Lithium NMC batteries - the energy density holy grail ?

[caveman22]

Hi All

Suggested this topic merits a separate thread.

Anyone have experience with the Victron Lithium-Ion HE Batteries ? They are Lithium Nickel Manganese Cobalt Oxide (NMC) technology, energy density at 180kw/kg is outstanding, 24V, air cooled but with only IP22 rating.

Apparently this technology is used in power tool and EV applications.

Yes expensive but if weight matters (and it does to me) possibly worth it. Wondering if they are robust enough for a sailing cat. Certainly marketed as "Marine".

Possible advantages in assembling a full Victron system from Batteries, BMS, to Inverter aside from the hit to the wallet ?

Like to hear your thoughts and any comparable NMC batteries out there.

On a separate but related issue for me, we are designing a 52' performance sailing cat. She will be around 9.0T fully loaded so easily pushed and expected to create enough apparent to sail in even light conditions. Propulsion and galley services all electric. Motors 2 x 15kwh (48V) + house loads DC 24V + 240V Ac via inverter.

Supply side planning a battery bank of circa 18kwh, 16kwh 48V DC Gen set, Solar around 2kw + what ever regen we get. Have sized Battery bank & Gen set to provide 100% for around 40mins on basis this should be enough to get us out of an infrequent but sticky situation where sailing wont do. Acknowledge that the Gen set will not give us enough power to motor continuous at 100% but from experience that requirement is unlikely after all we have sails, a light boat and will sail most of the time.

Am reflecting on battery placement and whether to centralise or decentralise. Desirable to locate batteries as close as possible to load, so with an elec motor in each hull would make some sense to create 2 separate battery banks for EP + one for house loads. Reduces heavy cabling & Vd, provides redundancy but adds complexity & cost -presumably would need 3 BMS, one of each bank + run charge sources to 3 places?

Im also puzzled about how 3 separate banks would interact in terms of relative charging and separate SOC?

Love to hear your thoughts and suggestions.

[john61ct]

Since NMC is much more likely to suffer thermal runaway (sudden fires, very hot), Victron must have good thermal management built into their protective systems to be targeting the marine market.

Check into qualifying for insurance.

Definitely not suitable for DIY, some talk about recycling scrapped EV packs, IMO foolish on a boat.

Note also NMC's cycle lifetime is a small fraction of LFP's, more in line with lead, likely even lower in a propulsion context at high C-rates.

That translates into pretty high long-term costs in comparison, $ per Ah per year.

If the weight issue can be designed away, I'd recommend sticking with LFP.

[nebster]

Originally Posted by john61ct

Note also NMC's cycle lifetime is a small fraction of LFP's, more in line with lead, likely even lower in a propulsion context at high C-rates.

I'm definitely not arguing for non-LFP on a boat, but let's keep the record accurate: the data don't suggest this statement is true, unless by "small fraction" you maybe mean something like 0.8?

It's hard to tease apart perfectly since high-rate definitely drives degradation faster. But the data we do have shows NMC performing well out into 2000 cycles and probably much closer to 4000 cycles, with new gains through the help of recent variants like NMC811.

Even Tesla NCA packs are looking great at 500+ cycles, very flat with ~95% SOH.

I think in a coddled low-rate environment, lifecycle is probably not really a differentiator for these chemistries. Lighting off a blowtorch in your boat, saving a ton of space/weight, and/or saving money are the first-order concerns.

[Boatsie]

I've been wrong about lithium ion. I'm probably wrong again if I comment.

Stoked that 3300 tonne of lithium ion landfill is about to be redirected into ♻
https://www.csiro.au/en/News/News-re...cycling-charge

Next web address might be useful sir.
Regarding charge rates and discharge rates to prolong battery life and make best use of battery energy

https://www.victronenergy.com/blog/2...um-ion-vs-agm/

Lol. A few years old, maybe out of date. A good read though. Just is there

[john61ct]

2-4000 cycles for LFP is manufacturer ratings, under care conditions much more abusive than I'd allow for a House bank, and easily avoided without significantly compromising performance.

IOW a gross underestimate compared to what is easily achieved in real life.

500 is of course barely broken in.

But I agree it's a moot point, since lead or LFP are the only chemistries should be on a boat, unless professionally purpose-built, likely engineered for propulsion.

[Steve_C]

Hi Caveman

Not to change the subject but

I would love to hear more about your project, maybe start a separate thread

I would be interested in the design of the vessel as well as the systems. Also rough budget projections and timeline. I am about 2 years or so away from retirement and have considered the custom route.

As for the battery chemistry, it would take a VERY compelling weight argument for me to forgo the safety of LFP as well as the tremendous amount of experience available to setup and use properly today! For a 18KWH bank of LFP you are at about 450 lbs (204 kg). How much weight could you really save?

Sounds like a great project, best of luck!

[css145hs]

I posted this reply on the LiFePO4 thread, but it probably got lost there, so putting it here too (edited slightly since you shared more info here):

They're being installed in my new cat (actually already installed, but the boat is a month from completion so no actual experience with them). Also a 52' 9T performance cat ;-)

I'm not worried about them being suitable for a marine environment. However, I would call out a few key differences. As you've noticed, they are half the weight of typical LiFePo4 batteries for the same nominal watt-hour capacity, but you can't discharge them below 20% (the HE battery pack will cut you off), so they sorta have just 60% more power per kg. Pricing wise, they are 19% more than the LiFePo4 victron batteries (in the US) per usable watt-hour (excluding the 20%) but 40% lighter per watt-hour.

Another major difference, is that they have a lead acid discharge voltage curve, its effectively linear from 100% -> 20%, not the flat voltage curve we're used to with LiFePo4 batteries. This does not mean that they have the voltage sag of lead acid when you apply load, just that their resting voltage drops with SoC. I personally like this, as its much easier to tell SoC from the voltage (in my experience, shunt monitors aren't as accurate as I'd hope). It also means you can easily program voltage-based cut-offs for inverters, chargers, alarms, generator auto-start, etc. That said, it will have an observable effect on voltage sensitive electronics.

For me, the inverters will hide the voltage sag for AC equipment. Most electronics don't care. However, the watermaker and the winches/windlass will behave differently ~3V lower. I don't care about the watermaker, I turn it on and come back later. For the winches/windlass, yea maybe I'm giving up a little umph, but whatever. If the solar is doing its thing, the voltage should be on the high end anyway.

I'm going 2.8kw of solar on the roof using the 360w sunpower hard panels. Was talked out of the flexible panels as they (even the expensive ones), can have significant efficiency loss in tropics due to no airflow beneath them. We're putting on a tiny FP 5kw genset, and have 15kw of the victron HE batteries (aka 3). It's a 24v boat, with 240v AC 2x5000kva inverters. Also electric kitchen.

Note, I was going to do the oceanvolt 15kw servoprops but we went with 2x40HP Yanmars instead. My builder put in 2 of the OV systems on previous boats and had nothing but problems. Even once that was all resolved, they were underpowered and it was an issue within a few weeks (motoring at 2knots under genset power into a 30kn headwind where an identical boat with diesels was going at 9kn right next to them (they built both of them, only the motor was different)). They were out of battery power on the electric boat, so only had genset output to drive it (I think it was a smaller 5kw genset, so maybe you'd get ~4-5kn with a larger one, but there's simply no comparison to sustainable max power output of diesels. I was going to make the bet that I'd only need higher output for short periods, but it's just a bit too optimistic, and there's not really an advantage. It's very easy, even likely, to start off with a non-full battery, so not even get the "40 mins" of high power output, etc. It costs more, it doesn't really save weight... yes its quiet and there's less maintenance... and while the regen is awesome, I don't know that it matters, I have plenty of power from solar and you're not sailing most of the time, so I tend to just ignore that. Anyway, don't mean to diss your plans, just sharing my journey. I totally agree your plans around a large lithium bank, no propane, big solar, etc, are spot on and definitely the way to go.

Anyway, my 2 cents.

[css145hs]

Caveman, crazy idea, not sure how far along you are in your build, but I'll be doing a Cape Town to Florida passage in June and July... if there's any chance you're interested in joining ;-)

Oh and regarding decentralizing the battery. As I understand it, it's highly advised against, as it's too easy to create an imbalance in resistance, but I'd recommend talking to a professional. We put the batteries and the inverters/chargers together, which is a lot of weight in one spot. You could counter balance with the genset. In our case, we counter balanced with laundry, watermaker, fridges, dive compressor.

[nebster]

Another important difference: these NMC packs have a fan, and so they have an air clearance requirement for ventilation. In some situations, then, their volumetric efficiency will be reduced compared to stuffing in more cells.

If you really are after high density, and you want to eat your cake, too, you can get close to those performance with the right LFP cells today. Of course, then you have a science project where you get to do a ton of stuff yourself. But somewhere in the neighborhood of 200Wh/L and 140Wh/kg appear to be achievable with the right cells.

[css145hs]

Originally Posted by nebster

Another important difference: these NMC packs have a fan, and so they have an air clearance requirement for ventilation. In some situations, then, their volumetric efficiency will be reduced compared to stuffing in more cells.

If you really are after high density, and you want to eat your cake, too, you can get close to those performance with the right LFP cells today. Of course, then you have a science project where you get to do a ton of stuff yourself. But somewhere in the neighborhood of 200Wh/L and 140Wh/kg appear to be achievable with the right cells.

This is true but at typical boat C rates, it should be minimal. We put them into a compartment with ingress/egress air channels, and wired the inverter fan circuit to also trigger the compartment fan circuit (the inverter being a much more likely source of heat generation). We also put temperature sensors/alarms in there, independent of everything else. In my case they are only likely to be stressed when charging, which could push 0.7C if solar + genset + alternators were all at max . My discharge rates are unlikely to exceed 0.2C, and charge rates are more typically also at 0.2C. There is certainly more to plan for than non-NMC chemistries.

[john61ct]

I wonder if the ABYC working group spec'ing non-lead standards is including NMC batts like this.

The rate this tech changes, they'll need to issue frequent updates after they publish v1.0

8-)

[nebster]

Originally Posted by css145hs

This is true but at typical boat C rates, it should be minimal.

Yeah, self-heating should be minimal. (Unless it isn't, at which point it will maximal. Sorry, couldn't resist! )

There is certainly more to plan for than non-NMC chemistries.

Agreed, and it sounds like you've thought that through.

[nebster]

Originally Posted by john61ct

I wonder if the ABYC working group spec'ing non-lead standards is including NMC batts like this.

The rate this tech changes, they'll need to issue frequent updates after they publish v1.0

8-)

Is it true that they still haven't released anything, after all these years? At the rate they're going, we might all be dead by the time they publish v1.0!

[css145hs]

Originally Posted by nebster

Yeah, self-heating should be minimal. (Unless it isn't, at which point it will maximal. Sorry, couldn't resist! )

Agreed, and it sounds like you've thought that through.

The other thing we've done is fully leverage the separate charging and discharge circuits, and the victron integration there. So if it is overheating somehow, it can instruct the chargers/genset to reduce load or turn off. It can cut off the solar circuit without damage. If it has to cutoff the engine alternators its likely to blow them, but better than blowing the boat up. The individual batteries can also take themselves out, and since there's 3 of them, can totally run on two.

[john61ct]

Originally Posted by css145hs

If it has to cutoff the engine alternators its likely to blow them

Kill switch for alt just works with Field input not power input