Design of AGM & LiFePO4 / Solar system for my boat

[tpenfield]

Greetings,

I'm new to the forum and looking for some guidance. I have a twin engine boat that has a fairly complex electrical system and has 7 AGM batteries divided into 3 banks (originally it was 4 banks, but I joined 3 & 4 into a single bank). I added a 100 watt Renogy solar charging system with a 10 amp Voyager controller feeding into a Victron ArgoFET 3 bank isolator. The system has worked well in maintaining the batteries, as the boat sits on a mooring away from shore power when not in use. The boat was made in 2015 and the batteries (original) are at their end of life (~ 9 years).

My plan is to re-design the electrical power system to have 2 AGM batteries (one of each engine) and 2 LiFePO4 batteries (100Ah) to supply the house & 2000 watt inverter (Xantrex SW2012). The engine alternators are rated at 70 amp and feed into a 2 alternator/3 bank isolator.

My design challenge has been to come up with a system that can charge/maintain the batteries via solar when the boat is not in use and then rely on the alternators for charging when the engines are running, and shore power when the boat is at a slip for overnights. I also have a ProNautic 1230 shore power charger when the boat is at a slip (occasional overnights). When on shore power, both the ProNautic and the Xantrex can charge batteries.

I'm fairly new to the details of lithium battery technology, but have learned the basics. After some guidance from several electrical engineering types, I have come up with the following schematic . . .

Looking for comments and guidance. . . . will it work? If not what changes may be advised to make things work (work better). Thanks in Advance . . .

 

[jdege]

I claim no expertise, but if you're looking to create a mixed lead-acid/lithium system, you should take a look at Emily&Clark's Battery Bank Manager System.

BankManager™ System

You may not decide to get one, but reading through his explanations of why he designed it will probably help.

 

[wholybee]

Ugg. Don't bother with the Bank Manager or similar hybrid systems. They are nonsense; a solution looking for a problem.

On a high level, one approach is to connect all your charge sources to one battery, then use a DCDC to charge the other. So, you could connect both alternators and the solar to the LFP(would require some alternator changes/protection), and use a DCDC to charge the AGM. Or, connect everything to the AGM, and use a DCDC to charge the LFP.

Or connect the Alternators to the AGM, use a DCDC to charge the LFP from the alternator. Solar could be connected to the LFP, but then it would then not charge the AGM. Adding a second DCDC for that would result in circular current and just drain everything.

Specific to your drawing, I would have each alternator charge it's own battery and not use the isolator. Then have 2 DCDCs, one for each AGM, to charge the LFP from the alternator. I would give up the solar for the start batteries and have solar direct to the LFP. There is no draw on the start batteries except starting, so they should be able to go a month or more just sitting. I would add a parallel switch, so that if one dies, you can parallel them together. If you are really concerned, add a small 50W panel to just the AGMs.

I don't understand the AGM feeding an input to the MPPT. That won't work. I also don't understand the "solar active" signal.

 

[tpenfield]

Thanks. I did have a design that used the AGM's to charge the LFP's (which would really be the alternators providing the power source).

The Renogy unit is a (newer) 'dual input' unit designed to combine solar charging and a starter battery (alternator) to charge a house battery (which can/would be an LFP).

I just encountered in the documentation of the Renogy charger that once the house battery is charged, it will trickle charge the starter battery (this would be with engine off and just getting solar as the input source). It might simplify things . . . as the AGM's typically would only need trickle charging in between uses of the boat and the LFP's would not loose much charge.

My boat does seem to use small amounts of power when not in use, which I think is from the Xantrex, the Stereo keep-alive, and the corrosion prevention systems.

The solar active signal 'energizes' the FET Isolator (alternatively this can be a battery or ignition switch source). I set my current solar charging system up this way, so the system 'shuts off' at night, turns itself back on in the morning, and does not use battery power to turn on the FET isolator.

 

[jdege]

Ugg. Don't bother with the Bank Manager or similar hybrid systems. They are nonsense; a solution looking for a problem.

The "problem" is that none of the commercial LiFePO4 charging options follow the manufacturers' recommendations with regard to charging regimes, because they're designed to charge based on voltage levels, and LiFePO4 cannot be safely charged using voltage levels to determine battery state. Or so his readings of the data sheets would suggest.

 

[wholybee]

The "problem" is that none of the commercial LiFePO4 charging options follow the manufacturers' recommendations with regard to charging regimes, because they're designed to charge based on voltage levels, and LiFePO4 cannot be safely charged using voltage levels to determine battery state. Or so his readings of the data sheets would suggest.

Thousands of boats have LFP installed without using a hybrid system. No problems. The hybrid LFP battery is snake oil, like the $15 dollar doodad that doubles your gas mileage on your car. The salesman makes it sound great, but it does nothing.

The data sheets I have seen use voltage. Granted I have only seen a few, but that's what they were. What's more, no one has demonstrated that charging by voltage is a problem. LFP are really durable, much more so than Lead. They are tough to kill from an incorrect charge. Many of those that installed Lithium Iron Phosphate Cells in their boats 15 years ago with privative or even no BMS, and charging by voltage, still have batteries that test at over 90% of original capacity.
However you chose to charge them, LFP are NOT damaged by undercharging them. 100% charge is defined by different manufactures as 3.7V or 3.8V. (https://en.winston-battery.com/?cnxdc/318.html) Practice is to charge to less than 3.65V. So it doesn't matter how you stop the charge, by current or voltage. The battery is fine as it isn't overcharging.
Seriously, TENS OF MILLIONS (maybe hundreds of M) of LFP batteries have been sold for several decades. If they were being charged incorrectly, we would know by now. The fact is that unless you do something grossly wrong they will die by calendar aging at more than 15 years old before they die because of the way you charge them.

 

[jdege]

Battery Bank Management System (BBMS) Technology: A Proposed Standard for Managing an Off-Grid Electrical System with Both Lead and LiFePO4 Battery Banks

 

[eXodus]

Seriously, TENS OF MILLIONS (maybe hundreds of M) of LFP batteries have been sold for several decades. If they were being charged incorrectly, we would know by now. The fact is that unless you do something grossly wrong they will die by calendar aging at more than 15 years old before they die because of the way you charge them.

that happens sometimes - but it's mainly in consumer grade products. Where some Smartphones, Tables or laptops randomly caught fire and recalls had been issued.

It usually is big in the press.

Most of the concerns with engine based systems is towards "protecting the alternator"
Which can be achieved in various ways, DC-DC is one option. Bank Manager is another one.

Lately my research is going towards current limiting BMS and high current diodes. (aka isolator) Not many batteries have that yet.

Which essentially mean - actively limiting the current in and out of the LFP, which is the main problem. If that would be solved within the BMS - we could just connect the lead and LFP permanent and don't worry about it.

 

[mikefitz]

Looking for comments and guidance

I may not be following the diagram correctly but you may have a charging loop.
STB engine running, associated AGM charges, DCC30 triggers on , charges lithium. Rising voltage on lithium battery triggers on DC to DC, output via fet isolator loops back to starter AGM battery.

Ususal setup for engine charging is, engine charges starter lead acid batteries , DC to DC chargers from starter battery to lithium house battery bank.
Solar controller charges house battery bank only.
Shore power charges house bank.

To provide a maintaining charge to starter batteries a DC to DC charger from house batteries to starter batteries.
..

 

[grizzzman]

Thousands of boats have LFP installed without using a hybrid system. No problems. The hybrid LFP battery is snake oil, like the $15 dollar doodad that doubles your gas mileage on your car. The salesman makes it sound great, but it does nothing.

I disagree. I have been running a hybrid system tied to a FLA CG2 while camping for 5 years now. There have been times that my drip coffee maker would not have worked in the morning if I did not have this set up. ( You would not like me without coffee in the morning.... Redundancy is as far as I'm concerned mandatory on the water.

The data sheets I have seen use voltage. Granted I have only seen a few, but that's what they were. What's more, no one has demonstrated that charging by voltage is a problem. LFP are really durable, much more so than Lead. They are tough to kill from an incorrect charge. Many of those that installed Lithium Iron Phosphate Cells in their boats 15 years ago with privative or even no BMS, and charging by voltage, still have batteries that test at over 90% of original capacity.
However you chose to charge them, LFP are NOT damaged by undercharging them. 100% charge is defined by different manufactures as 3.7V or 3.8V. (https://en.winston-battery.com/?cnxdc/318.html) Practice is to charge to less than 3.65V. So it doesn't matter how you stop the charge, by current or voltage. The battery is fine as it isn't overcharging.

That is a Yttrium doped electrolyte that allows for lower temp. charging and a higher charged voltage and is not generally used.

Seriously, TENS OF MILLIONS (maybe hundreds of M) of LFP batteries have been sold for several decades. If they were being charged incorrectly, we would know by now. The fact is that unless you do something grossly wrong they will die by calendar aging at more than 15 years old before they die because of the way you charge them.

Once again I disagree.

 

[jdege]

For me, the critical issue might be the risk of unexpected shutdown.

There really is no way to know when a BMS is likely to just turn off.

 

[wholybee]

For me, the critical issue might be the risk of unexpected shutdown.

There really is no way to know when a BMS is likely to just turn off.

That is ridiculous. A BMS will shut down when there is a problem. In a well designed system, it should never happen. My system is 4 years old. I have had one shutdown when I had to leave the boat for 3 weeks and it rained the whole time so my solar didn't work. Just as it should have. No drama. If you think there is no way to know when a BMS will shutdown, you simply don't understand the technology.

LFP is *FAR* more reliable than Lead Acid.

 

[wholybee]

I disagree. I have been running a hybrid system tied to a FLA CG2 while camping for 5 years now. There have been times that my drip coffee maker would not have worked in the morning if I did not have this set up. ( You would not like me without coffee in the morning.... Redundancy is as far as I'm concerned mandatory on the water.

That is a Yttrium doped electrolyte that allows for lower temp. charging and a higher charged voltage and is not generally used.

Once again I disagree.

Disagree if you like. If you think an LFP system without being a hybrid would cause your drip coffee maker to not work, you did something wrong. There are far more pure LFP systems than hybrid. And people need their coffee. It just isn't a common complaint.

I have lived aboard full time with LFP for 4 years. I have never had an unexpected shutdown. I have never had the LFP fail me and leave me without power. And that statement is repeated by boater after boater. "Lead is Dead" is often heard, because LFP requires so little attention or maintenance or paying attention to at all. It just works. Yeah, redundancy is good and I am all for that, but a hybrid system isn't redundancy.

I don't doubt your hybrid system works. But did you actually have a pure LFP system first, and have problems with it?

 

[grizzzman]

Disagree if you like. If you think an LFP system without being a hybrid would cause your drip coffee maker to not work,

Where did you get this idea from?

you did something wrong.

You are correct, I put my trust in a Chargery BMS ( The problem was corrected in a software re-flash ) I will not bother you with other issues.

There are far more pure LFP systems than hybrid.

Yes and 5 years ago many on this forum said "You can not do that" I decided to see for myself.

And people need their coffee. It just isn't a common complaint.

True

I have lived aboard full time with LFP for 4 years. I have never had an unexpected shutdown. I have never had the LFP fail me and leave me without power.

It is always a good thing to have total success. Can you say that to every LFP system out there?

And that statement is repeated by boater after boater. "Lead is Dead" is often heard, because LFP requires so little attention or maintenance or paying attention to at all. It just works. Yeah, redundancy is good and I am all for that, but a hybrid system isn't redundancy.

I don't doubt your hybrid system works. But did you actually have a pure LFP system first, and have problems with it?

No

 

[tpenfield]

Thanks for everyone's help on this. I think I am all set on the design.

I was hoping the Victron Orion XS units would be readily available when I am buying the components, but I may have to go with the older model (Smart TR) based on availability over the next couple of months.

I am liking all of the features of the Renogy Dual Input unit and Will has a nice review of it in his video series.

 

[markpj23]

That is ridiculous. A BMS will shut down when there is a problem. In a well designed system, it should never happen. My system is 4 years old. I have had one shutdown when I had to leave the boat for 3 weeks and it rained the whole time so my solar didn't work. Just as it should have. No drama. If you think there is no way to know when a BMS will shutdown, you simply don't understand the technology.

LFP is *FAR* more reliable than Lead Acid.

deleted so as to not hijack the thread....

 

[tpenfield]

OK - Quick Update . . .

The system is done and installed.

I went with the EPEver DuoRacer solar controller instead of the Renogy, and the Orion TR Smart DC-DC Charger (Looks like the XS is not readily available yet). Here is a picture of the component panel with a few finishing touches remaining . . . (post & bus bar caps, etc)
.

.
500 lbs of '31M-AGM' batteries (7 total) came out and 150 lbs of batteries (2 LFP and 2 LA) went back in, for a decent weight savings. . . . I sold the batteries and the spare boxes.
.


.
Here are the batteries being positioned. I re-purposed as much of the 2/0 cable as I could. The batteries are now in 4 of the boxes that are from the original 7 battery layout. (not shown)
.

.

.
I used the LiTime TM 100 Ah batteries, since they have the LTCP feature, which the other 100 Ah batteries do not have (yet). Will P. did a review of the batteries about 7 months ago.

Here is the final component layout & schematic for those following along at home. . .
.

.
Full System Schematic. . .

.
After some testing and a few tweaks, it is all good. The MPPT solar controller does a nice job is squeezing power from the 100 watt solar panel.

I need to re-position some of the original battery and device labeling, just for completeness.

 

[tpenfield]

Here is the engine bay as it was originally . . . 7 (seven) 31-AGM batteries at 72 lbs each. Lots of weight and not much open space. There were originally 4 battery banks

Starboard Engine ('Fred') - 1 Battery
Port Engine ('Barney') - 1 Battery
Accessory - 1 Battery
Inverter - 4 Batteries
.

.
I combined the Inverter and Accessory banks into one (now called 'House' bank) and switched them over to the 2 LiFePO4 batteries (in parallel). The engine banks are on the 2 L.A. batteries. I have slightly less usable Amp-hour capacity for accessory & inverter (180 ish vs 250), but not really a problem given my typical usage of the boat.

 

[grizzzman]

Why did you decide to stop solar charging when the engines are running?

 

[tpenfield]

Since the chargers are quite different I did not consider it a best practice to have them both going at once. Some technical bulletins advise caution in using dissimilar chargers in parallel. I came across stuff like this . . .

"Important: Not all chargers can be paralleled safely! Consult with your charger manufacturer before paralleling chargers. Non-isolated chargers in particular may be dangerous in a parallel configuration."

Plus, the extra electrons from the solar panel is of little benefit once the engines are running. I've had some folks advise that it would not matter having the chargers in parallel, but they have been known to be wrong.

 

[SparkyGage]

Umm,... it looks great, but I'm seeing two things that concern me: 1)I don't see hold down straps on the batteries, and 2) if I am correct that the LiPo's are in parallel, I would think it would be better for them to be cabled parallel at the batteries themselves and run the Neg cable from the system to the Neg of one battery and the Pos of the system to the Pos of the OTHER battery so that the draw load and the charge remains even across both batteries.

 

[tpenfield]

Thanks for your comments.

Actually the batteries are now in battery boxes which are screwed into the battery shelf. The picture shown is one of the more final work-in-progress pictures, but not completed. As for the battery connections, I did hook them up as per the manufacturer's instructions, and all cables to the LiFePO batteries are the same length, but they are a bit longer than I'd like (33 inches). I'll consider the method you described as it may be a bit cleaner.

 

[grizzzman]

Thanks for your comments.

Actually the batteries are now in battery boxes which are screwed into the battery shelf. The picture shown is one of the more final work-in-progress pictures, but not completed. As for the battery connections, I did hook them up as per the manufacturer's instructions, and all cables to the LiFePO batteries are the same length, but they are a bit longer than I'd like (33 inches). I'll consider the method you described as it may be a bit cleaner.

I suggest that you heed tpenfield's cabling advice.

 

[tpenfield]

Here are a couple of 'finished' pictures . . . battery boxes and ID labels in place.

The system seems to be working well. Not much solar, due to cloudy/rainy days. I have not fired up the engines either, since overnight temperatures still get a bit cold.
.

.

.

 

[SparkyGage]

I still think your Parallel batteries are wrong (sub optimal). That Yellow going from the pair (Positive?) should be connected to the other battery in the pair, so the power, goes in one battery and out the other, that way the load and charge is spread evenly across both batteries. The way you have it, most of the load and charge is hitting the first battery and the one in parallel is just boosting it. The front battery is likely to starting having a different state of charge than the back one in this configuration.