Lithium Batteries installation

[magnetman]

I have endless room in my head for batteries. 

 

I suppose people do have space constraints which is why the drop in thing exists in the first place. 

 

One day these problems will be solved when infernal combustion engines are banned. 

1 minute ago, Alan de Enfield said:

 

MPPT fed by 24v alternator ?

See above

33 minutes ago, magnetman said:

Maybe A 24v alternator, 2 LA batteries to make a 24v bank and the MPPT repurposed for winter to charge the LFP batteries via alternator-LA battery-MPPT.

In summer the solar does the job and the other alternator is still 12 volts. 

 

 

 

33 minutes ago, IanD said:

 

 

 

I'm not sure using an MPPT controller as a B2B is a good idea, the nature of the beast is different -- an MPPT controller is expecting to have solar panels on the input, and to adapt the input current and voltage to maximise output power from the panels. Not sure how this will react when the input side is an LA battery being charged by an alternator, maybe it'll just provide maximum output current all the time...

Its an interesting question. 

 

i wonder if it is good for the makers (lets say Victron) to keep these two products separate in order to shift more blue items in flash boxes. They might be electronically very similar. I don't know but have difficulty working out why the output from an alternator through a battery would be so different from a solar panel. 

 

[IanD]

15 minutes ago, magnetman said:

I have endless room in my head for batteries. 

 

I suppose people do have space constraints which is why the drop in thing exists in the first place. 

 

One day these problems will be solved when infernal combustion engines are banned. 

See above

 

Its an interesting question. 

 

i wonder if it is good for the makers (lets say Victron) to keep these two products separate in order to shift more blue items in flash boxes. They might be electronically very similar. I don't know but have difficulty working out why the output from an alternator through a battery would be so different from a solar panel. 

 

 

A solar panel acts as a light-dependent current source, but at high voltages the current drops off -- as some point the product of V*I is maximum power, and the MPPT controller tries to find this by changing its own input voltage up and down periodically to track this point, or to be more accurate by adjusting the gain between input and output. It can only step voltage down so Vin has to be bigger than Vout. It's not expecting the input to be a low impedance voltage source like a battery, it's expecting it to be a solar panel with very different characteristics.

 

A DC-DC B2B converter is designed to work with low impedances (voltage sources) on both sides and control the current between them, and works when the input is lower or higher than the output.

 

They're fundamentally different beasts that work in different ways, have different internal circuits (buck vs. buck-boost SMPS if you want to get technical), and different control and protection algorithms.

 

The only thing they have in common is they both come in blue boxes... 😉

Edited February 10 by IanD

[magnetman]

Is there any reason why the MPPT would object to a voltage which was not jumping around? Presumably with solar panels with no shade and a sunny day this would be quite a common outcome. 

[IanD]

4 minutes ago, magnetman said:

Is there any reason why the MPPT would object to a voltage which was not jumping around? Presumably with solar panels with no shade and a sunny day this would be quite a common outcome. 

 

It's not that it objects to it as such, it's what the MPPT controller tries to do with a solar panel on the input and a battery on the output. When trying to track maximum power, what it does is wind the input voltage up and down and measure the input current and power from the panel to find the maximum power point -- it can do this because a solar panel has a high output impedance so the current doesn't change much with voltage.

 

https://sinovoltaics.com/learning-center/components/mppt-charge-controllers-mppt-advantages/

 

If it tries doing this with a battery on the input which is pretty much a voltage source -- where small changes in voltage cause massive changes in current -- then the MPPT controller could well come a cropper. You'd hope it wouldn't destroy itself but for sure the control algorithms will toss a wobbler, it may end up with the current bouncing up and down, might shut off, or might end up just sourcing maximum current all the time -- or it might work as a B2B, who knows, it's just not being used in the way that is intended.

[Tony1]

2 hours ago, IanD said:

 

Are you planning to keep the LA domestic battery or just have LA starter + domestic LFP + B2B? In which case you'd connect both alternators in parallel to the LA (if this works...) and have a bigger B2B for charging the LFP

 

35A continuous out of your 100A domestic alternator sounds pretty poor, most people seem to be OK with maybe 50% current even when charging LFP banks. I'd have thought something like a Victron Orion XS 12V-12V 50A battery charger would do the job nicely -- programmable currents and voltages, maximum efficiency is 98.5%, I expect full-load is less than this but should still be well above 95%. About £300 though, and not out for a couple of months... 😉

 

https://www.victronenergy.com/dc-dc-converters/orion-xs-12-12-50a-dc-dc-battery-charger

https://offgridpower.solutions/shop/victron-energy-smart-buckboost-dc-dc-charger-non-isolated-50a

 

Looks like Victron have produced the ultimate easy-to-fit solution for lithium charging, for those who cant afford an alternator controller costing anything up to £900, plus a few hours labour for an electronics person to fit it- and its 98% efficient! 

 

AND it has the ability to change the charging current- that will allow people to buy it with confidence, knowing that it will work with any alternator from 25 amps up to 200 amps. 

For those folks with good powerful alternators, you could get a pair of them and have anything up to 100 amps charging current, which is pretty decent really. If its a super hot summer day and you're a bit concerned about the alternator heat, you could reduce the charging current a bit using the phone. 

Yes its maybe 100 quid more than the basic 30 amp B2B, but the ability to precisely set the charging current is a bit of a game changer for me. I dont think there's any other unit that will let you do that. 

All you have to do is get in the engien bay with a thermometer, and find out what your alternator can safely put out, and set the B2B current to exactly that value- you'll be able to get every possible amp from your alternators.

 

Erm, anyone interested in buying two used Victron 30 amp B2Bs? 

 

 

[nicknorman]

All this shenanigans with trying to interface a LA alternator to Li by means of heavy current electronics,  when the solution is to convert the alternator to Li-friendly charging. End of problem!

 

It is a shame that currently available alternator regulators are so expensive, but good to see the Zeus appearing. Maybe some other manufacturers will produce something, and with a bit more competition prices will drop. The crazy thing is that it can be done cheaply in terms of component cost, I think mine might be over £50 but certainly less than £100, and that is with CANBUS and a VE.Direct interface for the BMV712. All the extra cost is in the software development, testing etc. It would be good if there was another open source project such as the Wakespeed was originally, but with a better initial concept - the Wakespeed design was far too complicated in that it reinvented the wheel and all the regulation was done from scratch, whereas there are several digital automotive alternator control chips on the market for a few £ that mean 90% of the hard work is already done.

[magnetman]

I wonder how many others use lithium batteries for starting the engines. 

 

Alternators do need to adapt to lithium batteries at some stage and it should not mean a regulator costing hundreds of pounds. 

 

 

[dmr]

9 hours ago, nicknorman said:

All this shenanigans with trying to interface a LA alternator to Li by means of heavy current electronics,  when the solution is to convert the alternator to Li-friendly charging. End of problem!

 

It is a shame that currently available alternator regulators are so expensive, but good to see the Zeus appearing. Maybe some other manufacturers will produce something, and with a bit more competition prices will drop. The crazy thing is that it can be done cheaply in terms of component cost, I think mine might be over £50 but certainly less than £100, and that is with CANBUS and a VE.Direct interface for the BMV712. All the extra cost is in the software development, testing etc. It would be good if there was another open source project such as the Wakespeed was originally, but with a better initial concept - the Wakespeed design was far too complicated in that it reinvented the wheel and all the regulation was done from scratch, whereas there are several digital automotive alternator control chips on the market for a few £ that mean 90% of the hard work is already done.

 

There is a fair bit of design effort involved, including bluetooth and writing good apps for iPhone and Android, plus a good enclosure and robust manufacture for a harsh environment, then its a relatively small market so the develiopement cost is probably a significant part of the selling price.

An open source approach would be good, Wakespeed started off as an Arduino but had to move to a better controller which makes open source a bit less attractive.

I disagree with you (dangerous 😀) on the current control loop, I think its better to do all the critical stuff yourself rather than using "black box" software in a controller chip which is really designed for automotive. However you have to know what you are doing here, its specialist stuff and there are various occassional edge cases that need to be bandled (like the lithium disconnect) and experimentation during development can have expensive mistakes. (I have designed current control loops for electric motors in the distance past).

It would be lovely if Victron got involved but for now it looks like the Zeus is the way to go.

[cheesegas]

Just now, magnetman said:

I wonder how many others use lithium batteries for starting the engines. 

 

Alternators do need to adapt to lithium batteries at some stage and it should not mean a regulator costing hundreds of pounds. 

The issue with a lithium starter battery is that most BMSs are rated for 100-200 amps, whereas the starter motor can draw 300+ amps. You'd need either a contactor based BMS, or massive FETs to pass the current. It's fairly common in the cafe racer motorcycle scene as their starter motors draw a lot less current.

 

Most alternators we see on boats at the moment are based on car designs, which have no need to adapt to lithium batteries - it's only marine-specific models like Balmars etc which have lithium compatible regulators, and those are expensive external ones. You'd usually see Balmars on dual engined cruisers with a big house battery bank. I think the market is very small for an alternator with an internal lithium regulator so it'll be many years before one comes out.

 

There are however some large non-automotive alternators aimed at the modified car market with massive audio systems - they're built very well to withstand heavy loads at a low engine speed so it would be good to see them adapted for marine use. Some even have remote rectifiers and regulators to keep the heat in the case down. Again though, not cheap as the market is pretty small! 

[Alan de Enfield]

34 minutes ago, magnetman said:

Alternators do need to adapt to lithium batteries at some stage and it should not mean a regulator costing hundreds of pounds. 

 

Why not ?

 

If a small market wants (needs) something they have to pay the design, set up and low volume production costs amortised over a small number of customers.

[magnetman]

24 minutes ago, cheesegas said:

The issue with a lithium starter battery is that most BMSs are rated for 100-200 amps, whereas the starter motor can draw 300+ amps. You'd need either a contactor based BMS, or massive FETs to pass the current. It's fairly common in the cafe racer motorcycle scene as their starter motors draw a lot less current.

 

Most alternators we see on boats at the moment are based on car designs, which have no need to adapt to lithium batteries - it's only marine-specific models like Balmars etc which have lithium compatible regulators, and those are expensive external ones. You'd usually see Balmars on dual engined cruisers with a big house battery bank. I think the market is very small for an alternator with an internal lithium regulator so it'll be many years before one comes out.

 

There are however some large non-automotive alternators aimed at the modified car market with massive audio systems - they're built very well to withstand heavy loads at a low engine speed so it would be good to see them adapted for marine use. Some even have remote rectifiers and regulators to keep the heat in the case down. Again though, not cheap as the market is pretty small! 

 

It would be very easy to have two negative terminals. One attached to the BMS for loads which may discharge the battery while unattended and the other one for heavy discharge such as starter motor and connected directly to the terminals. 

 

I have a battery like this.40A BMS or connect direct. It was home made by someone who is an EV industry professional. Seems an interesting approach I think. 

 

 

 

 

5 minutes ago, Alan de Enfield said:

 

Why not ?

 

If a small market wants (needs) something they have to pay the design, set up and low volume production costs amortised over a small number of customers.

Yes that is basic economics but if lithium batteries become universal then it seems demand would go up. 

 

Is it not plausible just to have say a 65A alternator which will charge a LFP battery and switch itself off after a certain amount of time? If one knew how many Ah one wanted would it not make sense to put a timer onto the regulator. 

Edited February 11 by magnetman

[nicknorman]

18 minutes ago, dmr said:

 

There is a fair bit of design effort involved, including bluetooth and writing good apps for iPhone and Android

 

Bluetooth is easy to implement at the regulator end. It’s is just a serial interface and a £3 BLE module from China, plus the software to compose the appropriate sentences. iPhone and Android App development is beyond my pay grade and probably harder work, but I question why a phone app is really necessary for a well designed alternator controller that “just works”. Yes very interesting to see the field current etc for the first 5 minutes, but after that what are you going to do with it? Ok maybe configuration too, but again that is a one-shot usage.

 

18 minutes ago, dmr said:

I disagree with you (dangerous 😀) on the current control loop, I think its better to do all the critical stuff yourself rather than using "black box" software in a controller chip which is really designed for automotive.

I disagree, obviously! An automotive chip is designed to quite a high standard, much higher than stuff for domestic use etc, and of course it will have been “field-tested” in millions of vehicles world wide. It has to be 100% robust and with several layers of fail-safe.

 

In using one, all one is doing is specifying the voltage to regulate to and the maximum field current, data which would normally have come from the vehicle’s ECU. In the event of the vehicle ECU ceasing communication with the chip (wire come off etc) the devices go into a failsafe mode, in the case of the one I use, a default voltage of 13.5v which is vehicle-safe but also of course is lithium-safe. There are separate hardware (ie not software dependant) protections within the chip eg a hardware over-voltage shutdown. 

 

It would be a massive and pointless task to repeat all that design and testing work and I can see no reason other than masochism to follow that route!

[Alan de Enfield]

6 minutes ago, magnetman said:

Yes that is basic economics but if lithium batteries become universal then it seems demand would go up. 

 

But in a very large number of applications (most .?) the lithium batteries will not be charged by an automorive type alternator, they will be charged by solar or some 'mains' powered battery charger.

 

In the forseable future I really do not see that car starter batteries will change to lithium batteries due to the cost add-on, which leaves only niche markets, using automotive alternators, ie 'small' boats, and off grid camper vans.

[cheesegas]

11 minutes ago, magnetman said:

 

It would be very easy to have two negative terminals. One attached to the BMS for loads which may discharge the battery while unattended and the other one for heavy discharge such as starter motor and connected directly to the terminals. 

 

Is it not plausible just to have say a 65A alternator which will charge a LFP battery and switch itself off after a certain amount of time? If one knew how many Ah one wanted would it not make sense to put a timer onto the regulator. 

This would make it possible to severely damage the battery by trying to start the engine when the battery is discharged and the BMS is about to do a discharge disconnect. It’s also the most likely time that someone’s going to start the engine - leisure battery low, need to run engine to charge batteries. 
 

The timer would work for a lead battery but not lithium, and it would still be a bad solution. Say you’ve worked out that you need to put in 100ah with your 100a alternator, which works out to 1hr. Then, you go inside, switch the inverter on, charge your laptop and put on the washing machine…this all takes say 20ah so you end up with only 80ah in the battery. And if you were to start the engine in summer at 3pm to cruise, you’d need to check that the alternator is off or else you’d be holding the lithium battery at full charge which isn’t good for it. 
 

A better solution which is what myself and a couple others do on here is to use the relay output on a Victron BMV to disable the alternator at 100%. This is better but not great - without a float voltage, if you were to draw high power with the batteries full and alt off, it goes into a loop of short charge, short discharge and repeat. 

[magnetman]

17 minutes ago, cheesegas said:

This would make it possible to severely damage the battery by trying to start the engine when the battery is discharged and the BMS is about to do a discharge disconnect. It’s also the most likely time that someone’s going to start the engine - leisure battery low, need to run engine to charge batteries. 
 

The timer would work for a lead battery but not lithium, and it would still be a bad solution. Say you’ve worked out that you need to put in 100ah with your 100a alternator, which works out to 1hr. Then, you go inside, switch the inverter on, charge your laptop and put on the washing machine…this all takes say 20ah so you end up with only 80ah in the battery. And if you were to start the engine in summer at 3pm to cruise, you’d need to check that the alternator is off or else you’d be holding the lithium battery at full charge which isn’t good for it. 
 

A better solution which is what myself and a couple others do on here is to use the relay output on a Victron BMV to disable the alternator at 100%. This is better but not great - without a float voltage, if you were to draw high power with the batteries full and alt off, it goes into a loop of short charge, short discharge and repeat. 

 

LTO is an interesting option. I have been experimenting with 6S LTO modules and they don't seem to be bothered about much. I've had one of them down to about 4v (no BMS) and it charged up normally cells are still within 3mV of each other. 

 

Also the top voltage is 16.8v. Mid range maintenance voltage between 14 and 15v. 

 

Not done any alternator testing its just been solar so far but I will be doing some experiments. 

 

 

Also excellent for engine starting I use a 40Ah module charged to 15.2v to start the Perkins P4 its does it happily. 

 

They are generally rather expensive batteries but if one were to have a multi cycle charge discharge regime the battery banks can be smaller. 

 

I have marked the module which went right down (lights left on) and interesting to see if it suffered at all. It appears not. 

 

 

 

LTO have very high cycle life and also huge C rating. 10C is easy. 

 

i took apart a Suzuki Swift mild hybrid battery. It has a 3Ah 5S LTO battery which is nominal 12v 3Ah. Tiny.  The battery has a 150A fuse despite being so small. This is used in the car to add power while moving from a standstill then recharged. I believe the in/out rate is probably something like 20C in use which would mean it was supplying nearly 1HP for a short time. 

 

It might be more. The 150A fuse was interesting. 

 

 

 

 

Yes 6S is a bit high for normal 12v loads but easy to put a buck converter on lights and pumps circuit. Inverters cut out at over 15v which is about 85% charged so a happy battery. 

 

 

LTO ocv soc graph .  So roughly 40-90% is 2.2-2.5v which for a 6S would be 13.2-15v. Not too terrible. Space above for safety. 

 

Edited February 11 by magnetman

[Tony1]

10 hours ago, nicknorman said:

All this shenanigans with trying to interface a LA alternator to Li by means of heavy current electronics,  when the solution is to convert the alternator to Li-friendly charging. End of problem!

 

 

Even with my limited understanding it seems clear that an alternator controller is the most efficient and elegant solution, but I don't know if its going to be cost effective for everyone. 

Consider those unfortunates among us who have sub-40hp engines (such as my canaline 38), many with narrow V belt alternators that have quite modest current output (and I mean continuous output).

On top of the cost of the lithium batteries themselves, you're looking at perhaps £1200 to install one of these controllers on a domestic alternator, that can only put out say 45 amps safely on a continuous basis.

Its a lot of money to spend for such a modest output, And upgrading to a powerful alternator in some cases requires a crank upgrade to fit a poly V belt, so that adds maybe £250-300 for the alternator and maybe £500 for supply and fit of the crank upgrade. So perhaps another £800 on top of the £1200 for the Wakespeed controller? 

And that's on top of the cost of the batteries themselves.

I suspect that to many folks with older or less powerful engines, it will appear just too high a cost- so its easy to see the appeal of this new victron B2B unit for that group of people. 

They can still achieve their max 45-50 amp charging rate, but for only £300 - and they can probably install it themselves and save money there too. 

And if they have two alternators they can get a second B2B unit and get another 40 amps charge from their canaline 38 (if they're feeling a bit  brave).

I totally agree its not the best way to solve the problem, but I reckon for for a certain segment of the boating population it will look like a decent cost effective option. 

The biggest question in my mind for people pondering spending a lot of cash on lithiums is- which of these charging setups will meet any potential future regulations...

 

 

Edited February 11 by Tony1

[nicknorman]

3 minutes ago, Tony1 said:

 

Even with my limited understanding it seems clear that an alternator controller is the most efficient and elegant solution, but I don't know if its going to be cost effective for everyone. 

Consider those unfortunates among us who have sub-40hp engines (such as my canaline 38), many with narrow V belt alternators that have quite modest current output (and I mean continuous output).

On top of the cost of the lithium batteries themselves, you're looking at perhaps £1200 to install one of these controllers on a domestic alternator, that can only put out say 45 amps safely on a continuous basis.

Its a lot of money to spend for such a modest output, And upgrading to a powerful alternator in some cases requires a crank upgrade to fit a poly V belt, so that adds maybe £250-300 for the alternator and maybe £500 for supply and fit of the crank upgrade. So perhaps another £800 on top of the £1200 for the Wakespeed controller? 

And that's on top of the cost of the batteries themselves.

I suspect that to many folks with older or less powerful engines, it will appear just too high a cost- so its easy to see the appeal of this new victron unit for that group of people. 

They can still achieve their max 45 amp charging rate, but for only £300 - and they can probably install it themselves and save money there too. 

And if they have two alternators they can get a second B2B unit and get another 30-40 amps charge from their canaline 38 (if they're feeling a bit  brave).

I totally agree its not the best way to solve the problem, but I reckon for for a certain segment of the boating population it will look like a decent cost effective option. 

The biggest question in my mind for people pondering lithiums is- which of these charging setups will meet any potential future regulations...

 

Yes I get the  cost thing and that was my point really - the intrinsic cost (component cost) of a fancy smart alternator regulator is less than £100 so it is a shame that they cost nearly 10 times that. The only reason why they do cost that much is because the manufacturer and the retailer both need to make a profit, and a lot of up-front man-hours go into software development, testing etc.

 

But there is a lot of “open source” stuff out there, people give freely of their time to create stuff and publish it on the internet, which eliminates most of the one-off costs above. So it’s a shame that this doesn’t happen for an alternator controller project. Well it did, as we know, but it then went commercial.

[David Mack]

19 minutes ago, Tony1 said:

 

I suspect that to many folks with older or less powerful engines, it will appear just too high a cost- so its easy to see the appeal of this new victron B2B unit for that group of people.

Or even easier just to stick with lead acid batteries and established charging technology.

[Tony1]

9 minutes ago, David Mack said:

Or even easier just to stick with lead acid batteries and established charging technology.

 

I think it will depend on budget to a large extent.

Fortunately there are middle-ground options that sit between a simple lead acid setup and a full-on £3.5k lithium installation. 

Lithiums bring a lot of benefits, especially for liveaboard boaters, and people will always look for an affordable way of obtaining those benefits. 

So its easy to imagine that £1300 for a 460Ah Fogstar battery plus £300 for a victron B2B might be within the range of some people.

But yes- others will not fancy that, and will prefer to stick to lead acids.

 

Edited February 11 by Tony1

[nicknorman]

1 hour ago, David Mack said:

Or even easier just to stick with lead acid batteries and established charging technology.

Only someone who has not had Li on their boat would say that.

[Naughty Cal]

6 minutes ago, nicknorman said:

Only someone who has not had Li on their boat would say that.

Agreed.

[MtB]

2 hours ago, nicknorman said:

Yes I get the  cost thing and that was my point really - the intrinsic cost (component cost) of a fancy smart alternator regulator is less than £100 so it is a shame that they cost nearly 10 times that. The only reason why they do cost that much is because the manufacturer and the retailer both need to make a profit, and a lot of up-front man-hours go into software development, testing etc.

 

But there is a lot of “open source” stuff out there, people give freely of their time to create stuff and publish it on the internet, which eliminates most of the one-off costs above. So it’s a shame that this doesn’t happen for an alternator controller project. Well it did, as we know, but it then went commercial.

 

 

And back here in the real world, the above is not available to the average muppet boater like me with other stuff to do with their lives. The options for us are:

 

1) A cheap-as-chips hybrid drop-in regulated by a long bit of wire (tolerating the tiny risk of the internal BMS failing)

2) £6k+ dropped into the coffers of Ed Shiers or similar

3) Stick with the LA bank and keep on wearing down the patience of the neighbours with excessive engine running in winter

 

 

 

 

[magnetman]

The Victron 50A B2B does look good to be fair as it seems to allow user settings for charge profile and current. 

 

Quite attractive. Easy enough to have an alarm for alternator case temperature. 

[kris88]

What’s this Zeus that people are going on about? Anybody got a link? 
 

is this it https://arcomarine.com/products/arco-zeus-high-energy-alternator-regulator

Edited February 11 by kris88

[David Mack]

38 minutes ago, nicknorman said:

Only someone who has not had Li on their boat would say that.

Well currently I have nothing bar an unconnected alternator on the boat. So it's a reasonable question what to install.

But the choice seems to be between a costly professional install, a simple DIY hybrid system with a long bit of wire, getting up to speed with some sort of self-developed electronic control system or sticking with lead acid, the 2nd and 3rd options apparently falling foul of battery manufacturers' instructions, relevant standards and insurance requirements. Hence, slightly reluctantly thinking of heading down the LA route.

 

Incidentally, have any of the hire boat firms gone down the lithium route, or are they all sticking with tried and tested technology, widely understood and easily fixed if something goes wrong?