Pros & Cons of Lithium-ion batteries for solar application

PNjunction

Re: Pros & Cons of Lithium-ion batteries for solar application

karrak wrote: »

I can sort of see the logic in this, I would be interested in your results

Re: the Optimate TM291 Lithium charger and why I like it so much..

I got bored with cycling two Shorai LFX-14 parallelled together as they really didn't seem to go badly out of balance. Although I am cheating by charging and discharging from opposite polarities of opposing batts - ie I'll use the positive from batt#1, and negative from bat#2. Your links to those 4 year old threads were invaluable and made me grin because it more or less mirrors my experience so far.

So - another test was performed by fully charging the pair, measuring the overall balance for equality, disconnecting them, and discharging only one of them down to about 40% DOD. I reconnected both and let them sit for about 2 hours. I did this because I was in a safe environment, with a bucket of sand, buddy, and eyewash handy. WOW - sure enough major imbalance among the cells - although this was at the midpoint of the flat part of the curve, so these weren't taken as being entirely valid but still ...

I think there was some self-balancing going on.
Bat 1 discharged to 13.3v
Bat 2 discharged to 13.0v

After two-hour reconnect, they stabilized at 13.22v and 13.24 respectively.

Bat 1
Cell 1: 3.296v
Cell 2: 3.320v
Cell 3: 3.317v
Cell 4: 3.301v

Batt 2
Cell 1: 3.301v
Cell 2: 3.306v
Cell 3: 3.299v
Cell 4: 3.304v

While I can't be certain unless I break open the Shorai's, I *think* that in this parallel configuration I really have a 4S4P. Maybe.

The Optimate charged the unbalanced system back up, and only went through it's normal optimize cycle of cycling from 14.0 to 14.4v for about 10 minutes before stopping and passing the short-term voltage retention test. If it wants to do it again it will, but it didn't. I then put it on a Battery Tender set for 14.6v to get them near the absolute top to see how far apart they are.

I pulled the batts apart again, and only noted a 0.050 max cell gap as reported by the Hyperion EOS pack sentry. Did this on both batts and 0.050 was the worst. I was lazy and didn't break out the Fluke for this test.

Pretty happy with that. If I was looking to DIY my own simple controller, I might think about something along these lines. I might even be tempted to use this on a 12v 4S pack made of CALB cells - but then again this was only a one-off experiment so ymmv.

karrak

Re: Pros & Cons of Lithium-ion batteries for solar application

dak664 wrote: »

My take is not to balance, but take corrective action only when a cell drops below some low voltage alarm level.

For Electric Vehicles you have control of the charging, the variable is how much power you will need till you can recharge the battery. With off-grid applications you control the amount of power you draw from the system but don't control when the sun shines/wind blows. This implies to me that for Electric Vehicle use you want to know as accurately as possible how far you can travel until your battery is flat, while for off-grid use you are more interested in maximising the energy you can store for that rainy day or more correctly that cloudy week while also maximising the life of the battery. To me this implies bottom balancing of the battery for EV use and top balancing for off-grid use.

But when one cell hits the low voltage alarm its time to bring it up again, preferably with a single cell charger, next best by shunting all the other cells during the entire charge cycle, next best by reducing charge current and shunting high cells when they reach 100%, and absolute worst by cycling a high current charger on and off, momentarily overwhelming the shunts then allowing them to bleed off excess voltage. Unfortunately most BMSs use the last method. A compromise might be to shunt at 90%, but only when the longer life is more important to the end user than a reduced capacity. And of course shunting reduces the overall coulombic efficiency.

Due to the low charge current compared with EVs the balancing current for off-grid systems can be quite small. I have used these http://www.tmenet.com/products/lithium-charger/xtrema-balancer on a system i made for a friend. One big plus with these is that they do not balance to an absolute voltage like 3.6 volts but balance the highest cell to the lowest cell, so if the cells are balanced they don't do anything. I turn them on when the cell voltages are above an average of 3.375 volts at a charge rate of < .1C which is around 80% SOC.

So oversizing a lithium ion battery is an advantage, for fixed storage anyway. Used 30kWh automotive batteries would be a good match for a nominal 10kWh storage cycle, with excess PV not being wasted on balancing.

Yup, I think the old four times daily use is still not a bad rule of thumb, we have ~9.36kWh of storage for a base load of ~2.5kWh/day without a backup generator. Works well in our climatic conditions with our usage patterns.

PNjunction

Re: Pros & Cons of Lithium-ion batteries for solar application

karrak wrote: »

This implies to me that for Electric Vehicle use you want to know as accurately as possible how far you can travel until your battery is flat, while for off-grid use you are more interested in maximising the energy you can store for that rainy day or more correctly that cloudy week while also maximising the life of the battery. To me this implies bottom balancing of the battery for EV use and top balancing for off-grid use.

Agreed. However I'm going to approach it from a KISS principle (10-80% DOD) and merely de-rate the battery by 30%. That is, if my application calls for a 100ah battery bank, I'll want at least 130ah. I'll go out on a limb and say to just increase your size by 50%. This does not take into account any autonomy needs.

The problem is that if you don't do your power-budget homework right, the upfront costs can make for a very costly mistake. The numerous threads here on how to do this are invaluable.

I think the overall tendency at least for me, is to want to over-complicate a low-current (relatively) application. It really comes down to this:

* Set your upper voltage limit correctly.
* Buy quality stuff, and size appropriately to your application.
* Turn off your gear in time. This can consist of a full-blown monitoring setup, or a watch and a multimeter.

I should have joined the LiFepo4 party years ago. Unfortunately, it took away about 90% of the fun I had doing maintenance from my Pb batteries. I'm still freaked out by not having to worry about sulfation, end-amps, too-little or too-much EQ, etc etc. My relationship to my batteries has changed. I just wish Enersys / Deka or others would jump into the game for the *common man* who wants to support his local community with over the counter sales. Another day perhaps.

Once my budget recovers from this uber-expensive proof-of-concept misapplication with the Shorai's, I'm going big.

PNjunction

Re: Pros & Cons of Lithium-ion batteries for solar application

How about Real World Solar Issues with LiFepo4?

What about remote or intermittent applications like an off-grid weekend cabin, or areas that have very poor solar-insolation, or a charging system that fails at a poor time? LiFepo4's characteristics are ideal for handling this.

My usual recommendation for low solar insolation is AGM. High-quality ones that is that have extremely low internal resistance to make the most of a limited charge window. Of course low self-discharge is always a plus. But STILL, you don't want to leave these things at low states of discharge, or tickle them with less than the recommended minimum current. Wall-warts on an Odyssey actually hurts them long term. The reason I like the pure-lead AGM's is that I can hit them very hard when I only have a very limited insolation window in which to do it. Think Enersys on the high end, and Hawker Cyclon's on the low capacity end. I love 'em.

However, with LiFepo4, you don't want to have them fully charged and then sitting around doing nothing for long periods of time. If you aren't going to use them for a few days, it is best to store them at less than full charge.

What I'm getting at is that for the weekend cabin, your biggest problem will be scrubbing off capacity before you leave, rather than ensuring that you are fully charged. If it then becomes rainy/cloudy for 2 months or so, or you have no gas for the generator, no worries. And if solar is not your main source of energy, but is a backup system, you could also get by with a MUCH smaller solar setup whose sole objective is to just make sure that the batteries are just above 50% DOD. If it takes a few days to do so, then so be it. Now you can return when it is convenient, and not freak out trying to get somebody to drive you to the mountaintop in the middle of winter in a snowcat. Relax knowing that your LiFepo4's are doing the same.

If you look at LiFepo4 charge / storage characteristics, it may be the answer to many solar applications that are not daily or convenient for on-site maintenance, especially if you are in areas or seasons that have very little solar insolation.

PNjunction

Re: Pros & Cons of Lithium-ion batteries for solar application

Intermittent use presents a unique challenge

Using the example of an off-grid weekend getaway cabin, I just realized there could be a major problem to be dealt with.

Proper storage of Lifepo4 means that you don't store them fully charged if you intend to leave them alone for more than a few days. So what if you visit your cabin on weekends, and properly store the batteries during the week?

It means you are met upon arrival with a discharged (not fully, but lets say 50% DOD for discussion) battery bank. In order to use it to support your loads, unless you are willing to wait a whole day or so to get recharged via solar, you are probably going to want to fire up the genset - and obviously so if you arrive at night. Or, you could chance it, and take your bank dangerously low in SOC before the next day when hopefully you actually have some solar insolation.

Interesting challenge - and something I'd have to take into account, which ultimately could rule out Lifepo4 in favor of Pb and all the minutae associated with it's upkeep. Daily use - no problem. So despite my pleasure with lifepo4, it is NOT a universal solution and needs to be carefully matched to your application; not just from a technical standpoint, but from an operational standpoint too.

OR, just double your calculated capacity if you could afford it. At first I thought this was stupid financially, but by doing this it gives you the capacity you need on arrival AND affords you some autonomy. On the back-end just before leaving, with a large bank, you now may have to incorporate a way to use up 50% of the bank before departure, such as disconnecting any charging sources a day or so beforehand, or use a dummy load. Argh.

Ok, too many variables going around in my head. The first one being I need to buy a cabin to test this with.

Sun Dog

Re: Pros & Cons of Lithium-ion batteries for solar application

I don't know what time frame you are imagining between visits to "your cabin", but with the low discharge rate of Li batteries I don't think you would be met with a bank that had discharged that much.

PNjunction

Re: Pros & Cons of Lithium-ion batteries for solar application

Well, it's not a matter of self-discharge, but purposeful discharge prior to going away for more than a few days. You don't want to leave them fully charged for a long period of time to get the most cycles out of them. Of course once you reach that purposeful discharge level, they hold that state for a year or so.

I think my example of 50% DOD is too drastic. Even though primarily a starter application, the Shorai charger has a "store" mode, that will either discharge or charge up to only about 80% SOC which you use if the batteries are not going to be in service. Sooo, in my mind I think that if I stick to the KISS principle for maintaining balance, instead of doubling my capacity for emergency / infrequent use, I'll pare that back to maybe increasing my capacity by 3/4.

Unlike the LiCoo2 cobalt types that usually arrive at 40% SOC from the manufacturer, these Lifepo4's arrived at about 80% SOC, so I guess that makes sense.

That would provide me the capacity I need upon arrival if I don't go overboard... hmm.. It is strange to be dealing with issues of too MUCH power and not too little!

georges99

Lithium ion batteries are superior in all technical aspects. The prices are still high but this is balanced with the efficiency of the battery. Recently  I built in Greece an off greed power device with a 200 W solar panel and a 60 Ah Lithium battery. It works perfectly compared with the neighbors who are still  struggling with their lead acid. Don't hesitate. Go over to Lithium.

mike95490

inetdog said:

....
In principle, I think that you could get a situation in which the CC is sending current full blast into a pack with all of it going through the bypasses and heating things up.

Nope, the BMS will bypass a couple amps and the rest goes into the full cells !

And the common failure mode for the FET's in the BMS, is shorted, which bleeds the cell the vampire board is mounted on, down to 0V.  In theory, it works, in practice, not ready for prime time yet. 

What does work pretty reliably is manual balance and weekly verification it's not drifting much, till you get enough confidence to go to monthly verification.