battery well built will be far less likely to disconnect from your battery feeds then any lead acid. Li is far more rugged then lead scid, more tolerant of poor usage and more resistant to decay or mid term failure.

There should be NO reason LFP battery disconnects should EVER occur in a well engineered LFp system. Much more likely a lead acid will fail first and create a shorted cell.

For safety don’t parallel chemistries. It’s not recommended by companies like “ battleborn batteries “ and Manly batteries for example ( and many others ) and is expressly forbidden by the new ISO standard for lithium on boats for example

 

[Luthj]

An LFP battery well built will be far less likely to disconnect from your battery feeds then any lead acid

Engineering for failure courses are a goldmine, and they teach very differently.

A lead acid battery cannot disconnect from the bus. They can gradually fail, or become discharged. No matter what you do to a LFP pack, it will always have a disconnect system which relies on digital logic, and a variety of components. All of which are less robust than a simple wire link. No matter how much engineering is put into such a system, it will always have more possible failure points with regards to a sudden, unannounced disconnect.

There should be NO reason LFP battery disconnects should EVER occur in a well engineered LFp system.

Then why do LFP packs have disconnects at all?

There is no reason a well engineering airplane should ever crash. Therefore, seatbelts, escape slides, etc, are not necessary. There is a reason that aircraft have triple redundant systems in many cases. For aircraft operating over water, the same mechanic can't service both engines is another example.

Reality doesn't care how much we engineers try to plan. Failures happen. Human causes, unexpected events, just pure entropy.
Single event upsets caused by cosmic rays are a real thing, and cause occasional rare errors in microprocessors. It only takes one bit flipped to result in the pack disconnecting because the variable for max allowed voltage is now 10 volts lower than it should be.

The best solution is to have two independent batteries on the same bus, that way if one drops out due to corrosion, a loose wire, human error, mice, etc, the other will continue to supply power.

If I was operating a modest sized boat off battery power, I wouldn't trust a single LFP battery to provide life sustaining electronics/systems during a heavy storm.

 

[Luthj]

And now you have an agm linked to your LFP, that has an internal short. Given time, this shorted-cell batt goes into THERMAL RUNAWAY.

A properly fused system would blow a fuse. Not to mention the BMS should drop out.

Its a moot point, if you have more than one battery, the same thing can happen with two LFP packs.

One factor that is overlooked is that by charging an AGM (accidentally or otherwise) well beyond it's max current rating, is that by nature, AGM's use very-thin plates spaced closely together. This is what give them their superior lower internal resistance. However it reduces their true deep-cycle capability by being constructed this way.

I have over 20 years experience with every type of lead acid out there. There is no single design spec for AGM lead acid variants. Many AGMs have superior deep cycle performance and durability (lifeline for example). Physical support from the glass mat actually prevents plate damage in many cases.

The problem with nearly all lead-acid discussions is that it is far too easy to get a paper education by reading the interwebz, and never testing or encountering the hazards yourself.

You will notice I have not made any claims about your experience or knowledge. I would appreciate it if you kept your unsubstantiated speculation about mine to yourself.

even the kludge guys from themselves.

Is that a habit of yours? When someone disagrees with you, you resort to schoolyard insults? Thus far all the world ending issues you have described are either non-issues, the same for LFP, or are speculatory hyperbole.


I have personally seen big lead acid banks meltdown, and the aftermath of lead acid thermal runaway. We aren't talking about big banks though, a small lead acid battery isn't an issue. Thermal runaway risk of lead acid is directly tied to the surface area to mass ratio of the pack, ambient temperature, and the stabilizing alloy used in the plates.

You have claimed that lead-acid is self-limiting. Not really, but yes, if you accidentally charge a conventional AGM by slapping it to an LFP, which provides current well beyond the agm's max rush current rating, what happens is that it reaches the CV point real real fast. And it sure *looks* like it is self-limiting as it immediately goes into absorb and a lowering of current. The chemical reaction is so inefficient though, that the result if you do a capacity test you realize that most of that quick charge is a superficial surface charge. But that's not the concern - safety is:

That was a lot of dissembling to admit that its self limiting. Take a look at the internal resistance chart for a lead acid battery. Its very high at both ends of the SOC curve. If I thought it would be effective, I could link a couple of studies on dump charging lead acid batteries.

We aren't talking about a 500lb bank. A single <40lb lead acid battery as a buss buffer in this case.

 

[curiouscarbon]

BMS shutting down is the ultimate "load shed" i.e. 100%

lead acid cells should be on a BMS when cell life assurance is desired. 50% depth of discharge can be enforced by MOSFET or Relay/Contactor safety disconnect.

convenient load shedding can delay final BMS disconnect.

the culture i have seen is:

lead acid cell operators rarely use BMS to protect cells
lithium cell operators frequently use BMS to protect cells

this all begs the question to me,

"if an operator is comfortable with a lead acid battery without bms to avoid safety disconnect, why directly compare to a safety disconnect protected lifepo4 battery?"

comparing no bms lead acid to no bms lifepo4 seems fair to me.

comparing no bms lead acid to bms lifepo4 is not fair to in my humble opinion.

this all comes down to whether it is more important to run the equipment or protect the battery. that variable changes the operational priorities quite a bit

 

[curiouscarbon]

some devices have low voltage lock out to prevent excessive discharge (hopefully configurable). this is auto load shedding by voltage threshold.

systems that utilize devices lacking low voltage lockout, may benefit from a main under voltage lock out mechanism, which some use a MOSFET BMS to implement. solenoid relays are also common

 

[Luthj]

BMS shutting down is the ultimate "load shed" i.e. 100%

lead acid cells should be on a BMS when cell life assurance is desired. 50% depth of discharge can be enforced by MOSFET or Relay/Contactor safety disconnect.

convenient load shedding can delay final BMS disconnect.

the culture i have seen is:

lead acid cell operators rarely use BMS to protect cells
lithium cell operators frequently use BMS to protect cells

this all begs the question to me,

"if an operator is comfortable with a lead acid battery without bms to avoid safety disconnect, why directly compare to a safety disconnect protected lifepo4 battery?"

comparing no bms lead acid to no bms lifepo4 seems fair to me.

comparing no bms lead acid to bms lifepo4 is not fair to in my humble opinion.

this all comes down to whether it is more important to run the equipment or protect the battery. that variable changes the operational priorities quite a bit

The major difference is that a lead acid battery can self balance its cells during absorb through the shuttle reactions. Lithium batteries cannot do this. So balancing is needed, and a BMS can perform this.

Lithium cells do not self limit their voltage during absorb. So if a single cell exceeds the max safe voltage, the electrolyte decomposes, causing swelling, possible venting, and internal shorts/fire. Lead acid does not have this issue.

The next difference is that lead acid batteries, while they can be degraded by discharging to empty, will recover, and continue to safely operate. Lithium batteries can be destroyed if allowed to completely discharge, and can result in cell shorts/fire/failure during subsequent recharge.

With a lead acid battery, an external voltage sensing relay can be used to prevent over discharge, and shed any non critical loads. These are pretty robust units, and failures are rare.


With lithium cell level monitoring is required. This requires quite of a bit of wiring, cell connections, and a logic board.