Jump to content

Drop In Lithiums


Recommended Posts

Surely the BMS can be engineered to manage the profile better, given that most of the drop ins will be into a 12v system charging at 14.6. I realise that temp sensing and regulators are important especially in larger systems.

Link to post
Share on other sites

When you consider all the variables, 'not really' is the answer although it wouldn't be impossible. 

Solar/wind need to be managed differently. Alternators differently again. And Battery to Battery systems different again.  Mains chargers again differently. 

A2B and B2B chargers are very expensive ($1500 for 100a) and they generate a lot of heat that you really don't want in your battery pack. 

You'd either not have enough charge rate or you'd be paying for too much that you couldn't use. You'd either have charge sources you couldn't use/didn't need, or you'd come to add a charge source in the future and find out you couldn't. 

Anything builtin would be a massive compromise. 

Link to post
Share on other sites
20 hours ago, Psyche said:

Surely the BMS can be engineered to manage the profile better, given that most of the drop ins will be into a 12v system charging at 14.6. I realise that temp sensing and regulators are important especially in larger systems.

All the BMS does is monitor each cell and balances the charge to each individual cell. Apart from that, it is pretty dumb. This happens in all Lithiums, even EV ones. When it comes to the likes of the EV batteries, there is additional software on the vehicle than controls the more important charging/discharging parameters. This software is different between manufactures of course. But even then, I know of many vehicles that allow the bank to be charged to 100%, when it is known that this limits the overall life of the Bank. I guess it is more about marketing the range rather than life of the battery.

  • Upvote 1
Link to post
Share on other sites

It's worth pointing out, that if you get a lfp battery and an appropriate charge controller, you don't need to worry about not charging to "100%". 

Just let the charge controller do its thing.  If your charger goes faulty then the BMS will turn off the battery to stop any damage occurring. 

Modern BMSs are always set well below the true 100% charge capacity which is usually about 110% of the rated capacity. 

 

Link to post
Share on other sites

Are you referring to a single Alternator? If single Alternator then this is a safe and simple setup. 

Alt > Start Battery > B2B > Drop-In

Size the b2b at the lowest of

  • ~50% the capacity of your Alternator; or
  • ~70% the capacity of your BMS/Li charge rate

If it is a modern alternator, you trust the built-in regulator to do its thing when the temperature increases etc, or it is continuously rated (eg a Balmar) then you could ignore the 50% alternator warning. 

Ideally choose a b2b that has a on/off switch to give you more control over the remaining charge in the battery. Eg, not charging at the end of a weekend away, not charging if your starter was flattened, not charging when using shore power etc

This simple/safe solution forgoes one of the key benefits of Li, and that's fast charging because you're limited to the rate of the B2B or the drop-in BMS. 

2 hours ago, Guest said:

Wth the B2B as alternative load  I can see the start battery getting overcharged? Specially if gel or AGM.

When the start battery is fully charged the voltage at the alternator will drop away.  Mine drops to 13.7v. The B2B is a buck-boost converter, so it's taking the 13.7v at the Alternator and delivering ~14.6v (3.65v/cell) to your Li.   So your start battery will not be overcharged. 

If you're referring to a dedicated alternator then there isn't really a simple/safe and fullproof way.  There are simple ways, but they are not foolproof and they could become quite expensive.

I know a few people who do not run a BMS at all, 'they' are the BMS and they explicitly trust/monitor the charging system. After all, the BMS is the last line of defence, if it is ever activating something else has gone horribly wrong.

  • Upvote 1
Link to post
Share on other sites
1 hour ago, Guest said:

For those about to pull trigger, Marine deals has a Sterling 120A B2B at $900. One only. That is a good price.

That is a LA charger.  It supports neither custom or Li charge profiles.  For charging LiFePO4 you'd have to choose a compatible LA charge profile without an equalization charge - any of which will be a compromise.

1 hour ago, Guest said:

I don't see point in going to B2B unless you're willing to shell out for decent output.

I agree completely - however, the Alt > Start Battery > B2B > Drop-In pattern remains the simplest and safest for the typical NZ small yacht setup that has a stock alternator, two batteries and a VSR.

1 hour ago, Guest said:

The elegance of not using the lithium settings on the dumb but configurable ext regulator and a B2B  becomes apparent.

If your external regulator is dumb enough to limit current (albeit without reading a shunt eg, "belt-control") and it can obviously limit voltage then a better solution would be:

Alternator  >> Splitter >> Lithium >> B2B >> Start Battery

Connect your start battery to the splitter and get a small low end B2B to keep your start battery topped up.

Set the voltage on your Alternator to 3.55v/cell - that will still charge your Li sufficiently while not over charging the start battery.  If you have an AGM start battery then you could up this as AGMs generally need 14.7v to fully charge - ymmv - so check your battery specs.

Make sure the splitter is an ultra low voltage loss - both Victron and Sterling do a FET based splitters rated at up to 200A.

1 hour ago, Guest said:

For me, (scrooge mcgrinch), an alternator saver device at  10% of a B2B and multiple v meters (already there with alarms) and watch for the  "knees"

Personally, I would never take my alternator away from my start battery with a drop-in BMS - that's a catastrophe waiting to happen.

Those Alt Savers seem like a solution looking for a problem.  They also seem a bit sacrificial - I am sure they work as advertised the first time - it's the second, third and fourth times that concern me... definitely good insurance, however it's unclear to me how many times you have to pay the premium.

  • Like 1
  • Upvote 1
Link to post
Share on other sites

OK, here is my current level of knowledge. I'm not a LifePo4 specialist, and currently experimenting!

Why would you want LiFePo4 batteries. 

  1. Light weight - for my tests, I've put in 2 120 a/h batteries giving  192a/h useable at 80% depth of discharge. They can each be carried easily in one hand. They replaced 2 200 a/h VRSLA batteries weighing 62KG each, and gave 200 A/h useable capacity at 50% dod.
  2. Life span - the VRSLA's give 400-1000 cycles - the LiFePo4 about 3000 cycles, both at the manufacturers recommended dod.
  3. Charging abilities. A good LiFePo4 battery can accept huge charge rates right until full - no bulk/absorbtion/float required. It is amazingly quick to charge (provided your charging system is up to it - an alternator designed for lead acid may quickly fail. 
  4. Voltage stability - LiFePo4 batteries keep very stable voltages until almost empty, when it drops VERY fast... 

Why not?

  1. Your charging system will need careful consideration and planning - for all charge sources, Aternator, Solar, Wind, Shore Power, Genset - whatever you have will need to be set up for the right voltages - EVEN WITH DROP IN - because they are not! LiFePo4 is different. All these things either have to be set up correctly, or, if they cannot be, then replaced.
  2. LiFePo4 batteries, especially drop ins, are not starting or high current batteries. Check the specs for max discharge rates for high loads. Sometimes this can be overcome by several smaller capacities in a parallel bank rather than one larger one.

LiFePo4 batteries don't need - or want - to be kept at full charge all the time, so you have to change your mindset about this. Also, other stuff may change - like on Island Time for example, the watermaker draws 24a at 12v when in production, so we used to only use it when the engine was running. However, now, as the batteries can take the full alternator output (around 90-100a continuous in our case) it is now better to run  it off the batteries, and top them up if needed. 

The NZ regs state among other things, that lithium batts must have audio and visual warnings, at the normal operational station of the vessel,  before battery shutoff. This is a problem for virtually all drop in batteries.  So, at this point, there is no way to use a drop in battery and comply with the law. There is no device I'm aware of that will talk to a drop in system's internal  bluetooth BMS (Battery Management System) and facilitate this.

Next, the drop ins are normally made from cheaper grade cells, that may not be matched or balanced. Balance is the voltage control of the internal cells in the battery, normally 4x 3.2v cells for a 12.8v battery. Proper cell balance allows equal charging and discharging of the battery, as well as actual capacity. More on that below. 

What controls the balance, and can switch the battery on or off is the internal BMS (There are some drop ins without this - don't go there!). Most of the low(er) cost drop ins use FETs for control. A FET is like a solid state relay, it's not bright, it's just a switch. The cheaper FETs will not carry much current, so the ability of the battery to charge, discharge, and balance is or can be restricted. Look at the spec sheet on any battery you are considering, and ensure that the charge rate is good - at least 1c (1x the capacity of the battery in amp hours, so 100a for a 100a/h battery), the discharge rate is workable (what's your highest load the battery could ever see? - electric winches, inverters, anchor winches etc for your boat, and if you can, the balance current for the BMS. That last one can be hard to find! Also, a good manufacturer will be able to provide vibration test data for a marine battery. The most common drop in (therefore the cheapest) around are actually made for LED street lights - so the output can be low amps, the charge by solar low amps as well, and the balance currents can be small. 

So, the BMS. The BMS will switch off the battery during use if

  1. A single cell voltage gets to low (under about 3 volts)
  2. A single cell voltage gets too high (over about 3.65 volts)
  3. The battery total voltage gets too low (some under 12v, some under 10v)
  4. The battery total voltage gets to high (around 14.6v)

Some further data first. V=IR so the cells voltage will rise = to the current x resistance of that cell - the other cells may be different.

A battery can switch off well before full charge, and in fact never get charged if one cell is out of balance, and that cannot be remedied.

The BMS is supposed to balance the cells, usually either during charging or when the battery is resting. Good BMS units can control how much current goes to each cell, and even bleed from one (higher) cell to top up another. Cheap ones cant do that, they can simply add a little resistance to one cell to reduce the current to that cell, and bleed off the excess energy into heat. If the BMS cannot control sufficiently the voltage of a higher voltage cell during charging, then that cell voltage will rise to the cut off voltage, and the WHOLE BATTERY will shut down at least it's charge side (it likely, but not always, will keep the discharge side turned on). That means that the battery may stop charging when nowhere near its total capacity. 

This issue is why quality Lifepo4 battery manufactures select matched cells (virtually identical resistance and voltage at the fully charged state). The process of fully charging and measuring the cells is known as top balancing, and is done by individually charging the cells.

Unfortunately in a drop in, you have no access to the individual cell terminals to do this, and the cells are not usually matched at production either. So you have to rely on the internal BMS to do it for you. This means some will NEVER manage to balance the cells at full charge, and they wont produce the capacity that they should, as the BMS is not capable of correcting the imbalance, depending on time and their balance current. Some of the cheap BMS systems are only capable of a 50ma or less balance current. More info here http://www.liionbms.com/php/wp_balance_current.php for any geeks. Please note that a 1000a/h battery with a 10ma balance current wont balance in YOUR lifetime! This is why balance current of the BMS matters. I learned this with my experimental system - the internal BMS balance current is totally inadequate :-( . 10amp balance current in a 100 a/h batt would be good...and balance in 10 hours from one flat cell.

Remember from above - one cell can turn off the rest, so one cell at 3.65v then switch off when the others are at 3.4v means  the battery was only around 85% charged when the charging shut down due to poor balance.

When a LiFePo4 battery switches off without informing the alternator first, the alternator can fail immediately, blowing the diode pack. As the drop in units dont normally have the ability to do that, another method must be employed. I use an Argo Fet based battery splitter, and a SLA start battery, so that the circuit is never simply cut off. Works ok so far....

There's probably more to say, but this post is long enough, and tests are continuing...

 

  • Upvote 3
Link to post
Share on other sites
19 hours ago, Island Time said:

There's probably more to say, but this post is long enough, and tests are continuing...

How many amps are you getting out of your stock D2-40 alternator?

Are you limiting it with the VRC-200?

I assume that you have these two batteries in parallel and each BMS is rated at 60a max - so should in theory be able to max out the alternator and let the internal temperature regulator do what it does best.

Link to post
Share on other sites
2 hours ago, CarpeDiem said:

How many amps are you getting out of your stock D2-40 alternator?

Are you limiting it with the VRC-200?

I assume that you have these two batteries in parallel and each BMS is rated at 60a max - so should in theory be able to max out the alternator and let the internal temperature regulator do what it does best.

I have the  standard 115a Mitsubishi Alt on my D2-40F, other variants have other alts. Eric (VRC-200 designer) says this is an excellent alt, and so far I have no reason to disagree.  It has built in temp compensation, and seems to stabilize output at around 90-100 amps continuous (60 mins at that so far longest charge I've needed), which is surprisingly good. The batteries I have are indeed in parallels, but they are rated for 120amps EACH max charge and max discharge. No restriction in current programmed into the VRC-200, just letting the alt temp compensation take care of that, so the alt can go as high as it can without overheating...

Link to post
Share on other sites
2 hours ago, Guest said:

I use an Argo Fet based battery splitter, and a SLA start battery, so that the circuit is never simply cut off. Works ok so far...."

 

1.So if motoring for longer than 1hr do you just switch the VRC-200 off? (and thus the fields)

No, but when the VRC-200 detects <3a charge at 14.6v - (VRC-200 bulk setting), it switches to float (13.6v) which is the storage recommendation for this battery.

2.And if you need to charge longer than 1hr what stops the start battery overcharging? Small buck converter? Or 2hr would be a complete recharge on Li and not enough to bother the SLA? Same as above, float voltage is 13.6, no issue for the SLA start batt.

3.  Max discharge is 1C ? (Can't see in your post) Then in parallel you have the windlass hooked to the house? Or if  0.6C ( same as charge) and to start battery and always run the motor seeing the Argofet is capable of 200A.  Depending the windlass that you have spose. These batts are 1C in and 1c out max. But my windlass is still on the start batt. They could do the capstan though, its 1200w...

Thanks.

edit. 

4. Are your cells links bolted or laser welded? Supposedly laser welded, but can't verify, I'm not opening the case.

5. Do you have access to any more? Not sure I'd recommend these, the BMS is too small, and balance is pretty poor at high levels of charge. So Far.

 

 

Link to post
Share on other sites
On 15/12/2021 at 3:24 PM, Island Time said:

OK, here is my current level of knowledge. I'm not a LifePo4 specialist, and currently experimenting!

Why would you want LiFePo4 batteries. 

  1. Light weight - for my tests, I've put in 2 120 a/h batteries giving  192a/h useable at 80% depth of discharge. They can each be carried easily in one hand. They replaced 2 200 a/h VRSLA batteries weighing 62KG each, and gave 200 A/h useable capacity at 50% dod.
  2. Life span - the VRSLA's give 400-1000 cycles - the LiFePo4 about 3000 cycles, both at the manufacturers recommended dod.
  3. Charging abilities. A good LiFePo4 battery can accept huge charge rates right until full - no bulk/absorbtion/float required. It is amazingly quick to charge (provided your charging system is up to it - an alternator designed for lead acid may quickly fail. 
  4. Voltage stability - LiFePo4 batteries keep very stable voltages until almost empty, when it drops VERY fast... 

Why not?

  1. Your charging system will need careful consideration and planning - for all charge sources, Aternator, Solar, Wind, Shore Power, Genset - whatever you have will need to be set up for the right voltages - EVEN WITH DROP IN - because they are not! LiFePo4 is different. All these things either have to be set up correctly, or, if they cannot be, then replaced.
  2. LiFePo4 batteries, especially drop ins, are not starting or high current batteries. Check the specs for max discharge rates for high loads. Sometimes this can be overcome by several smaller capacities in a parallel bank rather than one larger one.

LiFePo4 batteries don't need - or want - to be kept at full charge all the time, so you have to change your mindset about this. Also, other stuff may change - like on Island Time for example, the watermaker draws 24a at 12v when in production, so we used to only use it when the engine was running. However, now, as the batteries can take the full alternator output (around 90-100a continuous in our case) it is now better to run  it off the batteries, and top them up if needed. 

The NZ regs state among other things, that lithium batts must have audio and visual warnings, at the normal operational station of the vessel,  before battery shutoff. This is a problem for virtually all drop in batteries.  So, at this point, there is no way to use a drop in battery and comply with the law. There is no device I'm aware of that will talk to a drop in system's internal  bluetooth BMS (Battery Management System) and facilitate this.

Next, the drop ins are normally made from cheaper grade cells, that may not be matched or balanced. Balance is the voltage control of the internal cells in the battery, normally 4x 3.2v cells for a 12.8v battery. Proper cell balance allows equal charging and discharging of the battery, as well as actual capacity. More on that below. 

What controls the balance, and can switch the battery on or off is the internal BMS (There are some drop ins without this - don't go there!). Most of the low(er) cost drop ins use FETs for control. A FET is like a solid state relay, it's not bright, it's just a switch. The cheaper FETs will not carry much current, so the ability of the battery to charge, discharge, and balance is or can be restricted. Look at the spec sheet on any battery you are considering, and ensure that the charge rate is good - at least 1c (1x the capacity of the battery in amp hours, so 100a for a 100a/h battery), the discharge rate is workable (what's your highest load the battery could ever see? - electric winches, inverters, anchor winches etc for your boat, and if you can, the balance current for the BMS. That last one can be hard to find! Also, a good manufacturer will be able to provide vibration test data for a marine battery. The most common drop in (therefore the cheapest) around are actually made for LED street lights - so the output can be low amps, the charge by solar low amps as well, and the balance currents can be small. 

So, the BMS. The BMS will switch off the battery during use if

  1. A single cell voltage gets to low (under about 3 volts)
  2. A single cell voltage gets too high (over about 3.65 volts)
  3. The battery total voltage gets too low (some under 12v, some under 10v)
  4. The battery total voltage gets to high (around 14.6v)

Some further data first. V=IR so the cells voltage will rise = to the current x resistance of that cell - the other cells may be different.

A battery can switch off well before full charge, and in fact never get charged if one cell is out of balance, and that cannot be remedied.

The BMS is supposed to balance the cells, usually either during charging or when the battery is resting. Good BMS units can control how much current goes to each cell, and even bleed from one (higher) cell to top up another. Cheap ones cant do that, they can simply add a little resistance to one cell to reduce the current to that cell, and bleed off the excess energy into heat. If the BMS cannot control sufficiently the voltage of a higher voltage cell during charging, then that cell voltage will rise to the cut off voltage, and the WHOLE BATTERY will shut down at least it's charge side (it likely, but not always, will keep the discharge side turned on). That means that the battery may stop charging when nowhere near its total capacity. 

This issue is why quality Lifepo4 battery manufactures select matched cells (virtually identical resistance and voltage at the fully charged state). The process of fully charging and measuring the cells is known as top balancing, and is done by individually charging the cells.

Unfortunately in a drop in, you have no access to the individual cell terminals to do this, and the cells are not usually matched at production either. So you have to rely on the internal BMS to do it for you. This means some will NEVER manage to balance the cells at full charge, and they wont produce the capacity that they should, as the BMS is not capable of correcting the imbalance, depending on time and their balance current. Some of the cheap BMS systems are only capable of a 50ma or less balance current. More info here http://www.liionbms.com/php/wp_balance_current.php for any geeks. Please note that a 1000a/h battery with a 10ma balance current wont balance in YOUR lifetime! This is why balance current of the BMS matters. I learned this with my experimental system - the internal BMS balance current is totally inadequate :-( . 10amp balance current in a 100 a/h batt would be good...and balance in 10 hours from one flat cell.

Remember from above - one cell can turn off the rest, so one cell at 3.65v then switch off when the others are at 3.4v means  the battery was only around 85% charged when the charging shut down due to poor balance.

When a LiFePo4 battery switches off without informing the alternator first, the alternator can fail immediately, blowing the diode pack. As the drop in units dont normally have the ability to do that, another method must be employed. I use an Argo Fet based battery splitter, and a SLA start battery, so that the circuit is never simply cut off. Works ok so far....

There's probably more to say, but this post is long enough, and tests are continuing...

 

Just a couple of points,  without picking at your post. 

Low balance current on BMS's in the 10 - 40ma range is perfectly sufficient for balancing small to medium lithium batteries. It's just not working how most people assume, as the majority of people are assuming you are removing current from each cell, while the other cells are receiving charge current. In fact what is happening, it the balancing is resistive [yes you do get a current measurement] and works from a voltage drop standpoint. The cells being balanced have a slightly lower voltage, therefore less charge acceptance, with the other cells are charging at a higher current, so they are catching up and not the out of balance cell having current removed. Balancing should only occur during charging and there is very little point is balancing happening until you are getting above 3.45v a cell, as below that the charge curve is to flat and you do need a difference in potential to be effective. 

FETs,  or MosFETs as in the case of what is used in most BMS's have very low resistance when full on,  they also are perfect for this role as because once the gate charge is applied there is no further current draw from the device,  also the resistance between on and off is such that they can be used as variable resistor, to resistively balance the cells in the pack. 


You can top balance your drop ins. Bring them slowly up to 14.6v, from 14.2 over a few days on a bench top power supply should do the job perfectly. There is no real capacity to be gained here, but is will equalise the cells if you are worried about voltage difference between them.


With regard to starting engines and running high loads, lithium batteries are vastly superior if the BMS can handle it. The fusing requirements for lithium based batteries is an indicator of this as you need an arc interrupt capacity of 20,000amps and currently, Class T fuses are the only accessible marine fuses I know with this rating. A lead battery of any type can not sustain the current output and voltage stability of most lithium battery types of equivalent capacity.  


Audio and visual warnings, a battery monitor like a Victron 712 would be sufficient and this is what I use. 


Perfectly matched cells is another internet myth that is persistent. It is true to a degree, but the internal resistance is such that you would need to be pushing some pretty massive current or likely have faulty cell in order to have a real world issue and that might be a couple of percent lower capacity for the total pack in most circumstances. If you have a faulty cell, you just have a faulty cell like any battery.  Failures can happen.

  • Like 4
Link to post
Share on other sites

A little update on this topic, I spoke to a marine sparky recently who said that he had fitted quite a few drop ins quite successfully provided a quality external regulator was used (around $600) and he mentioned about 1300 for a 100 a/h battery so sounds like good quality but obviously this is a very basic system. Still we are talking close to 2k and given that using my existing system with a carbon foam LA was only $450. Its confirms that lithium is not quite there yet for me anyway.

So I took a snap of a mates setup which is way, way above my pay grade!

Made me feel a bit better :)

 

IMG_0507.thumb.jpg.35f2be96a8d93dc4961ea30bd43d49dc.jpg

Link to post
Share on other sites
Just now, Psyche said:

A little update on this topic, I spoke to a marine sparky recently who said that he had fitted quite a few drop ins quite successfully provided a quality external regulator was used (around $600) and he mentioned about 1300 for a 100 a/h battery so sounds like good quality but obviously this is a very basic system. Still we are talking close to 2k and given that using my existing system with a carbon foam LA was only $450. Its confirms that lithium is not quite there yet for me anyway.

So a mate has this setup which is way, way above my pay grade!

 

IMG_0507.thumb.jpg.35f2be96a8d93dc4961ea30bd43d49dc.jpg

Looks a lot like a big 62’ cat 😏

Link to post
Share on other sites
On 15/01/2022 at 11:57 AM, Psyche said:

my existing system with a carbon foam LA

 

IMG_0507.thumb.jpg.35f2be96a8d93dc4961ea30bd43d49dc.jpg

I would be very surprised if your existing gear can be programmed to suit carbon foam batteries, and then not be programed or made to work with lithium.  Carbon Foam batteries do not share the same bulk, absorption, float voltages with any of the other LA batts.  What brand Carbon Foam are you looking at?

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...