LiFePO4 Batteries - Okay Tear Me Apart ;-)

[Tinkerer007]

I got the MiniBMS installed (wires everywhere, I have to figure out a good solution to package this up), and started charging.

I started with a car battery charger that got the battery from 13.1 V to 14.4. After I got my Xantrex 40 charger from the boat and hooked it up, it started at 20 A for about 20 minutes, and then dropped to 10 A for a couple hours. 5 hours later when I went to disconnect it, it was down to 1 A @ 14.6V.

Tonight I will try the equalize button, and if that doesn't work, there is a temperature switch that will increase voltage to 14.8.

The Hitachi 55A alternator on my Yanmar 30 only puts out 14.4 V. How much capacity will I loose by not being able to top the battery up? I am expecting not to charge the last 10% anyway due to the decreased efficiencies at the end of a charging cycle. My preliminary energy budget is ~80 AH per day. So I am hoping I can replace that with 2 hours of engine time per day.

@ Electric1: What voltage is the HVC event set at on my MiniBMS?

[s/v Thea]

If your only charging at 1A @ 14.6 volt, sound like batteries are already as fully charged as your going to get them. Do you have meter you can attach to them, then attach a load and discharge to LVC, then see how may amp-hrs you get out of them?

[Tinkerer007]

I have a Digital Multi meter that can measure to 10A. So if I could figure out a 10A load, that would take 14 hours to get down to a theoretical 30% SOC.

Any recommendations on a suitable load I might find in my garage? I don't think the siren I have laying around will be an appreciated load.

Or, I can wait till I install in the boat and use the Link 2000 to watch it. I could leave the fridge on for a load.

[s/v Thea]

Couple of car headlights? Cigarette lighter element from your car?

[s/v Jedi]

Originally Posted by s/v Thea

If your only charging at 1A @ 14.6 volt, sound like batteries are already as fully charged as your going to get them. Do you have meter you can attach to them, then attach a load and discharge to LVC, then see how may amp-hrs you get out of them?

See, that is in theory not true. The low current means that the batteries can't be charged further at this voltage but if the voltage would be brought up to 14.7V they would charge again. It is only when the voltage is at the maximum allowed and the current tapers off to almost nothing that the battery can be declared fully charged.

For LFP however, I'm told that a full charge isn't really needed and some say it evens shortens their lifespan, which would make 14.6V a good compromise.

cheers,
Nick.

[LiFeTech Energy]

Hi Guys (and Girls),
This is my first post here after being alerted to this forum by a customer.
I have had a quick read through this thread and there appears to be lots of confusion and misinformation when it comes to LiFePO4 batteries.

I will answer as many questions as I can here so please feel free to ask anything you are not sure about when it comes to Lithium Iron Phosphate (LiFePO4) battery technology.

I understand I am not allowed to mention anything of a commercial nature here so I will try my best to speak in general terms only.

A bit of background about myself and my qualifications/experience. I am the general manager and senior electrical engineer for the world's leading manufacturer of LiFePO4 batteries and associated products. We are the official licensees for the patent holder (Phostech/Hydro Quebec Canada) from the inventor of the LiFePO4 battery (Dr John Goodenough from the University of Texas, USA)

Before becoming involved in the LiFePO4 battery industry I was a senior electrical engineer in the critical power protection (UPS) industry and responsible for installation, testing and maintenance of lead acid batteries used to provide critical backup power for many well known companies including IBM, Apple Computer, Coca Cola, McDonalds, GE and several government departments and hospitals.

It is true the DIY electric car guys are far ahead with the understanding and real life experience using LiFePO4 batteries so hopefully the boating fraternity will catch up soon.

There are lots of websites with technical information for those wishing to research LiFePO4 batteries and their advantages over lead acid batteries.
For starters the Australian Electric Boating Association (EBAA) is a wealth of information not just for those wishing to build a fully electric boat but as a general guide for those contemplating using lithium batteries as a replacement for lead acid batteries for house power in sailing boats.
The EBAA website is at- Electric Boat Association of Australia

I would suggest anyone considering using LiFePO4 batteries do lots of research since there are some very good products on the market and some very poor quality products from China. Buying lithium batteries is a bit like buying running shoes. You can buy a pair of genuine Nike running shoes or a cheap copy made in China which will fall apart much faster than the genuine article. Many people have been burned as a result of making a poor choice by buying a cheap lithium battery thinking they would be saving money by not using one of the more expensive brands. When it comes to your lithium battery investment you really do only get "what you pay for".

[LiFeTech Energy]

I see that my membership status should here be changed to vendor status. I am not sure how to change my status so could a moderator please change my status or otherwise provide instructions how I can do this myself.

Thanks.

[electric1]

Originally Posted by Tinkerer007

@ Electric1: What voltage is the HVC event set at on my MiniBMS?

Since you have Thundersky cells, your HVC is 3.8V per cell. As I have been repeatedly saying here, under normal use you will never see HVC. HVC is like an insurance policy against severe imbalance, which you will never see unless you abuse your pack. You'd have to charge to 15.2V to get an HVC. Other cells like HiPower and SkyEnergy have HVC at 3.6V , or 14.4V per bank. Its more likely to get HVC when using those cells. Mind you, these HVC levels are still below max safe voltage per manufacturer's datasheets, so even hitting HVC every day would be perfectly normal and not have any negative impact on your cells.

Specifically for Thundersky cells the "knee" during charging is 3.6V , this is the point when cells are practically full ( 95% or so ), at this stage if you continue to charge, you would see rapid voltage rising and hit HVC within few minutes. So, from practical every day point of view you can consider your cells as "full" at 3.6V or 14.4V per bank. Charging them further is a waste of time and energy.

LifeTech, nice to see you here, I know you from DIY EV forum. I agree that you generally get what you pay for, but I would disagree that everything from China is poor quality. Most of the Western world lives on stuff made in China, you just need to pay attention to what you buy and who you work with. I understand that you are promoting highest quality product and its what I call a "Cadillac solution" , but there is a big world out there and plenty of people who are perfectly satisfied with "Honda solution". My product falls into that category and I stand behind my quality and reputation and will not let anyone bash it just because it costs less than alternatives. I do my business with minimum overhead, but I never compromise on quality and support, so please try to keep a humble attitude when referring to your competition.

[s/v Thea]

Originally Posted by electric1

Specifically for Thundersky cells the "knee" during charging is 3.6V , this is the point when cells are practically full ( 95% or so ), .

Electric1,

I'm looking at Thundersky's charge/discharge curves in literature for the 200AH battery, looks to me like they show charging at 0.5C until voltage is over 4V, and at 3.6V/cell they show batteries about 60% charged. Also on discharge curve at 90% charged and 0.3C discharge rate they show voltage of about 3.75V.

If this data is correct then looks to me like it would be tough to bulk charge at a fairly high rate up to even 80% using LA chargers?

Doug

[hellosailor]

LifeTech, it is nice to have a professional join us, thank you.

I've been less than comfortable with the failures of LiOn batteries in laptops. In my experience, even prime cells often fail around 4 years of age, regardless of use or lack of use, even without significant use. And that's from packs built with onboard computers and charging cell by cell. One day they show 90% capacity, the next day, they simply refuse to recharge. With 50 charge cycles tops, on the pack.

I've also heard that Li-anything technology requires each cell to be monitored and charged individually, with differences of as little as 0.1V per cell leading to imbalances that can cause catastrophic failures.

Bottom line? Great power, high cost, relatively high risks and an expensive charge controller needed. (At least with a marine setup the controller wouldn't have to be thrown out with each battery change.<G>)

So, how does your industrial experience, or LiFePo4-vs-LiOn, jibe with my consumer experience of "lithium" as being expensive and unreliable? Where's reality here?

[electric1]

Originally Posted by s/v Thea

Electric1,

I'm looking at Thundersky's charge/discharge curves in literature for the 200AH battery, looks to me like they show charging at 0.5C until voltage is over 4V, and at 3.6V/cell they show batteries about 60% charged. Also on discharge curve at 90% charged and 0.3C discharge rate they show voltage of about 3.75V.

If this data is correct then looks to me like it would be tough to bulk charge at a fairly high rate up to even 80% using LA chargers?

Doug

Doug, Thundersky datasheets, especially their graphs, are full of baloney. Their charging graph makes no sense at all and their discharge graph is the right shape, but not the right values.

OceanPlanet posted a PDF earlier in this thread which combines charging and discharging curves for 180AH Sky Energy cells. That graph is the true picture. I will also post some graphs I can vouch for here.

This is a true 1C discharge curve for Thundersky cells.



This is the probably the best set of charts I found for a bank of 4 cells, which is precicely what you can expect from 12V nominal LFP house bank.

See how charging percentage climbs linearly, while voltage stays almost leveled until the end? This is what I was trying to explain earlier.

[s/v Jedi]

So, for a 12V bank charge voltage is 14.6V (stop charging at 14.6V) and it's best to stay in the range 30%-90% SOC giving 60% of capacity as usable for about 4,000 cycles (but that is out of the graph range) with max. discharge rate 0.5CA

This means I can replace my 1275Ah LA (425Ah 6V batteries) bank with 600Ah LFP (200Ah cells):

no. of LA terminals to wire: 12
no. of LFP terminals to wire: 24 (double the wiring plus BMS wiring)

weight LA : 275 kg
weight LFP : 87.6 kg -> less than 1/3rd of the weight of LA

max. discharge / charge current for long life LA: 255A
max. discharge / charge current for long life LFP: 300A (more current and much more reserve for surges)

bank recharge time LA: 1.8 hours @ 255A
bank recharge time LFP: 1.8 hours @ 300A (check these, must be wrong)

price LA : $2,220.-
price LFP : $2,640.- plus BMS

cycles LA : who knows, 1,800?
cycles LFP : who knows, 4,000?

looks like LFP comes out on top but a lot depends on the BMS price & features.

cheers,
Nick.

[electric1]

Originally Posted by s/v Jedi

looks like LFP comes out on top but a lot depends on the BMS price & features.

cheers,
Nick.

Nick,

this is a great high level summary, right on the money

This is exactly why I am designing a drop-in replacement with built-in BMS, to take all the guessing out of it, just drop in and enjoy ( minimal control wiring to feedback into charger regulators and control panel, can be done by anyone who knows the business end of voltmeter ).

I am making progress on collecting all the parts for a prototype and drawing out all the features for BMS controller.

[Tinkerer007]

Nick,

Is there a reason you are looking at the 200AH cells?
Do you currently have 3 banks?

There are 400 AH and 800AH cells available. Might reduce the wiring mess.

Your evaluation is compelling for the LiFePo4 technology. Now to see if implementing is practical.

In my garage I found a spotlight to hook up to my battery. It started draining at 6.6 A, but after 20 minutes, it was down to 6.4A. This morning (12 hours later) it was at 6.37 A. I expect that the battery will shut down this evening around 5 PM (23 hours later). I hope I can get home before 5PM so I can witness the event.

Bob

[electric1]

Originally Posted by Tinkerer007

In my garage I found a spotlight to hook up to my battery. It started draining at 6.6 A, but after 20 minutes, it was down to 6.4A. This morning (12 hours later) it was at 6.37 A. I expect that the battery will shut down this evening around 5 PM (23 hours later). I hope I can get home before 5PM so I can witness the event.

Bob

Did you implement LVC cutoff to automatically disconnect the load? Did you test it by manually breaking signaling loop to simulate LVC event? Your 200 AH cells should handle 6.5Amp load for almost 30 hours, but make sure you tested LVC cutoff if you aren't planning to be there when voltage starts dropping.

200AH cells are most practical to implement drop-in replacement, that is why I will be using them in my product, but for DIY job you can go as high as you can afford, although I think having at least 2 strings is better from reliability perspective, if you lose one cell in one string while in the middle of a long trip, you can still run on the remaining string.