Help with replacing lead acid house battery with LiFePo4

ruvane

Member
Joined
24 Nov 2005
Messages
188
Location
Ashkelon Marina
www.techno-laser.com
Hi everyone,

I have a SmartGauge and SmartBank Advanced installed in my boat to control the batteries and it's been working great for a while. Not a large or complex setup - two lead acid batteries, one for engine start and the other for auxiliary electronics.

My aux battery has failed and I'd like to replace it with a LiFePo4 100Ah. I'm aware of the two main problems with simply dropping a LiFePo4 in place of the lead acid:
  • Alternator burnout from overloading it with high charge current to the LiFePo4.
  • Alternator diode destruction when the LiFePo4 is suddenly disconnected for whatever reason.
I'd appreciate all feedback on my suggested system in the attached graphic. I think that I'm covered for both the above issues.

Obviously this diagram ignores fusing and switches, it's simply to see whether I'm missing something critical in the overall system.

Many thanks and have a wonderful new year.
 

Attachments

  • lifepo4 wiring.jpg
    lifepo4 wiring.jpg
    174 KB · Views: 68

Poey50

Well-known member
Joined
26 Apr 2016
Messages
2,309
Location
Chichester
Visit site
Hybrid systems in which a lead acid battery is kept in the system certainly avoid the two problems you mention. The great exponents of these systems are Zwerfcat (you can Google it) and, closer to home, Phil Brooke-Little of the '12 Volt Boating Group' on Facebook. In both cases a fully user-configurable BMS is needed. The system makes use of the differing charging voltage profiles of lead acid and LFP. The latter charges at a lower voltage so the high voltage disconnect is configured to cut charge to the LFP at around 90-95% state of charge leaving the lead acid to be charged alone to a higher voltage and taken to full charge.

The two things unclear to me is whether yours is a fully configurable BMS - smaller 'drop-ins' tend to have fixed settings. I'm also unsure how nicely the SmartBank Advanced will play with the above. You could take advice from Merlin.

Not everyone likes to use a BMS high voltage disconnect in this way, preferring to use it as the final backstop for catastrophic-level protection only rather than for routine control of the BMS charging. Nevertheless, I can see the appeal.
 
Last edited:

Tranona

Well-known member
Joined
10 Nov 2007
Messages
41,065
Visit site
I am wondering why you think you need to change from lead acid. You only have one house battery which suggests a pretty small boat, or at least low demands for 12v power. What are you hoping to achieve which might be different from replacing like for like? or upgrading your current capacity?
 

migs

Active member
Joined
5 Jul 2011
Messages
130
Visit site
I would also question the cost/benefit of swapping to Lithium in this scenario, but aren’t there a couple of issues with the design?

When the op says that the BMS is "charge current limited to 20A" I suspect he means that the BMS is rated for 20A charge current, in which case the 80A alternator will blow up the BMS and/or the alternator will overheat when it is run continuously at full output

Also, when the BMS trips out on full battery, boat electrics will get disconnected (i.e. instruments, nav lights, vhs etc. ) will shut down without warning - not ideal
 

Poey50

Well-known member
Joined
26 Apr 2016
Messages
2,309
Location
Chichester
Visit site
I would also question the cost/benefit of swapping to Lithium in this scenario, but aren’t there a couple of issues with the design?

When the op says that the BMS is "charge current limited to 20A" I suspect he means that the BMS is rated for 20A charge current, in which case the 80A alternator will blow up the BMS and/or the alternator will overheat when it is run continuously at full output

Also, when the BMS trips out on full battery, boat electrics will get disconnected (i.e. instruments, nav lights, vhs etc. ) will shut down without warning - not ideal
.
You are right to point out that the charge limit on the BMS is poorly matched to the alternator. But the only good reason for leaving a lead acid battery permanently in the system is to prevent a voltage spike that will kill the alternator diodes when there is a BMS disconnect. It's not clear to me how the hybrid of lead acid and LFP will respond to the mismatch but it will not kill the battery or the BMS, or melt the alternator (if the charge is limited as proposed); it will simply disconnect when the charge limit is reached. The diagram does not show the loads but these should be connected to the the LFP in which case they will not shut down when the LFP is isolated as long as charge shutdown still allows discharge as is the case with, for example, the Overkill BMS. If high voltage disconnect isolates the battery completely with no discharge possible then the system is unworkable as you suggest.

Just a note on whether it is worth switching from lead acid to LFP which Tranona raises. It is the case that even a 100ah LFP has benefits in terms of the additional usable capacity it offers (around double that of lead acid), its ability to harvest solar more efficiently, and the non-issue of sulfation and corresponding tolerance of lower states of charge. However the great virtue of lead acid is that, while it is unwise to neglect it and be imprecise in meeting it's needs, it can mostly be done safely. LFP has much higher demands for being treated properly to get the full value from the investment. Those choosing cheaper drop-ins may find it a quick way to waste money if the potential longevity can't be realised.
 
Last edited:

ruvane

Member
Joined
24 Nov 2005
Messages
188
Location
Ashkelon Marina
www.techno-laser.com
Thanks everyone for the feedback - much appreciated. Rather than creating individual posts to address each point I'll try to do it all in one post.

The smart BMS is rated at 150A but I can limit the charge current to anything up to this. 20A seems to be safe to start with and measure things as I go.

I'll contact Merlin about Smartbank being used in this way.

I'm sick of lead acid as the prices here are truly exorbitant and the quality of even makes such as Varta is awful.

I've had the lead acid fail during a sail and I can't see how this is better than having the LiFePo4 fail in the same way.

The house load will be available even when the BMS stops charging as this is programmable.

Is there anything serious I'm missing? TIA
 

migs

Active member
Joined
5 Jul 2011
Messages
130
Visit site
Ok; a BMS that can control charge current. What type is it? And how does it control the charge current (maybe communication with the alternator regulator and mppt controller)?
 

Poey50

Well-known member
Joined
26 Apr 2016
Messages
2,309
Location
Chichester
Visit site
Thanks everyone for the feedback - much appreciated. Rather than creating individual posts to address each point I'll try to do it all in one post.

The smart BMS is rated at 150A but I can limit the charge current to anything up to this. 20A seems to be safe to start with and measure things as I go.

I'll contact Merlin about Smartbank being used in this way.

I'm sick of lead acid as the prices here are truly exorbitant and the quality of even makes such as Varta is awful.

I've had the lead acid fail during a sail and I can't see how this is better than having the LiFePo4 fail in the same way.

The house load will be available even when the BMS stops charging as this is programmable.

Is there anything serious I'm missing? TIA

For interest, what LFP are you considering?

Lead acid usually fails more slowly than LFP - voltage sag is generally the warning sign e.g. lights fading. A BMS inspired blackout is sudden and can have multiple causes some of which the user has no control over.

I have a SmartGuage but had to retire that from use with the house bank when converting to LFP as the algorithm is lead-acid specific. As this remains the brains of the SmartBank Advanced that could well be a problem.

Here is the Zwerfcat link that I mentioned above. Lithium-Hybrid
 

Tranona

Well-known member
Joined
10 Nov 2007
Messages
41,065
Visit site
I've had the lead acid fail during a sail and I can't see how this is better than having the LiFePo4 fail in the same way.

There is something seriously wrong if you have LA "fail" while sailing. Certainly not my experience and no doubt thousands of others. Either FLA or AGMs are extremely reliable and predictable if you look after them, which in itself is not difficult. The AGMs in my current boat are 6 years old and no sign of declining capacity. The cycle life of LA batteries in "normal" cruising use ranges from 5-10 years depending on the specific type. on the price front an AGM bank with a usable capacity (50% discharge) is about 35% of the equivalent LP. Difficult to see what benefits you gain from spending 3 times as much.

As you have only one house battery this suggests that you either have very modest demands or you are overworking an under capacity bank. Rather than jump into a very expensive way of solving what is an undefined problem, perhaps go back to basics and identify your power requirements and look for the most appropriate solution.
 

Poey50

Well-known member
Joined
26 Apr 2016
Messages
2,309
Location
Chichester
Visit site
Ok; a BMS that can control charge current. What type is it? And how does it control the charge current (maybe communication with the alternator regulator and mppt controller)?

BMSs which directly control chargers are generally high-end kit. I'm assuming the OP has something like a Daly Smart 150amp BMS. You can set a charge limit on this but it is basically on or off. When "off" the alternator then simply charges the lead acid start battery in a hybrid set-up..
 
Last edited:

migs

Active member
Joined
5 Jul 2011
Messages
130
Visit site
Given a BMS with a maximum charge current feature, and a charge current greater than the limit set, wouldn’t charging always be shut down resulting in no charge at all?
 

Poey50

Well-known member
Joined
26 Apr 2016
Messages
2,309
Location
Chichester
Visit site
Given a BMS with a maximum charge current feature, and a charge current greater than the limit set, wouldn’t charging always be shut down resulting in no charge at all?

Yes, that looks like the main risk if the charge current can't be limited at source. The fact that LFP is not self-limiting for charge is one of the major differences to lead acid. The latter will only take the charge that its internal resistance will allow, LFP will take everything thrown at it for much longer than is good for it.
 
Last edited:

ruvane

Member
Joined
24 Nov 2005
Messages
188
Location
Ashkelon Marina
www.techno-laser.com
It looks like there's a fatal flaw, as I suspected! As Migs points out, charging might indeed never happen as the BMS will simply disable charging with the alternator trying to supply full current.

That leaves an external DC-DC charger limiting current to the BMS.

Thanks Boater Sam, I'll go see what the canalboat guys are doing.

And thanks to everyone for the input.
 
Top