Testing a Shorai LiFePO4 on solar!

[Sunking]

IE will 100 mv difference between cells in a lithium phosphate pack cause a life threatening situation.? I doubt it.

John there are two points I think you are over looking.

1. Lithium batteries are extremely expensive. Some 4 to 10 times more expensive per watt hour than their lead acid cousins. That alone justifies a balanced charger.
2. Initially you are right a 100 mv is not a lot to worry about. But liithium batteries are extremely sensitive to over charging. The weakest battery in a series string will have the highest voltage applied during recharge. That results in an over charge condition of that cell and under charge of the normal cells. With each cycle the damage is compounded and accumulative causing premature failure. Now go back and look at reason #1.

[Sunny Solar]

After putting into service 4 systems using 12x 10ahr cells per system in constant use over 4 yrs and another now over 3 yrs.As I said in a previous post they don't get very far out of balance and easily kept in check by simple means.. Its not necessary to spend a fortune on doing this. Same goes for charging a PWM charger than can have the charge voltage altered and equalizing turned off works fine.
Having tested hundreds of various sized cells at work have found they are NOT that sensitive to overcharging. We have long term tested many to 14.5v and never had one die from that.. Low voltage is another story. 2.3 v is rock bottom. and its safer have disconnect at 2.5v.. I have destroyed cells from 10ahr to 400 ahr with just one discharge to 2v.and some even at 2.2v

In real life you shouldn't get to that low a figure as 4 cells at 2.5v =10v . Any half reasonable inverter will shut down before 10.5v.. But there is no way I would trust that only, a LV cut off switch is really mandatory.

[Sunking]

Cant' say I know of any practical application in a solar application that only uses a 12 volt 10 AH battery John. A good 4S 10 AH LiPo pack cost around $300 + a $100 state of the art balance charger. a Good 12 volt 10 AH AGM battery cost $20 and uses a $50 state of the art charger. If I am going to pay for LiPo, you can darn sure bet I am going to pay a small price for a good balance charger. That same Balance Charger I am speaking of can charge any battery chemistry at optimum charge rates and algorithms up to 400 watt input. That means I could charge a 12 volt 30 AH LiPo in 1 hour or a 12 volt 120 AH LiPo in 4 hours. You can darn sure bet I am willing to pay a measly $100 for a good balance charger for a $1000 battery investment. You can use your $50 charger all you want.

It is economics 101.

[PNjunction]

Solar on my two Shorai's is working well.

I have no need to actually charge up to 14.4v, so I put back the jumper in my little Morningstar pwm CC's which set at 14.1v. Batts are still fully charged by the time I get to them. What a joy it is not having to deal with long absorb times and watch the lifepo's just gobble current until the very last minute. Nice since in my location in winter, I don't have much solar insolation due to long shadows from neighborhood trees, etc.

Basically, I love how fast and how hard you can hit them - much like my beloved Odyssey tppl agm's. However, unlike the Odysseys, I have a bit more latitude with low-current charging, although I want to hit them with about 0.5 to 1C. Apparently this is good for the best reformation during charge, although lower currents are not supposed to damage cycle life, unlike an Odyssey which basically demands 0.4C to keep them healthy.

I've been searching, and see references to charging LiFEpo4's between 0.5-1C for best life, but don't have any hard data on what <0.5C does to them. Being an Odyssey agm fan, I'm obsessive over supplying the proper amount of current. If anyone knows if <0.5C is harmful overall to lifepo4's, I'd appreciate an update.

After an initial balancing with the Shorai charger, I have been through about 10 cycles with charging from my carefully selected standard agm charger limited to 14.4v, and subsequent recharges with the balancer don't seem to take any longer than it did the first time out. Still waiting for my more advanced charger and Hyperios EOS voltage / balance checker to arrive...

And yes, my low-current application is at least 3 times the cost of an Odyssey agm. (Based on ACTUAL ah/wh, NOT that funky Pb/Eq value that is marketed towards SLI and CCA's) Knew that going in, BUT if I take care of these Shorais, maybe I can eek out twice or more the cycles out of them in the long run making them a bit less of a hit on the wallet.

[Sunking]

I've been searching, and see references to charging LiFEpo4's between 0.5-1C for best life, but don't have any hard data on what <0.5C does to them.

Completely dependent on a manufacture to manufacture, and model to model. Generically you can say do not exceed 1C charge rate. However many recommend much higher rates up to 5C for a maximum 15 minute recharge.

The driving force today and for the foreseeable future is the Electric Vehicle market. Tons of money both public and private is being poured into Lithium Battery development. Two of the most critical characteristics of a lithium battery in a EV is very high charge/discharge rates. They will have to have that in order to be able to compete with ICE cars. The market (mass public) demands a 15 minute or less recharge time similar to filling up with your car with gas today. Without that moderate long distance trips are not possible. Right now technology exist to get 5C charge rates and up to 50C discharge rates. However it is very expensive and cycle life is not acceptable at 500 or less cycles to 80% rated capacity.

[Sunny Solar]

Cant' say I know of any practical application in a solar application that only uses a 12 volt 10 AH battery JohnI cant think of one either that's why I made the systems I built 30ahr.
After looking at the build dates its now over5 yrs since the first solar lighting ones assembled.. Nothing has been replaced or needed and maintenance. Its exactly as how the customers wanted it.

[PNjunction]

Completely dependent on a manufacture to manufacture, and model to model. Generically you can say do not exceed 1C charge rate. However many recommend much higher rates up to 5C for a maximum 15 minute recharge.

The Shorai's seem to indicate about a 3C max on their label. I have no need for that fast a charge, and will limit myself to probably 2C max just to be conservative. Since they have such low internal resistance, my latitude is probably going to be 0.1C to 2C. This is unlike my fav Odyssey agm's, which need 0.4C minimum to be healthy. I like having that wider range of charge options without any chemical degradation from a low current input.

My other toys came in yesterday. The Optimate TM291 Lithium charger is functioning just fine when monitored with my Fluke, and both Shorai's which were subject to many previous standard terminal charges passed with flying colors. The Hyperion EOS Pack Sentry unfortunately looks like it needs a Shorai-specific cable, and goes into *automatic* top balancing. Not really liking that.

What became immediately apparent when I laid out all these toys is that there is only ONE thing I can count on - my Fluke 87V. Epiphany! Because of this, I'm not going to depend on anything but my own bottom-balancing with a break-out cable as opposed to top-balancing from devices that I have no idea of their accuracy or ability to maintain it over time or differing temperatures. There are a LOT of disputes over this in other forums so I'll just say that I'm not going to have my batteries depending on top balancing from unknown and possibly uncalibrated little devices. Maybe if Fluke got into the bms market, I might reconsider.

In reality, I'm just going to depend on the Optimate TM291 which uses a different method of balancing than top or bottom, and instead relies on pulsing the last 10% of charge compared to internal resistance templates for lifepo4, among a bunch of other useful tests. Seems reasonable enough, especially since the Shorai is only a 4S configuration no matter the capacity model chosen. And this will not be used each and every time, since solar is part of my equation for buying them in the first place.

Well, that's the plan anyway. Might take some time to get back with figures unless I encounter early failure.

The best part from this experiment is that for me, it looks like If I go with larger cells, I'll be doing my own 4S configuration, with no BMS, and instead rely on quality cells that may have simply been bottom-balanced at the start, and from then on just quality monitoring. Looks like I'll need three more Flukes.

[Sunny Solar]

PNJUNCTION From all the testing and experimenting we have done at work over the last 6 yrs or so . Generally its a lot harder to balance large cells. Over 200ahr are really a problem..They just seem to drift regardless of who manufactured them..
10ahr is the "perfect" size we found..and easy to replace if one goes faulty..(rare).

Its most likely why car manufacturers use thousands of very small cells. Big cells just are to much trouble??

[Sunking]

The Hyperion EOS Pack Sentry unfortunately looks like it needs a Shorai-specific cable, and goes into *automatic* top balancing. Not really liking that.

I use Hyperion batteries in my RC planes and know exactly what you are talking about. Depending on your charger it can be fixed real easy. For example my 3S Hyperion batteries come with a 4S balance plug (5 pins). All I have to do is shift the wires over one pin. That fixes the problem. Take a close look at the balance plug, and I bet you will notice it will have one extra pin and all the wires are shifted by 1-pin.

[PNjunction]

THANK YOU! Being one pin off was indeed the case, although transitioning from a 5-pin balance port to a 7-pin charge port cable had me scratching my head. Had to make a minor case mod to the EOS sentry and it works great. BUT LURKERS - this VOIDS YOUR WARRANTEE as nothing but the dedicated charger is allowed for attachment to the Shorai port. If you don't know what you are doing, or cannot ENSURE that you are doing this safely, then STOP.

Now I just have to hope that the dc voltmeters in the Hyperion EOS are somewhat accurate. Nice way to collect data, but I probably would have preferred the earlier version which didn't go into auto-balance. At any rate, good to see it working, and again thanks, otherwise it was going to be an expensive little dust catcher.

[Sunking]

THANK YOU! Being one pin off was indeed the case, although transitioning from a 5-pin balance port to a 7-pin charge port cable had me scratching my head. Had to make a minor case mod to the EOS sentry and it works great. BUT LURKERS - this VOIDS YOUR WARRANTEE as nothing but the dedicated charger is allowed for attachment to the Shorai port. If you don't know what you are doing, or cannot ENSURE that you are doing this safely, then STOP.

Now I just have to hope that the dc voltmeters in the Hyperion EOS are somewhat accurate. Nice way to collect data, but I probably would have preferred the earlier version which didn't go into auto-balance. At any rate, good to see it working, and again thanks, otherwise it was going to be an expensive little dust catcher.

You are welcome. Hyperion does that for sales so you have to either buy an adapter cable or charger from them. I have an RC LiPo charger. That charger has a external board on it to enable any Balance Port Plug to be used. My Hyperion are 3S aka 3 cells, however the Hyperion 3S batteries come with a 4S plug. All you have to do is shift th3 wires over 1-pin.

[PNjunction]

The OES pack sentry seems to be about .01 to maybe .02V different from my Fluke 87V. Maybe that is good enough for real-world balancing / scrubbing.

I'm definitely going to try my hand at bottom-balance, and for convenience I'll use the Hyperion there too once I draw down to the discharge knee. I've observed that rest periods are probably the most beneficial to scrubbing, as doing this right off charge is a long dance. AND after many hours of rest, it seems to need a balance yet again! Makes me wonder about the efficacy of top-balancing while under charge - even if taking the accuracy of the scrubber into account.

Since I'm not intending to live on the scrubber, but use it occasionally for drastic capacity checks, I'm going to take the Shorai into the basement of the discharge knee, and the Optimate will recognize that and charge gently until it is out of the knee before applying full charge.

I guess I'm getting a bit too RC here, but what I want to do is find out what is more effective for a standard two-terminal charge from say under the typical solar charge controller. More to come...

[Sunking]

PN don't sweat .05 volts or split hairs with voltage. Most important parameter is INTERNAL RESISTANCE. When you have a single cell in a string go 10% higher than the others, you have big trouble.

[PNjunction]

I think I was looking for trouble and not really finding any. Perhaps because I'm only discharging at a 0.2C rate.

Starting from a fully-charged Shorai, I let the EOS pack sentry top balance / scrub it so that it stops itself at a max cell-gap of 0.005v.

At 80% DOD (roughly 12.883v total / 3.22v per cell), no balance issues reported by the Hyperion EOS - all cells still at 0.005v max gap.

At 95% DOD (roughly 12.5v / 3.125v per cell), the max cell gap was 0.043.

Revived with the Optimate since it drops to low-current recovery below 12.8v so battery doesn't get hammered charging at very low SOC, and it brought it back to full charge eventually and passed all tests.

EOS reports a max cell gap of 0.135v at the top end. Eh, I just let EOS balance it again.

I think that I'm good, but not going to learn much from this other than my gear is working as it should, and the cell quality appears good too. Will stay out of the knees for best life.

There is really no drama here. Simple charging along with common-sense monitoring without obsession seems to be the way to go. I hate to say it, but this was kind of anti-climactic. But I'm glad I went through it.

[Sunking]

PN just like lead acid batteries, especially AGM, for LiPo or any of the lithium batteries is to measure and keep track of the Internal Resistance from the start. Most of the better Smart Chargers made for Lithium will do that for you and I assume yours does?

For example just pulling out the first Hyperion in my box of batteries for my planes is a 4S 3600 ma, 5C/50C pack. When I bought it and after a couple of cycles I took a permanent marker, measured the internal resistance/capacity, and wrote it on the battery with the date. It reads 3760, 8/8/9/9, 8/11/13. So when I start seeing those numbers slip I know my battery is slipping away from me and to get rid of it.

A few last tips I can give you on LiPo is to store them @ 3.8 to 3.85 volts per cell or roughly 60% SOC. Keep them in an air-tight bag like ZipLock and store them in your refrigerator just be sure to warm them up before charging and using. Only charge them to 90% SOC and do not go below 20%. Do those things and you will extend the life of the batteries as much as possible.

One last thought or thoughts. LiPo's do not exhibit the Peukert Effect and are roughly 99% efficient on both charge and discharge. Discharging at high rates assuming you do not exceed the maximum rate will not harm the batteries. As for charge rates do not exceed 1C unless the manufacture specifically states you can do so. Even then keep to 1C or less. Lastly when charging do not leave them unattended and touch them while charging. If they start to feel warm, STOP because something is WRONG. LiPo should NOT get warm while charging because they are about 99% efficient with very low resistance.

Good Luck

SK