I was just heaving a huge sigh of relief that I was finally finished with lithium batteries, at least for a while, when two members asked my opinion on adding lithium in parallel to an existing lead-acid system. And that’s gotta be the fifth time in the last year. Time to write an article.
And, hey, I get it. After reading five highly technical chapters of stuff we need to know before plunking down money for lithium, a shortcut would look attractive to anyone.
The good news is that this article is short since I have already written extensively on the underlying reasons for most of the points I will make.
So you can either just take my word for it by reading what follows and move on, or dive deeper by reading the linked content as well.
Update July 13th
To be clear, my thoughts on why lithium add on systems that rely on connecting lithium and lead acid in parallel without isolation are a bad idea apply to all, not just the product I mention several times in what follows.
I fitted these lithium extender batteries https://www.buildsolar.com.au/products/bos-le300-lithium-extension-battery in parallel with my AGM batteries. They are quite effective as they discharge first and charge last, reducing the cycles on the AGMs. 3 of them provide enough power for overnight sailing (autopilot, lights, navigation, fridge) and they charge up quickly so I figure I’m getting the best of both worlds
Those units are tiny, only 25 Ah each. 3 of those units is approximately equal to one relatively small lead acid battery. I didn’t see anything in brochure that provided any information regarding actual technology used or safety considerations.
Hi Richard,
I have looked at that solution before and it’s certainly an option, but, as I explain above, using it means not realizing many of the major benefits of a lithium installation. Also, given that they cost US$330 each for 26 Ah at 12 volts (322 Wh) it’s an expensive way to get the job done. For example our AAC standard example lithium bank size of 4800 Wh it would require 15 modules at an nearly US$5000, as against $4192 for a Victron system with top of the line external BMS and monitoring built in.
And of course the Victron system would take up a fraction of the space and weigh far less.
And then there is the issue of ABYC compliance and insurance which I think might be doubtful with the add on modules, at least compared to the Victron solution.
So, in my opinion, while it might have made sense when you installed it, and for your usage profile, I can’t recommend it for others today.
Hi John, I agree it’s not for everyone but I think it does provide the main benefit of lithium (fast recharging) – also the price is closer to US 200$ – not 330$, that was AUD. I will change to completely Lithium in the next few years so I’m reading your articles with great interest
Hi Richard,
Good hear they are less expensive than I thought.
As to faster charging, that’s actually not true in many cases. For example the LE300 lithium packs are limited to a max charge current of 12.5 amps which is .5c but lifeline AGMs will happily lap up twice that until about 70% charged and the same is true of maximum discharge current. See my answer to Evan for more on why that matters.
Also while looking at the LE300 spec I noted that they recommend at 70-125 Ah of lead for each 28 Ah LE300, which will result in an even heavier total bank than I realized, so I stick by my thinking that this is just not the way to go, at least today and into the future.
Thank you for another good one, John.
I would be interested in your take on the X2 BMS from BatteryBalance in Sweden, which uses separate load and charge busses with high capacity contactors to the lithium bank. A lead auxiliary bank is permanently connected to the charge bus and can be manually connected to the load bus in case of a lithium blackout. The BMS controls charging devices and manages top balancing of the lithium bank.
In our particular case, the AGM bank that supplies the windlass and bow thruster will be connected as the auxiliary.
TIA
Hi Torsten,
I have looked at that BMS before and it looks interesting, that said, for me it has one fatal flaw: it relies on using a contactor to parallel lead and lithium, which, as I explain above, is fundamentally bad practice. And then, as always, there’s the issues of compliance and insurance. So adding it all together I can’t see the benefit when compared to a plug and play compliant system from Victron or maybe Mastervolt that also will be much less likely, I think, to upset our insurance company.
Once again, this is a solution that might have made sense once, but with recent price drops from companies like Victron probably does not now. And then there is the lead to lithium parallel connection without isolation to put the final nail in the coffin, at least to me.
Seems the old saying, you get what you pay for, also applies to electrical systems advice in the internet. The value of free advice is worth about what it cost: nothing! This is why I gladly pay to be part of this knowledgeable community.
My lead batteries needed replaced and I was planning on going lithium, but the complexities of a well-designed system convinced me I’m not ready. So I went with AMGs for now and will continue to study these great articles and discussions.
Hi Ben,
Thanks for the kind words, much appreciated.
There’s a good argument that it’s not you that’s not ready for lithium, but rather lithium batteries are not ready. My thinking is that we are on the cusp of lithium being ready for main stream usage on boats, so it can be argued either way, but there is no question that lithium will get less expensive and easier to implement going forward, so there are definitely good reasons for waiting.
Hi John,
I think your analysis of the parallel Lead-Acid / LFP option is a very good article. Thanks.
I feel that your Point #1 though is not entirely fair. (Yes, I have read the referenced article.)
The increased power capacity for given weight and size of an LFP bank is indeed a big advantage over a Lead Acid (LA) house bank, allowing one to install a lot bigger house bank capacity in the limited space one may have but there are some other very compelling reasons to consider LFP.
These are just a few of the game changing realities of a well designed LFP house bank system.
Yes, the design and installation takes a lot of thought and does involve more complexity; but once it’s up and running, one spends a lot less time and energy worrying about the SOC and managing the health of the house bank. It just works.
As I’ve said before, I do not know anyone who, having switched from lead acid to LFP, would ever willingly go back to LA.
I think your point #1 is too simplistic and short-changes the many advantages of LFP. It’s not just the fact that you have more energy in a lighter more compact bank that puts LFP into the preferred column, it’s a lot easier to live with when out on the water and on the hook.
Hi Evan,
I agree with all your points, but not your conclusion. In my view most-all are actually just contributory (secondary) benefits that add up to the one primary benefit: higher energy density.
For example lead acid will charge at just as high, and in some cases higher currents than lithium—most lithium .5c, lifeline AGM 1c—until they reach about 70% at which point lead acid tail off and lithium do not. But all that means is that more of the rated capacity of lithium can be used on each cycle, which in turn means that a smaller lithium bank will fit a given set of needs, and that means that lithium has a higher effective energy density and so the banks takes up less space and weighs less. Back to my primary reason.
The same applies to all your points to at least some extent, which is why I classed said benefits as secondary.
Also, some of the points are no longer problems for lead acid. For example the current tail off issue is easily solved these days by running the primary charge source, say generator, until lead acid reaches about 70% and then letting solar finish the job. This also solves the problem of sulphating that plagued lead acid for so long.
To be clear, I’m not saying that lead acid is better than lithium, just that when we make the decision between them the criteria are simple, less size and weight for lithium, simpler, and less to go wrong, for lead.
Hi John,
thank you for another very good article.
However, I agree with Evan that the first point is greatly simplified. For me all four Evans points are not due to higher energy density.
After switching to LFP I can recharge my batteries from 20 to 95% in less that 2 hours. My 700Ah Winston bank is taking 300+amps until 95% SOC after which the current begins to decline.
It’s impossible to charge AGM’s@1C. Maybe for 3 minutes. After that 3 minutes you’ll reach bulk voltage @ around 55% SOC (if you start from 50%).
Rod Collins made a test with Lifeline GPL-31T battery. He was charging them from 50% of SOC at 0.2C and 0.4C.
At 0.4C the charger left Bulk stage after only 19 minutes @ only 63% of SOC. So after only 19 minutes you start to waste diesel from the generator, or the engine. I don’t want to rewrite this article. Here is the link https://marinehowto.com/how-fast-can-an-agm-battery-be-charged/
For me the conclusion is
Here’s a short story for those who are still hesitant.
On our yacht Crystal we have 700Ah DIY bank. 4 Winston cells with REC BMS. We have 2 alternators that can charge the bank with constant 300+ amps. I can manually set them to 50% of power(when I need power at low RPMs) or totally disconnect them.
We have also induction stove and electric oven.
If I know that we move next day and we have only 20% in the afternoon I wait with charging until next day.
One day we woke up and SOC was 10%. After breakfast it dropped to 3%. I started the engine (starter is connected to the house bank) and started to raise the anchor (50 meters of chain). After 30 minutes after we left the anchorage I realised that the alternators were disconnected!
So than means that I started the engine and then raised the anchor from almost depleted batteries 😉 And the BMV was synchronized to 100% 3 days before so these values were quite accurate.
So I also agree with Evan that if someone try lithiums there is no way back. I like my induction, electric oven and very fast charging. I like that I can run my 2kW watermaker from the batteries (sometimes for 2 hours) without starting the engine. I lived on a yacht with AGM batteries for 12 years and it was ok, but after another 4 years with LFP I can’t imagine going back to AGM. Life is so much simpler with lithium.
Hi Michael,
Sure, if you go with a very high load scenario like induction cooking, then lithiums are pretty much required, for just the reasons you state. But to then extrapolate that to say, or even imply, that lithium are better for all is, in my view, a mistake.
As to Rod’s testing, I suspect the AGMs were slightly sulphated after six months of use on a trawler since most people don’t condition nearly enough. I have huge experience with Lifeline AGMs and have often observed that if sulphated, even slightly, they will go into acceptance way too early, in fact that’s how I knew that my batteries needed equalizing. And liquid filled lead acid don’t suffer from that problem as much, nor do gell.
Anyway, even if Rod’s testing is bang on, I agree that with most usage scenarios charging over about .3c is not required. And there is no question that all lead acids tail off after about 70% charged, as I have written many times.
That said, when people make blanket statements that are wrong, like “lithium accept more charge current than lead acid” without qualifying what part of the charge cycle we are talking about, I think it’s important to correct the facts so each of us can evaluate our needs properly rather than being deluged by things that are not quite true by the fans of either camp, who are often trying to “prove” their way is better.
There is no best way, just best for each of us and my job is to make getting to that best as clear a path as possible for each reader, which I think zeroing in on energy density does well.
Hi Michael,
One more thought. I did not say that all Evan’s points are due to higher energy density, rather all of Evan’s points, which I said I agreed with, result in higher energy density. The difference in wording is subtle, but the difference in meaning is huge.
You might want to read my original article on this to see what I mean, and also to see the other benefits I accord to lithium : https://www.morganscloud.com/2023/01/10/a-simple-way-to-decide-between-lithium-or-lead-batteries-for-a-cruising-boat/
Thank you for the great examples of daily use Michał.
The reality of LFP storage is that you can access the energy and put it back far more easily and efficiently than with LA.
Sorry to sound like a fan-boy but LFP provides a profoundly significant change in how you live on the boat for extended cruises off-grid. The energy management is no longer such a source of concern and hassle. It is radically better.
I agree with both commenters above. Point #1 is oversimplified.
The uptake of energy over the useable charge range (roughly 20% to 95% SOC in a lithium battery) is much more efficient (both from a time perspective and an energy in/energy out perspective). Lithium’s charging cycle occurs at a steady, predictable 0.5C across a wide, useable range. We’re looking at the total energy (area under the charging curve, not just a specific region under the curve where charging current demonstrates particularly good performance.
AGMs may be able to take up energy at a faster rate to 70% SOC, but this is only about 20% of the useable energy total (assuming you cycle your AGMs down to 50% SOC).
Additionally, in my experience, lithium batteries stand up to charging neglect much better than a lead chemistry deep cycle battery. Lead is prone to sulfation. If lead-based batteries are allowed to discharge too far, they can be irreparably damaged. This can occur from simply sitting for too long without being attached to a trickle charger. This has stymied me on a couple, different occasions. Once when I had a DC electrical leak that took me a few weeks to find while the boat sat on its mooring (eating a couple of lead acid batteries in the process), and again recently when I went to use my small, Duracell SLA battery that I use to power a fishing light. It sat too long off a trickle charger and died forever.
LiFePO4 batteries stand up better to partial charging cycles, deep cycling, and resting for long periods of time. It’s important to setup the system initially with a mind toward safety and follow recommended best practices (note: connecting them to different chemistry batteries isn’t a recommended best practice, but I can appreciate people experimenting and reporting their experience, given that they take precautions to manage the risks).
After the system is upgraded safely, LiFePO4 batteries live a much simpler life. They just perform without a lot of babying. John, I noticed that you mentioned below that “most people don’t condition [their batteries] nearly enough”. That’s part of the point: to maintain high performance out of lead chemistry batteries, you need to keep a close eye on their SOC, charging parameters, and maintenance schedule. This is much less of an issue for LiFePO4 batteries.
Hi Geoff,
Pretty much everything you say I agree with, but I don’t agree that I over simplified. Everything you point out is simply a an explanation of the basic fact that lithium have about 400% energy density advantage over lead acid that I clearly explain right at the beginning of the article, which I suggest you read: https://www.morganscloud.com/2023/01/10/a-simple-way-to-decide-between-lithium-or-lead-batteries-for-a-cruising-boat/
The one place I disagree with you is the idea that lead acid is more vulnerable to damage. Given that lithium batteries can be destroyed by one single over or under discharge, whereas lead acid can put up with an amazing amount of abuse, although certainly not unlimited, lithium are way more vulnerable, in my eyes.
As I keep saying, I don’t care what anyone does, but when proponents of either side bend the facts to make a point I will step it.
The idea here is to move on from “this is better than that”, to these are the facts to help each of us make the right selection for our own needs and wants.
Hi Geof,
One more disagreement. The popular idea that lead acid batteries quickly self discharge and therefore must be kept on trickle charge was proved to be wrong years ago. Typically a fully charged lead acid battery will be perfectly happy for several months without charging. This varies a bit by type, for example a lifeline AGM should probably be recharged about once every 6 months when not in use, but should never be left on trickle charge. The optimal recharge frequency also varies by temperate, but it’s months, at least, for all types.
Sure, if we leave a small load on a lead acid battery for weeks and don’t recharge it, we will damage it, although it can often by brought back by reconditioning.
On the other hand if we discharge a lithium just once past about 20% it’s toast. For example, Victron specifically warn against the dangers of leaving a lithium bank connected to a battery monitoring system that bypasses the BMS low state of charge cut off, as many do.
Point being, both chemistries can be damaged by bad practice. The nature of the bad practice is just different.
Hi John,
Thanks for your thoughtful responses and, in general, engaging with me on this issue. I think that the ability to read the comments and talk about these issues adds a huge amount of value to the AAC experience.
I read the article that you linked, and I think you do a great job laying out the considerations between each battery type. I think where you, me, and the couple posters before me are disagreeing is about the overall lack of maintenance and ease-of-use (due to the LiFePO4 chemistry and how well it takes to short-charging and very deep-cycling. This is partially related to energy density, but also partially related to other factors) and the robustness that comes from having a BMS manage the the battery. To your point above, Victron is warning against bypassing the BMS.
As you know, the BMS is critical in these battery systems. It can be your best friend or your worst enemy. I have generally been very skeptical about cheap LiFePO4 batteries because I’m a fairly strong believer that, especially in the marine world, you get what you pay for. With high-end Lithium batteries, you’re not only paying for top-notch cells, but also a top-notch BMS system.
I can understand anyone’s skepticism toward the BMS (a computer) and how this can seem like a recipe for disaster with the LiFePO4 chemistry’s more inherent proclivity for damage should the computer fail. In practice, I don’t think that people are finding this to be commonly (or uncommonly) the case. I certainly haven’t. There’s always edge cases of failures, and I think it’s important to consider how different types of edge cases will be handled if you’re going “way out there”, but I think that vast, vast majority of LiFePO4 battery users are finding after racking up some years on them that the BMS is so much better at managing/protecting the batteries than a human.
A quick, personal story. I recently lived aboard only 200 AH of LiFePO4 batteries for a year and a half. At one point, during the winter, I left the boat to visit a friend out of state, leaving the boat for a week. At one point during that trip, the older charger I had on board failed and the solar setup I had at the time couldn’t generate enough power to keep the batteries fully charged while running my refrigerator and freezer up here in the Northeast during the winter. The weather had been cloudy during a couple of the days that I’d been gone, and my lacking solar setup at the time just couldn’t keep up. When I returned to the boat, I looked at the battery charge history through my Victron MPPT charge controller. The data told the tale: after the main shore power charger failed, solar would take the batteries up to 60% charge (or so) during the day. During the night, the batteries would nearly or totally flatline. The BMS kicked in at least a couple of times overnight to protect the batteries. When I returned back to the boat, I cycled the power to the charger off and on, resetting the charger, and quickly brought the batteries back to full charge. No damage was done. If this was a lead chemistry battery, I don’t think that things would have gone so smoothly.
There’s all kinds of things I could have done better in this story, but the point is that the BMS saved my butt and the batteries soldiered on. Life happens when living your life aboard. Since then, I’ve run into the BMS shutdown limit a couple of other times on anchor or mooring, mainly because I spend so little time worrying about battery SOC unless I’m relying on them to power instruments underway.
I don’t want this to come across like I’m disparaging AGM batteries or saying one is “better” than the other. I’m simply taking issue with boiling the issue down to an energy density issue, which is only partially true.
My best sailing friend has used AGM batteries for a significant portion of his 20-year sailing career. He knows them, understands them, and is comfortable with them. They can serve a sailor well. I recently guided him through his first solar installation. It’s been really fun to watch his excitement at how freeing a solar installation is. The journey was less exciting for me, as it was more a matter of necessity since I telework and, being off the grid, needed power in order to hold my job down. As you mention in your article, a quality solar install is a truly fantastic thing for AGM batteries, and a cruising boat’s energy management in general. All of that being said, when conditions are less than perfect and “life happens”, LiFePO4 managed by BMS has proven extremely robust as I’ve lived my life on the sea, off the grid.
Hi Geoff,
As I said in the note I added to the above article, I’m getting sick of all these lithium fan boy comments that are totally off the topic of the article and therefore I will not further engage. I will also delete any further comments that are off topic.
Only one reason it is essential to install a lead acid battery in parallel with lithium: Failure Mode.
When failures happen, nothing can beat the pure physics and simplicity of copper protected by a properly rated fuse.
Any other arrangement will be more complex and will introduce new and often unpredictable failure modes. Especially anything with electronics.
Install that new lithium system with a fancy BMS but realize it may fail. Install that impressive alternator controller but realize it too may fail. Hopefully you have one that fails leaving the alternator to still function in its original manufacturer designed simple voltage feedback mode.
If that lithium BMS disconnects or it’s class T fuse blows and the alternator controller fails then your alternator blows up.
Unless you have a lead acid battery directly wired in parallel via a proper fuse.
Hi Kelley,
I disagree, there’s a way better way to have lead backup without the fundamental negative issues of paralleling lead and lithium: https://www.morganscloud.com/2022/07/03/building-a-seamanlike-lithium-battery-system/
Also, while a parallel lead will protect against alternator damage, there are much better ways to solve that problem: https://www.morganscloud.com/2024/01/29/lithium-batteries-buyers-guide-part-1-bms-requirements/
https://www.morganscloud.com/2022/01/30/wakespeed-ws500-best-alternator-regulator-for-lead-acid¹-and-lithium-batteries/
And finally, modern alternators with avalanche diodes (most good ones) do not tend to blow up or produce huge spikes when disconnected although it is worth adding an alternator surge suppressor as a further backup.
And if none of that appeals, there is our option 8 here: https://www.morganscloud.com/2024/06/17/lithium-buyers-guide-budget-economy-options/
In summary, while I agree that paralleling can solve these problems, today there are better answers, in my view.
The problem with your alternate solutions is that they are all more complex and add new failure modes that don’t exist at all in the simple parallel hookup with proper fusing.
While lead acid in series seems nice it introduces a new failure point in the dc-dc charger which could cause loss of power to the critical systems. The 1-2 switch could allow power to be restored assuming the alternator wasn’t already damaged when the BMS failed.
The other “better ways” both require more complex controllers or BMS to function properly.
And while “option 8” works, it loses the fast charging ability as you mention. And of course again adds new failure modes and complexity.
In fact, what are “the fundamental negative issues of paralleling lead and lithium”? Could it be that these issues are actually less fundamentally negative than the increased complexity failure modes you suggest?
With proper wiring, sensing and fusing a simple direct parallel lead battery will survive any complex lithium BMS or alternator controller failures. And provide instant power to critical systems without any blackout. And allow safe fast alternator charging. The only possible downside is a lead battery that is constantly maintained at a typical 13.3 volt lithium charge state with some variation between 13 and 14 volts. Basically always topped up on a trickle charge (a small waste of energy). And this matters little since it will never be used except in a failure scenario. See https://nordkyndesign.com/electrical-design-for-a-marine-lithium-battery-bank/
Note that this does not in any way alleviate the requirement to design an excellent lithium and alternator control system. It just provides the dead simplest, least failure prone, most reliable insurance against those new complex systems failing.
John, I think we need some probabilities in here. Like what is the % chance of complex lithium system failure verses a copper wire to a fuse on a lead acid battery.
Hi Kelly,
It’s time to agree to disagree, let’s leave it there. I have already listed the fundamental drawbacks of paralleling in the article above. If you don’t agree with those points that’s fine with me, but I don’t see any point in arguing it further.
John thanks for your time to consider this. I really agree with you almost entirely. I am just unable to set down my Occam’s razor.
Best Kelley
Hi Kelley,
Thanks, a good way to finish up the discussion.
I wanted to leave a comment to say I have a bank manager. The only reason I do is our boat gets hauled out and sits on the hard over cold winters. But while sitting there we have don’t always get a lot of shore power. So I disconnect the house AGM bank. I have fully rewired the system, including an Electromaax 250a alternator and ALL Victron charging equipment. Something interesting about a Multiplus is you can’t get AC voltage unless it’s connected to a 12v battery bank. So, for now I have to leave our AGM bank in place. I have a fully fused and installed 500ah LFP bank that has worked flawlessly with the bank manager this summer. Easily Maintaining the LA, and LFP. The LFP will be removed during the winter even though they have internal heating. It’s just safer. I don’t know if I will keep the bank manager in place when we move aboard full time and cruise to warmer latitudes or set it up direct as Victron expects.
Hi Kevin,
Glad to hear it has worked for you. That said, given you have a full Victron system, I would remove the bank manager and go with a serial backup: https://www.morganscloud.com/2022/07/03/building-a-seamanlike-lithium-battery-system/
I am inclined towards doing just that… when I’m full time aboard OR when the boat is not being left in a cold climate in storage for months at a time (and I need interior AC connection). The only other way I could achieve this is running multiple 15a shore power cords, or building a complete shore power supply system that BYPASSES the multiplus and goes to a switched fused etc AC panel. All because in order to get AC to a system connected to a multiplus (3000/12/120) it has to have a (mostly)charged 12v battery in place. Reaearch and checking with Victron groups this was the consensus. So this bank manager allowed me to have a bigger house bank in LFP and I can remove it for winter and the system doesn’t know or care. Perfect for my unique situation.
Cheers