Batteries - What The Future Holds.......

[Maine Sail]

So what is the goal here.... just a science project for your personal use, or are you looking to go into some level of production?

Mostly for my own knowledge and experimenting at this point in time. I am waiting on the ABYC standards to catch up before I would build packs for customers. Some guys in my field are already building them for customers, a friend in Florida has two builds going, but I am not willing to do that at this point.

With a good amount of homework, LOTS and LOTS of reading and a good understanding of electrics a DIY could build one for a similar cost to quality AGM's.. For most boaters though flooded lead acid batteries still represent a great value, especially 6V wet cells.

Even smart guys make mistakes though. A knowledgeable guy on another forum mistakenly left the temp compensation enabled on his charger. The temps dropped, the charger compensated the voltage up and the cells were over charged and "bulged" / ruined. A simple mistake that cost him a lot of money. These pack builds need to be carefully thought out, planned and well executed. If they are they are simply amazing batteries..

 

[Witzend]

Yea, obviously at this point LIFePo4 technology would not be suitable for the layman, but as always...production volumes go up, prices will come down, reducing risk. Systems will become bundled and better integrated for less confusion. Your work here is a step in that direction.

 

[Kettlewell]

IMHO these aren't foolproof enough for the average cruiser at this point. I have had so many failures over the years of various charging components, that then destroyed batteries, that I can see very expensive problems and then what do you do to replace the batteries and the failed electronics when you are in some atoll in the South Pacific or the Out Islands in the Bahamas? It will be interesting to see how Maine Sail does with these over the course of a few years, keeping in mind that his testing set up will probably be the best-case scenario.

 

[Maine Sail]

Just an update on the battery pack as I have had a lot of questions.

Expenses so far:

4 - 400Ah Winston Cells = $1032.00 (this was a special deal today they are 2k/$500.00 per cell)
Shipping Cells = $250.00 (they come in very well packed wooden crate truck freight)
Tyco EV200 Contactor = $79.00 (this is the load cut relay)
House Power BMS Board = $75.00 (Clean Power Auto LLC.)
Cell Modules = $54.00 (Clean Power Auto LLC.)
Aluminum for end plates/compression = $64.00
3/8" Polycarbonate = $27.00
Anodizing Aluminum = $60.00 (minimum charge)
Nuts, Bolts, Threaded Rod, Washers etc. = $12.80
Bus Bar = $4.80 (Blue Sea)
Switches, Alarm Buzzer etc. $44.00
Terminals = $9.50
High Voltage Cut Relays $24.00
Total Materials (my cost) = $1731.60
Total at current cell pricing $2704.10

Things I already had that you would need:

Balmar Regulator for alternator = $$$
Solar or wind controller with fully adjustable voltage parameters and NO temp sensing = $$
Class T fuse = $
Benchtop DC Power Supply - 30A minimum (for balancing the cells) = $$$
140A alternator with serpentine pulley kit = $$$$$$$$$$$

Labor:

This is an involved process and I am not counting my hours of "testing" that I did. I conducted at least 55-60 cycling tests and re-charges to discern how I wanted to best charge this bank. The bank was run in testing for months before installing it in the boat.

Unpacking and setting cells up for balancing - 2.5 hours
Balancing Cells - 11 hours (did both a top and bottom balance and decided on top)
Building custom aluminum cell compression case - 7.5 hours
Copper bus bar + & - take off jumpers - 1 hour
Wiring & building BMS box - 6.5 hours
Wiring cell modules and volt sense harness - 2.5 hours
System design/design changes - 6 hours (not including research)
Building new battery mount shelf in vessel - 11.25 hours (complete new battery location)
Running 2/0 wire and volt sense wire to battery distribution area - 4.5 hours
Re-wiring battery distribution for charge bus, load bus, BMS, regulator etc. - 6.5 hours
Wiring & programming MC-614 alternator regulator - 1.75 hours
Wiring / programming new solar controller - 1 hour
Wiring battery charger cut off relay - .5 hours
Installing new reserve bank (AGM) - 1 hour
Removing temp sensor for battery charger - .25 hours
Re-programming battery monitor - .1 hour

Total Labor Approx 65-70 hours (I also did about 16-18 months of research before doing this)

Shipping:

Initial Cell Balancing:

Capacity Testing:

BMS Construction:

Case Construction:

Case Completed:

Cell Module Wiring:

Battery Compartment Construction:

Battery Hold Down Design:

Distribution Layout:

Installing The Distribution & BMS Wiring:

Battery Installation:

 

[Bazzer]

Hi, just for interest a sailboat at our marina has lithium cells installed, I don't know who did the installation, but here are a couple of points I have found out from the skipper.
It appears to have two cells fitted which I have seen.
The motor, a yanmer turbo, has two alternators for charging, one for each each battery apparently,
When charging starts, the motor has to be run at 2000 rpm otherwise it loads up and stalls.
It was recently used sailing from California to the tip of Baja and back.
It reaches nearly full charge in about one hour.
There is a third alternator, the original Yanmar alt, that is used to charge the starting battery
The skipper is very pleased with the system

 

[Chris Patterson]

Wow Maine. Pretty work. Makes me want to print the pictures, and frame them.

 

[Sun Coast]

That is some interesting developments.

Chet

 

[Bob S]

Maine,I can’t believe how beautiful your work is. I do have a question as to how you mounted this into the junction box. Did you have to drill holes through the base? I’m mounting some terminal blocks into 2 junction boxes. I did one by drilling 2 holes into a piece of Lexan and mounted the terminal strip to that. I used some butyl tape and saw how it depresses to seal the back. What did you use for a backing plate on this and how did you mount it to the box? Thanks for putting up the small terminal connections on your Compass Marine site!

 

[Gunni]

Does it make any sense to start building toward a future Li bat install. eg: upgrade the alternator to a high output externally controlled system, install a very powerful AC power charger that can make use of my 5.5 kW AC generator? Then wait for the future to unfold? Or do you anticipate that changes in the battery technology will obsolete existing charge hardware?

 

[Maine Sail]

Does it make any sense to start building toward a future Li bat install. eg: upgrade the alternator to a high output externally controlled system, install a very powerful AC power charger that can make use of my 5.5 kW AC generator? Then wait for the future to unfold? Or do you anticipate that changes in the battery technology will obsolete existing charge hardware?

Yes it can pay and gets you there incrementally. I would focus on things like the alternator and a serpentine kit to drive it. The Balmar AT series (hairpin wound) are probably going to be the future of alternators for charging Li banks. You could also wire a charge and load bus ahead of time.

When you buy a charger you want one that allows for custom profiles to be created and one that allows for the complete turning off of any temp compensation either internally to the charger (ambient sensing) or an external battery sensor.

 

[Maine Sail]

Some observations thus far.

This weekend we went out and I decided to run all electric for our refrigeration and chose not to use the engine driven side of the holding plate. Our electric side draws about 10A on high and on low about 5A. I wanted to run the bank down to 80% DOD and see how she performed on the voltage side and cranking side as well. The only augmentation we used was our 140W solar panel but only on day three when we were getting close to 80% depth of discharge..

We went for multiple days in pure silence with zero engine use. We were barely able to draw the bank to 80% DOD over multiple days by turning off solar and being energy hogs. We purposely used as much power as we could and still the voltage held well above 13V. We left the laptop on and open, charged two iPads, iPhone, Android, used tons of cabin lighting, listened to music, had three cabin fans running 24 hours per day as well as a fan in the ice box to circulate the cool air.

The lowest voltage I noted seeing was 13.17V and that was at a 17.4A load at 41% SOC...

When I fired up the motor it was like the battery was fully charged and the 80% depth of discharge had zero perceptible effect. All instruments and fridge and other items were left on during starting and nothing dropped out. The voltage barely budged during starting. (confirmed with Fluke/in-rush/hold feature)

When I fired the motor at 1200 RPM the alt was pumping out 97-100A while we were getting ready to leave the anchorage. When I put her in gear to leave and hit cruise RPM she bumped up to about 144A +/- then slowly dropped back to about 120A (where she runs when hot). After about 25-30 minutes the regulator went into alt temp limiting and cut the output back to 60A. The alt would cool and then pop right back to 120A. It continued this temp limiting dance for the remainder. I will again tweak the reg to limit the alt output so she can run continuously at the highest current possible without entering temp limiting..


Unfortunately we motored for roughly 3 hours today, due to zero wind. The best part was that by the time we got back to our mooring today the battery bank was back at 97% SOC!!!! We never ever would have seen that with our lead acid batteries, because it is simply not possible.

One other observation I saw was when at 90% SOC, and it went back into temp limiting, and the bank was only getting 60A of current, that 60A could barely get the bank to 13.5V and the 120A would bring it to about 13.7V. It takes a LOT of current to raise the voltage on these batteries.

Just a few more tweaks to the regulator and it will be dialed in. I may go with a 200A alt and run it at a max of about 150A +/- so it can run cool all day and not enter temp limiting but it will give me a full recharge from 80% DOD to 100% full in just two hours.....

So far the toughest things to get over are the natural instinct to re-charge the batteries when they get towards 50% SOC. Letting them run to 80% DOD was tough and takes some personal reprogramming....

So far this battery has been simply amazing technology...

 

[Witzend]

Some observations thus far.

It takes a LOT of current to raise the voltage on these batteries.

The same goes the other way which is why you get such great performance.

 

[Maine Sail]

The same goes the other way which is why you get such great performance.

Totally true. For the average loads we see on sailboats it barely budges the voltage of these batteries through the entire depth of discharge range.. All our systems are MUCH happier. The Espar loves it, the fridge compressor loves it, it even performs better. On low our fridge compressor used to barely keep up now it keeps the plate at about 3F and cycles on and off. With the LA batteries we needed the medium setting to achieve the same cooling performance. Bilge pumps and water pumps work better even the stereo plays louder and cleaner before distorting..

All and all I have had to do very little tweaking since the bank went in. It has been performing flawlessly. I did mess around with charging voltages, now at 13.8V, and I will install a regulator cut switch in my engine panel but nothing else has needed to be done. The cells are still within .002V - .004V (Fluke 179) of each other so balance has been a non issue....

 

[Maine Sail]

Just wanted to update this thread for those interested in this new technology.

I am finally closing in on 200 cycles on our LiFePO4 battery bank. I have not pushed charging voltage to cell balancing levels and only performed a single top balance on the cells, 192 cycle ago or last winter. We do have a BMS system but I treat it as an insurance policy and do not use it. At high charging voltages the BMS can shunt small amounts of current to other cells in order to keep them in-balance. For EV guys who need every last ounce of energy and push the voltages into the knee range a BMS is a good tool to have. For house bank use, which more close mirrors off-grid, I have decided not to push my charge voltages into the "upper knee range"...

Since July the charge voltage has been limited to 13.8V which puts back in nearly 98-99% of the bank capacity but has far less risk of over charging and damaging the cells. All of our charge sources are limited to 13.8V.

During the winter I played with numerous charge voltages and settled on 13.8V to 14.0V and in July decided 13.8V was the perfect balance point on our boat. 14.0V gives a little shorter acceptance taper but at 13.8V the taper to full from bulk is still about 30 minutes.

In July I also had to current limit our 160A alt to 115-120A. This allows the alt to run there indefinitely without exceeding 225F. The temp limiting features of most external regulators reduce the current too deeply for my tastes and the alt goes from full bore to half speed and the over all charging is slower. I tried playing around with temp settings but found Belt Manager on the MC-614 worked better. If going Li you will need an alt significantly larger than where you actually want your current or it will get VERY, VERY hot and cook itself. I still have the temp sensor and it is still set for 225 but at the new current limit it has yet to exceed about 220F...

I deem the bank "full" when the current drops to 5A at 13.8V. Pretty simple stuff. I thn manually synch the battery monitor and cycle to 80% DOD again. If this bank was not for experimental purposes I would likely cycle only to 60 or 70% DOD... At 5A - 13.8V I shut off solar and the alternator and do not turn them back on until the bank drops to 80% DOD.

Here's where the cell balance just tested:

*Resting
*Loaded
*Bulk Charging
*Absorption Charging (Voltage Limited to 13.8V)

I am not real happy with my Cell Log 8 for accurate data and the Fluke 179 is the meter I have been basing everything off since the beginning. Thus I am using my NIST calibrated Fluke for all measurements to keep everything the same. The Cell Log 8 is a good tool, and lots of Li guys use them, but not what I want to use for accurate measurements.

Here's what I just measured:

Unloaded resting @ 74% SOC - Cell temps 49F:

Cell #1 = 3.362V
Cell #2 = 3.361V
Cell #3 = 3.361V
Cell #4 = 3.361V

Loaded @ a -17.4A average load @ 72% SOC - Cell temps 49F

Cell #1 = 3.313V
Cell #2 = 3.314V
Cell #3 = 3.314V
Cell #4 = 3.313V


Charging @ 121A 13.72V - Cell temps 49F

Cell #1 = 3.431V
Cell #2 = 3.430V
Cell #3 = 3.429V
Cell #4 = 3.430V

Charging @ 36A - 38A (current dropping quickly) 13.80V (regulator voltage limiting) - Cell temps 49F

Cell #1 = 3.452V
Cell #2 = 3.452V
Cell #3 = 3.453V
Cell #4 = 3.452V

Not half bad for 192 cycles, approx 70% of them to 80% DOD, and no cell balancing since the initial top balance 192 cycles ago.

It has NOT been easy to get to 192 cycles, lots of tally-marks in my note book. Almost half of them were done in my shop at .25 "C" or 100A loads and the rest were done on the boat often with the help of a ceramic disc heater to accelerate testing and DOD. I did a lot of invoicing on-board this summer rather than in my home office as it allowed me to turn on a 100A load while doing it and get some more cycles in.

Essentially we've simply been turning off all charge sources and letting the bank draw down to 80% DOD before even considering to re-charge. Even at 80% DOD with our average loads we don't seem to drop below 13V as a bank voltage or roughly 3.25V per cell. At average house loads of less than 10A the voltage barely moves at all. In fact right now I am sitting here drawing nearly 14A running the Espar, inverter, stereo and a bunch of electronics. The loaded bank voltage is still 13.26V. I find this pretty darn good for cells that are currently at 49F...

One observation I have noted over the years is that with lead acid batteries and coastal cruising boats the actual cycles to 50% DOD are pretty minimal before the banks are dead and replaced. When I poll my own customers on how many overnights and trips they take per year it gives me a good understanding of cycle life. I think it is pretty safe to say that the average coastal cruiser barely sees 130-150 cycles, over 4-5 years, before the banks are dead. Heck many customers I pick up do as little as 15-20 cycles per year. I often find a 1 week cruise and a hand full of Fri/Sat weekends. In less than one year I have already completed 192 cycles on this technology and most to 80% DOD not the 50% of lead acid... Just some more musings...

 

[Maine Sail]

I know it has been a while but I just wanted to post an update on this new technology...

First, this bank has still yet to be "balanced" beyond my first top balancing 550 cycles ago. My BMS system is set up to NOT automatically balance the cells. Considering I have yet to see many lead acid banks go 550 cycle on a boat, and none to 80% DOD and 550 cycles, this is proving to be less of an issue than you would have in equalizing lead acid batteries.

Of course a big reason for this steady balance is because I have employed safe & sane charging voltages, I do not float, and I do not charge the batteries at temps below 32F.

Just yesterday completed my 12th full capacity test, on this 400Ah rated bank, and this was at cycle number 550. This test was done at a 30A constant load the bank delivered 419.2 Ah's of capacity, this after 550 cycles with most of them being to 80% depth of discharge.


Pictured here is the 400Ah bank, a 60A lab grade power supply for charging and the 40A lab grade DC electronic load tester. Most of the smaller wires, in this spaghetti mess, are voltage sensing leads...

One of the things I offer my customers is 20 hour capacity testing for their expensive AGM or GEL batteries. Some even take advantage of it for flooded deep cycle. The only true way to test a lead acid battery, for use as a deep cycle bank, is to physically capacity test it using industry standard testing procedures. The 20 hour test is most representative of the loads used on boats and thus a true 20 hour capacity test is what I use.


Many years ago I built an Ah capacity tester using an Ah counter, relays and DC loads but the accuracy was not as good as I would like because the DC load needs to stay steady the entire duration and this means manual manipulation. As the voltage decays the current increases thus changing the batteries discharge rate. With an non steady discharge rate, on lead acid batteries, this results in less than stellar accuracy. With LiFePO4 holding the current steady, while capacity testing the bank, is less critical because of the low Peukert effect.


A few years ago I had Mark Grasser, of Mark Grasser DC Solutions, build me a custom DC constant load tester for Ah capacity testing batteries at the 20 hour rate. This device worked well but still I wanted better control, easier set up and better overall accuracy.


I finally invested in a lab grade DC electronic load with battery testing capability. What a world of difference! The accuracy of these devices is amazing but they are not inexpensive.


What does all this have to do with LiFePO4?

At capacity test #12 (cycle #550) I switched this battery bank over to the new DC electronic load tester. All testing from this point forward will be done using this device. My new baseline was established at cycle #550 and the bank delivered an astounding 419.2 Ah's at a 30A constant load after 550 cycles!


I hesitate to directly compare this capacity figure to the previous 11 capacity tests but it was very close to them and the bank delivered 419.2Ah's. From any 400Ah rated bank with 550 cycles on it, most to 80% DOD, this is something I would have never believed, had I not done the testing myself.


This bank is perhaps proving my initial cycle life skepticism as being wrong or misguided. While the discharge rate was lower, using the DC electronic load, the Peukert effect on this bank is also low. What really matters, or should, to myself or anyone choosing to use these as a house bank, is how many Ah's can you get at your average house loads? The answer for this bank is in excess of 400Ah's even at 550 cycles into it...


I do believe I have lost some capacity in those 550 cycles, but not much. The last test, using the old testing rig, delivered 423 Ah's.. Tests using the old method & equipment yielded anywhere from 421Ah's to a high of 426Ah's. This test was at 30A, not 100A, but with a higher accuracy piece of equipment and no inverter, just a pure DC load. Still this bank delivers more Ah capacity, at a 30A load, which is multiples more than our average on-board load, and that is good! We have 419 Ah's of capacity, at a 30A constant load, after 550 cycles. I'm a happy camper!!



Here is a close up shot of the data the DC electronic load captures when set to battery Ah testing mode. I set the capacity tester to 30A so as not to over heat it by running at full bore for 10+/- hours. The unit is rated at 40A but like anything electronic it likes to be run at less and will last longer doing so.


It should be noted that both my charge source and DC Electronic Load tester both utilize dedicated voltage sensing leads that do not carry any current. This means accurate charging voltages physically measured at the battery terminals, and also accurate cut-off voltages, measured at the battery terminals.


The cut off voltage for this capacity test was set to 11.2V or 2.8VPC. A Cell Log 8 was used to trigger alarms if any cell dropped below 2.78V.

For comparisons sake here is a 100Ah AGM battery that is less than 3 years old and has under 100 cycles on it.. The owner claims to have never let the voltage dip below 12.1V...

Tester Screen Shot

Needless to say I am pretty happy with the results of my LiFePO4 experiment thus far. I still don't think it is ready for mass "prime time" DIY builds though..

For more in-depth information and back ground on LiFePO4, I have opened up my yet to be completed article on LiFePO4 at the link below..

LiFePO4 Thoughts & Musings

 

[Chief RA]

Long but interesting read. I can see that I need to study this battery concept further.
Thanks, Chief

 

[Maine Sail]

Long but interesting read. I can see that I need to study this battery concept further.
Thanks, Chief

While we are surrounded every day by Li batteries when you get to large prismatic cells, that cost large sums of money, you want to protect your investment. Learning as much as you can is a good idea..

Unfortunately there has been very little research into the use of large prismatic cells at fractional "C" charge and discharge rates. I have 30-40 white papers on my hard drive and none of them deal with what I call "fractional "C" use". Lots of good info in them just not always applicable to use as a house bank..

Those of us using them off grid, and for house banks on boats, are really learning how well these batteries do when not used in an EV or other high draw application.

New technology is always exciting..

 

[Chief RA]

Yes, new technology is exciting and drove me for many years. At my current age I find my curiosity level still pretty high, it isn't what it once was. Of course I study less but my background is rather extensive for an ol' Tech/Engineer. I spent many years as a military tech and maint. officer before attaining engineering degrees in Industrial and Electronic fields.
Chief

 

[monkeybars]

That's about all it is, a curiosity and entertainment. Li has been around over a decade as far as availability for auto-marine battery power and has yet to even capture a percentage of a percent of the market share. AGM and Gel are just barely holding on to a large enough share of the market to where they are actually available in retail stores full time without having to make a special order. And most mechanics-repair facility's still aren't fully versed on their different charging requirements. It will easily be several decades before Li and lastest hi=tech toys will be a common use item.

 

[Chief RA]

monkeybars: I started to comment but admit I am not that versed on Lithium technology so will leave you for MaineSail. (kinda sounds like you are not either) I am an Electronic Engineer, not Electrical. Chief