LFP/Solar chat - deals, design, Q&A, etc

[rslifkin]

you are saying the LFP is directly connected to the start FLA without a DC-DC and therefore directly to the alternator in this setup you allow to BMS to shut off charge when it does not want any more.
This goes against all other opinions that say a standard ALT cannot run full power continuously without burning up.
I can only see this working with a small depletion of the house bank where the recharge is fast and the ALT is no longer in high charge mode.

It depends on the alternator regulator. If it has a good external regulator with alternator temperature sensing, it will limit the output current as needed to avoid overheating the alternator. You won't get full output continuously from most alternators that way, but compared to the DC-DC you'll get a bit more, as it'll be able to give a bit more power on a cooler day, or for the first bit of runtime until everything is up to temp.

 

[Iggy]

My understanding is LFP can accept much higher charge rates than LA - one of it's most desirable features. Does putting a LA in front of a LFP effectively kill that benefit? No need to groom the charge profile off the alternator to suit the needs of an LFP battery? Would be a simple solution - since it goes against orthodoxy, a credible citation would be great

Peter

Yes, on the other hand. LFP will in general charge faster at the same amps as AGMs. Also, being on the hook or at anchor. I can start the genset and use the on board charger. Delivering, more than 120 amps. Better than the DC2DC charger at 60A. Keep in mind, my running usage is about 25 to 30A. For me this is all good.

 

[DDW]

I think you mean to say an LFP will charge to full faster from the same charge source than AGM? Because at the same amp charge rate, it's the same charge rate.

I very much agree that there are disparate installations trying to solve different problems and with different goals. But if you have spent the money on LFP, certainly spend the money on an externally regulated alternator. Pretty much required if the alternator is hooked to the LFP bank directly, just to maintain the correct voltages.

In this and other threads there have been discussed two methods of saving the alternator from ruin when charging LFP: one is to throttle the charge to the LFP though a DC-DC, the other is to derate the alternator output by actively measuring its temperature. I much prefer the latter for a variety of reasons. On a small system, the DC-DC may be a little less expensive, on a larger system the externally controlled alternator will be cheaper.

 

[rslifkin]

I think you mean to say an LFP will charge to full faster from the same charge source than AGM? Because at the same amp charge rate, it's the same charge rate.

It will charge faster due to much less taper at the end of the charge. But LFP charging is also more efficient. So putting 100A into a lead avoid battery (including AGM) for an hour will put a bit less actual power into to the battery than doing the same with LFP. It's not a huge difference in practice though, mostly academic.

 

[DDW]

Yes as I said it will charge faster to full. From say 40% SOC to 80% SOC not very different, the difference in efficiency is negligible for most of us. But also importantly, it isn't important to charge to full on the LFP, while the AGM requires it (for any lifetime anyway). The old school sailboat thinking was to use flooded LA, and cycle between about 50% and 80%. And occasionally an equalize cycle. That works nearly as well operationally as LFP, except that you need a lot of flooded cells (as you are only using 30% of the capacity). AGM changed that because cycling like that will fairly quickly kill them.

 

[Mambo42]

But also importantly, it isn't important to charge to full on the LFP, while the AGM requires it (for any lifetime anyway).

That is what I thought as well, until I was told just recently by a Victron specialist that I have to charge the LFP's to full at least once a week. I also used to charge them to about 90 % and then stop the charge, thinking it would not be calculated as a cycle, which would increase lifetime.
However, apparently not charging to 100 % and also balancing the cells, will result in the battery thinking that a lower state is now 100 %. You don't lose a lot, he told me is maybe about 1 % capacity each time, but as soon as you go to full again and balance the cells you will get back to the full capacity of the battery again.

I have no idea if this is true or not, but the guy is known in the Netherlands as a Guru on solar and LFP, so I tend to believe his reasoning until someone can explain me he is wrong.
In other words, not charging to 100 % will cost you battery capacity with LFP as well. Good part is that you can get it back and that is not the case with other types of batteries.

 

[rslifkin]

You won't lose actual cell capacity by not charging to 100%, but if you go a long time without hitting full charge then the cells may start to drift out of balance, which would cost you a little bit of capacity as one might hit low voltage cutoff before the others when fully drained.

The bigger concern is metering accuracy. SOC metering will often drift a bit if not synced-up at 100% periodically, so if you go too long without hitting 100% you won't have an accurate indication of how much power is actually in the batteries (from either the BMS or a separate shunt based monitor).

In my mind, you don't need to avoid hitting 100% necessarily, but also don't worry much about how often you hit 100% charge as long as it happens occasionally and you're not seeing balancing problems when you do get to full charge.

 

[DDW]

As general advice charging to 100% to achieve balance seems the consensus. Will Prouse has a whole video on it.

But with newer batteries, there is no mystery about cell balance, you can read it off your (BLE app) cell phone at any time. You don't permanently lose capacity like an AGM, but the BMS will disconnect when the highest cell goes over voltage, leaving the others (perhaps) undercharged.

I think if you want to ignore all the visibility that batteries now have and treat them like an AGM or LA, you probably should charge them to 100% periodically. But it seems better is to lower the absorb charge voltage to say 14.1V, and set an absorb charge time of 1/2 - 1 hour. Then make sure the BMS is set up to balance cells starting a little below this voltage. I think this is better for them: you aren't banging the corners of the operating envelop, not much chance of a BMS disconnect, and plenty of opportunity to balance. I've been running an Epoch this way for a little while and the cell balance is always around 0.002V and has never more than 0.010V , probably limited by the measurement method.

 

[SteveK]

Bluetooth monitoring, suppose the SOC says 100% but you app says 2 or more cells are not equal to the rest, what will you do?
Will Prowse did make a firm statement that while you do not need to charge to 100% you can every time and not hurt the batteries. He went on to say BMS balancing starts after 95% as I recall.
While I agree you can set the absorb voltage to whatever, I did not know you can tell the BMS at what voltage to balance.

 

[Marco Flamingo]

I did not know you can tell the BMS at what voltage to balance.

Yes, with some apps. That's what makes the Overkill app so nice. LFP arrive with a plain vanilla BMS program set for some generic use, like maybe a large shore-side solar array.

I think of the BMS SOC "guesstimate" as the least informative measurement. In fact, with the Overkill app, I can change the % SOC numbers to alter the stock guesstimates. That shows the calculated % given is a flexible metric. A 90% SOC could really be off by 5-10 percent given irradic usage and alternator charging on a boat.

I have slowly weaned myself from fixating on the calculated SOC reading. It is another "issue" that is eliminated by my having much too much LFP amperage. With over 1,100A onboard and a daily usage of 150A, I don't care if my 90% is really 75%. That will be rectified in a few cycles.

 

[rslifkin]

But it seems better is to lower the absorb charge voltage to say 14.1V, and set an absorb charge time of 1/2 - 1 hour. Then make sure the BMS is set up to balance cells starting a little below this voltage. I think this is better for them: you aren't banging the corners of the operating envelop, not much chance of a BMS disconnect, and plenty of opportunity to balance. I've been running an Epoch this way for a little while and the cell balance is always around 0.002V and has never more than 0.010V , probably limited by the measurement method.

This will still get you 100% charged. You don't have to get all the way to 3.65vpc to reach full charge. That just gets you there faster. But a somewhat lower voltage (like the 3.525vpc you're charging to) will still get you there, just takes slightly longer (it'll have a longer absorb period to reach full charge).

 

[DDW]

Plenty of tests to look at (and fairly easy to do your own). The difference in absorb time to 100% between 14.1 and 14.6 is small, just a few minutes. And the extra minutes add to the time the balancers are working. In addition, you can set the absorb time to 30 - 60 minutes at 14.1 without much worry of a disconnect, to allow balancing to occur. Holding them at 14.6 for an hour risks disconnect.

 

[DDW]

Bluetooth monitoring, suppose the SOC says 100% but you app says 2 or more cells are not equal to the rest, what will you do?
Will Prowse did make a firm statement that while you do not need to charge to 100% you can every time and not hurt the batteries. He went on to say BMS balancing starts after 95% as I recall.
While I agree you can set the absorb voltage to whatever, I did not know you can tell the BMS at what voltage to balance.

Depends on the BMS and your visibility into it. Using the Overkill app on the Epoch, I can see when the balancers are scheduled to work, and when each is turned on. You can also change those values, but they looked fine to me as delivered. Just turn your absorb voltage down a bit and give them time to work.

 

[fryedaze]

Sorry I am behind on posting responses.
Here are clarifications to some questions.
1. My system charges the Li Bank directly off the Pb acid start batteries. The alternators, chargers and solar charge the start bank.
2. The Li Bank protected from overcharge by the BankManager system that controls a 200 amp contactor between the Pb and Li.
3. We have two 120 amp alternators with Sterling regulators. They have Hi temp protection. There is no danger of burning up alternators. The routinely charge at 200 amps on start up and quickly drop off.
4. If my BMS shuts down Lithium the start batteries prevent "Black Ship"
5. If the BankManager fails, there is an override for the 200 amp contactor between Pb and Li.
6. The start batteries are not discharged. The Li discharge profile voltage stays higher than Pb . The Pb essentially floats on the Li during discharge. If Li gets down to 12.7 vdc they are probably at 20% SOC or less.


A day in the life of our LiPo/ Pb BankManager system.
Our DC system on our 42 ft trawler Fryedaze consists of two Gp31 start batteries for each engine and two Epoch Essential 460 amphr house Bbank batteries. The BankManager contactor separates and combinds the Pb and LiPo batteries as determined by the BankManager (BBMS).
Charge sources are two 120 amp alternators, controlled by Sterling regulators, 3- 160 watt solar panels and a TrueCharge + 3 bank 40 amp charger. All charge sources have been set to comply with Epoch absorption and float voltage recommendations. All charge sources are connected to the Pb start batteries.

At the dock DC loads are less than 5 amps. The system floats on the solar and the Li and Pb are separated due to the Li being at 100% and the BBMS contactor opening. When the sun goes down the Pb voltage begins to drop. When the Pb and Li voltage are within about .2vdc the BBMS closes the contactor and ties the Pb and Li banks. This occurs at about 13.4vdc. The Pb floats on the LiPo overnight. When the sun comes up solar begins charging the Pb and Li. Overnight the Li bank drops about 50 amphrs. The BBMS calculates POC and when the Li is at 100% it again separates the Li and Pb banks and the system runs on the Pb and solar.

For a cruising day the banks are separated and the system runs off the alternators and the Pb batteries. At anchor when the engines are secured the Pb voltage will drop after solar can no longer support loads. At that point the BBMS will again tie the Li and Pb as the voltages are within the ~.2vdc range.

Fryedaze uses 150-250 amphrs overnight. 450 amphrs if AC is in service. At engine start for anchor pull the alternators begin charging at approx 190 amps and quickly lower to around 120 amps. The amps continue to lower as the alternators charge the Pb and Li banks. The majority of the charge time is between 50-70 amps. When the BBMS reaches 100% POC the Li and Pb are separated by the contactor. A 400 amphr recharge takes about 4 hrs. The Sterling regulator occasionally shutdown on alternators temperature.

At the dock we are not using the battery charger and let solar panels handle it. The 3 bank charger is maintained off and is also isolated from the Li bank and will probably only be used for cell balancing if needed.

We have had the system for four months. Recently we cruised 1300 miles in four week, where we spent 15 nights at anchor. The system has worked flawlessly for that period.
Just want to share our experience with the BBMS.

Video of system

Dave

 

[mvweebles]

Will Prowse just posted a video on fire risk of LiFePO4 batteries. In a nutshell, he says they are very safe - "A block of wood - it will burn if an external heat source but won't self-combust." That said, he emphasizes proper wiring and circuit protection......and a good BMS. It is a battery with a LOT of energy potential. But self-ignite? Nope....

Peter

 

[mvweebles]

I came across this side by side review of 200w rigid panels. Some very efficient and affordable options out there in a size common for boats.

He likes this Calsun bifacial N-Type for $180

Peter

 

[Mambo42]

I have seen quite a few reviews of bi-facial solar panels and basically they all come down to the same conclusion: these bi-facial panels barely have a higher output than a normal panel. Figures are between 4 and 10 % and the 10 % is then when you invest quite a bit in reflecting material and set up.
When I then think of the added risk of damaging the panel I can only come to the conclusion that it is not worth the money (at least for me when you want to use it on a boat).

 

[Marco Flamingo]

Worked great on my boat. 200W panel extends out from the SS railing. When the sun is on that side, light reflects up off the water. I've seen 220A, whereas most never seen full rating from a panel.

Hard to say if the improvement in partial shading for single face would be the way to go. I only have an unstayed mast but still odd how often it gets a shadow and output drops. Bifacial doesn't drop as much as single face (when properly installed).

 

[mvweebles]

All of the panels he reviewed were "new 'N-Type'" so it may be more than bifacial. If you looked at the review, the shadow beneath the panels had narrow slots so sunshine went through them. They were also divided somewhat to be more shade tolerant. Seems there's some newer technology improvements.

Anyone know what "N-Type" panels are?

Peter

 

[Bmarler]

I think it's getting into the weeds a bit when discussing n type or p type. But n type have more free electrons as a result of the make up of the cell. Gives it a negative bias( I'm not an expert) as opposed to p type.
Like I said, getting into the weeds of cell construction. Advances are being made so quickly in this field that no matter what you buy, it's obsolete in a month.

 

[mvweebles]

Interesting video from Will Prowse advising against Bluetooth enabled batteries. Apparently the folks on Solar DIY Forum are finding they do not load-share in a balanced manner. Strong recommendation for basic non-Bluetooth batteries.

He publishes a list of recommended batteries.

LiFePO4 Batteries for Offgrid and Mobile Solar

Tired of getting scammed by cheap LiFePO4 batteries? Not getting the capacity that was advertised or the features that were promised? Worried about safety? Check out my updated recommendations below!...

www.mobile-solarpower.com

Peter

 

[Barking Sands]

He is not correct. It has nothing to do with Bluetooth and just depends on whether or not the battery is programmed to have FCP or full charge protection. Something we have discussed here many times for nearly a year. You see people also describe it as a problem with "Smart" batteries.

Not all batterries labeled smart have FCP
Not all batteries that have bluetooth have FCP

You just have to find out via research before you buy.

 

[Iggy]

I am completing my 4th year of using my Blue Tooth batteries. So far so good. What I like and sorry, I did mention this before. One battery kept reading 0.7 lower than the rest. IF my memory is right. I had a bad fuse connection. I never would of found it if it was not for Blue Tooth.

Now on the small load thing, he is talking about budget batteries. I don't have that problem since having Blue Tooth I have seen as little as 1amp from LED light being on from each battery. Key being "budget".

Personal, I would not buy them! Safety is my first concern.

 

[mvweebles]

He is not correct. It has nothing to do with Bluetooth and just depends on whether or not the battery is programmed to have FCP or full charge protection. Something we have discussed here many times for nearly a year. You see people also describe it as a problem with "Smart" batteries.

Not all batterries labeled smart have FCP
Not all batteries that have bluetooth have FCP

You just have to find out via research before you buy.

Its become impossible to track all the batteries. Some that claim certain features (temp cutoff sensors, OCP, etc) fail the relevant tests so a buyer cannot go by manufacturers specs. Prowse notes that Forum feedback is this same discharge balancing issue is noted on all Bluetooth enabled batteries including his otherwise endorsed Litime. The only exception is the Epoch which he states in his later video he's still not endorsing because of the firmware brick problem from a few months ago.

I am not knowledgeable on the internal working of stuff like this. I trust @Barking Sands as he's been one of TFs most generous contributors, and a SME on the topic of LFP batteries (many thanks !). So my question to you is: Any suggestions on navigating the world of lower cost LFP batteries? Given Prowse ' testing, going without Bluetooth makes sense, at least for now. I wouldn't know if I had an issue with unbalanced discharge - I wouldnt think to put a clamp meter on each battery to see if they're discharging in unison.

Advice?

Peter

 

[SteveK]

Peter, I saw the video differently than you. Had to watch twice as I did not understand why he was paralleling with a different batteries, would you do that in a bank of batteries? Instead he was demonstrating the full charge shutdown feature which did not turn on with load if another source was present.

 

[fryedaze]

Going without Bluetooth doesn't make sense. The Bluetooth allows you to monitor battery performance. Knowing individual cell voltages and cell voltage differential is important. Bluetooth and the app doesnt control the battery or change settings unless you tell it to.

 

[Barking Sands]

I am not knowledgeable on the internal working of stuff like this. I trust @Barking Sands as he's been one of TFs most generous contributors, and a SME on the topic of LFP batteries (many thanks !). So my question to you is: Any suggestions on navigating the world of lower cost LFP batteries? Given Prowse ' testing, going without Bluetooth makes sense, at least for now. I wouldn't know if I had an issue with unbalanced discharge - I wouldnt think to put a clamp meter on each battery to see if they're discharging in unison.

Advice?

Peter

There is no tried and true way I know of. The Wattcycle 314 having FCP was a total surprise to me. I would have never guessed it. A couple of things to watch out for that may give a clue:

1) I dont know of any mid to low cost batteries using the tried-and-true JBD BMS that have FCP. This includes the Epoch Essentials and several other similar batteries. I have a Vatrer 300ah in the shop right now that is getting tested and I think it has a JBD BMS. I will test it for FCP. Usually you can spot a JBD BMS using Bluetooth as sometimes the model number is in some detail in the info in the app. If not, connecting with the Over Kill Solar app should show the BMS model. And if somehow a battery with a JBD BMS had FCP, connecting to the OKS app you should be able to delete it.

2) telltale signs of FCP are of course complaints of uneven discharge. But another one is to just charge until you get a mosfet disconnect and then check voltage at the terminals and compare it to voltage displayed in the app. If voltage at terminals is .4-.6 volts less than what the app says then it has FCP.

I suppose if you are eyeballing a certain battery, reading all the reviews you can to spot these odd behaviors is enough. But look how long it took for people to spot the Watt cycle.

I said this before and I could get in trouble for saying it again..if anyone has the Epoch 460V1 and wants to remove their FCP, just message me and we will get it removed.

 

[Barking Sands]

Peter, I saw the video differently than you. Had to watch twice as I did not understand why he was paralleling with a different batteries, would you do that in a bank of batteries? Instead he was demonstrating the full charge shutdown feature which did not turn on with load if another source was present.

Yeah..I am still not sure why he was using 2 different batteries.

 

[SteveK]

Yeah..I am still not sure why he was using 2 different batteries.

IMO, He was replying to an issue found by large solar panel users that have multiple brands of batteries. As long as one battery takes care of load, these fcp batteries will stay shut off.

 

[Barking Sands]

IMO, He was replying to an issue found by large solar panel users that have multiple brands of batteries. As long as one battery takes care of load, these fcp batteries will stay shut off.

Yeah...those off grid guys slap together anything and everything..lol.