LFP battery chatter ( AKA LiFePo4, Lithium Iron Phosphate )

[Living Large]

Wow - great response. The washcloth has been pulled out, and dipped into ice water.

I do admit, I was shaving yesterday, and the thought of my LI-Ion computer battery dying after a few years came to mind. I wondered, "gosh, what happens if my LFP bank dies in year 3? What if these things don't last the 2000 cycles they claim?"

Very good points. Thanks for giving me some perspective.

[Sunking]

FWIW I know of many Calb CA series EV users are on year 3 since they came came out and still going strong.

Keep in mind what is in your razor, cell phone, power tools, and the like are LiCo, not LFP. No commercial pure EV's use LFP. Only Chevy Volt uses a nano LFP made by A123 Systems and has been doing good. Trick is A123 Systems LFP cost roughly 4 times as much.

[reed cundiff]

A fun and informative thread. Thanks to both for driving on and getting things discussed
Reed

[Living Large]

FWIW I know of many Calb CA series EV users are on year 3 since they came came out and still going strong.

Keep in mind what is in your razor, cell phone, power tools, and the like are LiCo, not LFP. No commercial pure EV's use LFP. Only Chevy Volt uses a nano LFP made by A123 Systems and has been doing good. Trick is A123 Systems LFP cost roughly 4 times as much.

Heh heh. I use a blade razor, but the reference is a good one. Yes, I know I don't have an (dead) LFP in my laptop. There is a lot you've opened my eyes to as I look further into this subject. I am still excited about the product, but then again I fell victim to a Ponzi scheme 5 years ago that took 1/10 of my wealth. I was excited about that too. Also GM preferred stock in 2007, which became worthless. Mortgage backed securities as well, which met the same fate. I need to watch my enthusiasm - I'm on a bad roll.

Your friends - did they modify an OEM vehicle? You said OEM doesn't use LFP, right?

[reed cundiff]

We are on year 2 of CALB LFP (Manzanita Micro packaging) and are extremely happy with them in a 34' 5th wheel. System has been solar autonomous in this time frame. LFP is excellent for RV'ing in 5th wheel or travel trailer where weight and cube can be a major considerations. The LFP battery bank saves us around 500 to 600 pounds over lead acid. Frame and pin weight would be over specifications.

Liberty Coaches had tested their LFP systems through 2000 cycles as of one or two years ago. A Liberty Coach owner ($1.8M) wrote me that they have continued testing far beyond 2000 cycles. However, this is just "word of mouth" via RV.Net
Reed and Elaine

[Mike90250]

Ok Mike I get it - get the ComBox! I'll add it to the budget. One question - doesn't the SCP (which I won't be buying ) allow you to set the charge settings? I believe Schneider told me those are synched on the bus between the XW and MPPT. Maybe that is just thresholds, and not modes.

Can you tell me why I would choose the 6048 over the 5548 (I misspoke it isn't "5448")? Future expansion? I believe both are current - they dropped the 4548. As you know, I'm a newbie - the 6048 looked like overkill to me, though I believe one of the installers I spoke with recommended this upgrade also.

Yes, the SCP will let you set everything. manually. 0.1V at a time. Put it at a site you can sit down at.
The combox has no buttons, it's a computer box, and everything happens via a web interface (built in) just hook up the ethernet. Combox also has memory for logging, some smarts for fancy graphics on the web/computer interface if you can't have data draw a chart for you in excel. And it will let you update the firmware on your 3 XW devices, the SCP won't do that at all.

5548 vs 6048 I don't know whats different inside
xw6048 6Kw 132lb
xw5548 5.5Kw 165lb
I had assumed the 5548 was a modernized, budget version of the 6048, and I thought it would weigh less. But I was wrong, so I don't know what the difference is.

And you WILL want the Midnight Epanel that holds all the wiring, gives you a beefy battery bank DC breaker for the inverter, and has lots of buss bars and knockouts for all the breakers and wiring stuff you have to do. (simply connect the inverter to the battery and you have grid quality power) But there is a whole lot of AC and DC wiring to hook up, and the epanel helps manage the mess of wires. It's better then buying parts at the hardware store and trying to get it to work.

[Living Large]

Yes, the SCP will let you set everything. manually. 0.1V at a time. Put it at a site you can sit down at.
The combox has no buttons, it's a computer box, and everything happens via a web interface (built in) just hook up the ethernet. Combox also has memory for logging, some smarts for fancy graphics on the web/computer interface if you can't have data draw a chart for you in excel. And it will let you update the firmware on your 3 XW devices, the SCP won't do that at all.

5548 vs 6048 I don't know whats different inside
xw6048 6Kw 132lb
xw5548 5.5Kw 165lb
I had assumed the 5548 was a modernized, budget version of the 6048, and I thought it would weigh less. But I was wrong, so I don't know what the difference is.

And you WILL want the Midnight Epanel that holds all the wiring, gives you a beefy battery bank DC breaker for the inverter, and has lots of buss bars and knockouts for all the breakers and wiring stuff you have to do. (simply connect the inverter to the battery and you have grid quality power) But there is a whole lot of AC and DC wiring to hook up, and the epanel helps manage the mess of wires. It's better then buying parts at the hardware store and trying to get it to work.

My read was this is not a cheaper version of the 6048, but an upgraded version of the 4548. Here is a side by side comparison (scroll down 1/3 the page):

http://solar.schneider-electric.com/product/conext-xw-na-solar-inverter/

Yep, you have mentioned the ePanel before, and Amy chimed in also. That is in my design, as well as a 6 or 12 position MS combiner.

[inetdog]

My read was this is not a cheaper version of the 6048, but an upgraded version of the 4548. Here is a side by side comparison (scroll down 1/3 the page):

http://solar.schneider-electric.com/product/conext-xw-na-solar-inverter/

That link may be what showed up in your browser, but what you were seeing was the result of a lot of site state and maybe cookies.
From your link, which takes you to the main page, you need to go to battery inverters, then look in XW-NA for the 4548 and 6048 and and in XW+-NA for the 5548 and hit the Technical Info tab. I did not try to find a side by side comparison tool on the site.

[Living Large]

That link may be what showed up in your browser, but what you were seeing was the result of a lot of site state and maybe cookies.
From your link, which takes you to the main page, you need to go to battery inverters, then look in XW-NA for the 4548 and 6048 and and in XW+-NA for the 5548 and hit the Technical Info tab. I did not try to find a side by side comparison tool on the site.

No, you don't have to do all that. I pasted that link into a different browser after clearing cookies, and it takes me directly to the the XW+NA page with the specs side by side - they are down below in a section called "Technical Data". The page is a bit odd, and has a menu bar at the bottom when you load it. It is a long page and takes 5-10 seconds to load on my slow DSL connection. You can either scroll down, or click the Technical Data button and it scrolls for you. I apologize if it doesn't take others to the right page - I can't duplicate that problem. It isn't a "side by side comparison tool" - it is right on the XW+NA page. If the link doesn't work, all you have to do from the main Schneider Solar page is choose the drop down Solutions->Residential->Off grid Solar, and then scroll down and click the XW+NA icon. Two clicks. On the page that comes up, the chart is right there for you - 1/3 the way down.

[Living Large]

Paralleling LFP batteries to obtain needed capacity

I came across a product called MiniBMS, which has a manual that is pretty easy to understand and sounds like it would be a good candidate for someone looking for a BMS that includes balancing. Then I tripped across the charging (BMS) sticky. Nope - I hadn't seen it before. So I saw the debate over whether a BMS is needed. But I digress quite a bit from what is currently vexing and troubling.

The downside of considering a "not ready for prime time" battery chemistry is there don't appear to be a lot of options. I am comfortable with building a battery pack, but finding batteries is a chore. Let's say for the sake of argument that instead of looking for a 400 Ah @ 48V pack, I was looking for 500 Ah or 600 Ah @ 48V?

The first thought I had was, there are CALB 400 Ah and 100 Ah and 200 Ah 3.2V batteries - is there any chance that I could parallel 400 and 100 Ah batteries, and put the result in series? I tried researching paralleling different capacity LFP batteries and came up empty. Can it be done? I'm not sure these have the same physical size. Same question would apply to paralleling 400 Ah and 200 Ah to make 600 Ah.

I saw 180 Ah in a different prism, and wondered about paralleling groups of 3 to make 540 Ah, and this became complicated so I thought I would ask. I saw 300 Ah prisms by ENERGETECH, but I don't know how they stack up to CALB. If they can be paralleled, that would give me 600 Ah.

Basically, I am wondering if I can parallel dissimilar capacities. It appears I could parallel two 400 Ah to make 800 Ah building blocks, but I'd rather not go all the way to 800 Ah.

[Mike90250]

I would think that paralleling strings of batteries, of any chemistry, is going to be as bad as parallel lead acid batteries.

If I was to parallel them, I would only parallel identical cells, 300ah & 300ah, but not a 100 & 400.

what happens if the BMS of one string fired off, what does that do to the other string?

[Living Large]

I would think that paralleling strings of batteries, of any chemistry, is going to be as bad as parallel lead acid batteries.

If I was to parallel them, I would only parallel identical cells, 300ah & 300ah, but not a 100 & 400.

what happens if the BMS of one string fired off, what does that do to the other string?

Mike, my gut feeling is the same as yours - stay with parallel identical cells. But I thought I would toss it out there.

As far as BMS, the cell pairs of 100 & 400 (for example) would have one shunt, and their voltages would obviously be the same. I'm not sure I see a difference between paralleling different capacities or the same as far as BMS goes, but we'll see if others chime in with thoughts.

The only reason I am suggesting this is I can't find 250Ah or 300Ah or 350Ah packs I trust. Maybe someone knows of a supplier. As you are keenly aware, my cost will go up dramatically if I need to go to 800Ah with parallel 400Ah packs. By the way, I didn't know you are blazing a trail with NiFe - good stuff. I enjoyed the pictures you posted.

[northerner]

Dealing with Mr Peukert

I was thinking about the possibility of using sodium ion batteries for off grid storage. The down side with them is they are less efficient at higher discharge rates and their voltage range is quite high from fully charged to discharged. The voltage drop is also considerable with high current draws. This means that when you do have higher usage, (to borrow a phrase) you are dealing with Mr Peukert, and have higher losses, and thus less efficiency than optimum.

The other potential problem that could crop up is that the voltage of the bank could sag below the minimum for many inverters (designed for FLA batteries) and shut down with a considerable load placed on the bank when the bank is at partial state of charge.

The advantage of sodium ion batteries is that they potentially have a long cycle life and low cost overall if managed carefully (ie no heavy loads, and not discharged below about 50%). They also have little concern about balancing, and are environmentally friendly.

The advantage of LFP batteries is that they are capable of a relatively high rate of discharge/charge and maintain a relatively stable potential.

What I was thinking about is adding a smaller set of LiFePo4 batteries to a larger bank of sodium ion batteries. These LFP batteries would only be called into use when heavy loads are placed on the system. This would help stabilize the system voltage under heavier loads and prevent the inverter from kicking out. It would also prevent the heavier draws from the sodium ion bank, thus keeping overall storage efficiency high. Overall cost would be kept low, as the sodium ion batteries have a potentially good return if managed carefully, particularly if battery prices drop, as expected, near the end of this year.

I know that mixing battery types is generally not a good idea, but if managed carefully could work to one's advantage. Thoughts?

[Sunking]

Do not Parallel LFP of different capacities.

All EV manufactures parallel lithium batteries. Tesla does it on steroids. 69 cells in parallel forms a brick of 3.75 volts at 13 AH. 99 Bricks in series to make a Sheet of 375 volts @ 13 AH. 11 sheets in parallel to make 375 volts @ 144 Amp Hours. In all over 6000 batteries. Try finding that one or two bad cells on your own. Tesla uses water to both cool and heat the batteries. Tesla uses both BMS and Thermal Management. All EV manufactures use BMS and Thermal Management.

Question is do you need to use BMS and Thermal management for a grid tied solar system?

I will answer from my POV and let you decide. If I were building an Energy Storage system for a power plant to charge at night, I would use the best BMS and thermal management money can buy to protect a multi-million dollar investment. On a off-grid solar system you are not operating the batteries at maximum limits of charge discharge capacity like an EV or utility would. Not are you using LiCo which requires thermal management. So if it were my system just a simple BMS with some capability to monitor cells to protect a $10,000 investment. Well worth the $200 to $400 it cost to sleep well at night IMO.

[Living Large]

Do not Parallel LFP of different capacities.

All EV manufactures parallel lithium batteries. Tesla does it on steroids. 69 cells in parallel forms a brick of 3.75 volts at 13 AH. 99 Bricks in series to make a Sheet of 375 volts @ 13 AH. 11 sheets in parallel to make 375 volts @ 144 Amp Hours. In all over 6000 batteries. Try finding that one or two bad cells on your own. Tesla uses water to both cool and heat the batteries. Tesla uses both BMS and Thermal Management. All EV manufactures use BMS and Thermal Management.

Question is do you need to use BMS and Thermal management for a grid tied solar system?

I will answer from my POV and let you decide. If I were building an Energy Storage system for a power plant to charge at night, I would use the best BMS and thermal management money can buy to protect a multi-million dollar investment. On a off-grid solar system you are not operating the batteries at maximum limits of charge discharge capacity like an EV or utility would. Not are you using LiCo which requires thermal management. So if it were my system just a simple BMS with some capability to monitor cells to protect a $10,000 investment. Well worth the $200 to $400 it cost to sleep well at night IMO.

OK, there was an intervening post asking about paralleling different chemistries, and that blew my mind. For my needs, keep it simple - same chemistry, same exact thresholds.

My thought was the BMS (yes, I still plan to have one - though PN made some decent points I'd like to discuss another time, and probably in that thread) would work the same because the LV and HV thresholds are the same be it 100Ah or 400Ah. The shunt would hopefully work whether it is 400Ah or 400 + 100 = 500Ah. But this may by whistling Dixie - heck what do I know? If it is a gamble, it isn't worth it and I'll move on to another solution.

The problem arises, as I described, from a thin selection of capacities to choose from. To get 500Ah, my choice now is to parallel three 180Ah packs * 16 = 540Ah, or step up to two ENERGEX 300Ah * 16 = 600Ah. I don't know if ENERGEX has a comparable rep to CALB.