Lithium 48V Battery

sub3marathonman

karrak said:

When you say one cell was deeply discharged, do you know what voltage it got down to, for how long and how many times?

Do you know how much usable capacity is left in the damaged cell and in the other cells in the battery?

Looking at some of your old posts I see that the battery has a "Clean Power Auto" BMS and balancing system. As far as I am aware this system has a reasonable reputation. What makes you think it is one of the BMS modules that has caused the problem?

One major flaw with that BMS is that it does not send the voltage readings to the outside world so you have no idea what is happening with the individual cells.

If there was only one cell damaged you can just replace it with a new one.

Simon

I am thinking it was about 2.12V or so, for several days.  

Now, that's another interesting story of incompetence at the USPS.  So, I found a great place for a Li battery charger, Powerstream.  I called them up, desperate to get their charger from Utah to here in Florida as soon as possible.  I paid for Priority Express, Powerstream had just a short time to get everything boxed up and shipped, but they did it!  Spectacular!  Now all I had to do was check when it would be delivered.  Hours went by, nothing on the USPS tracking.  It turned out, the guy drove around with the package all day, not returning to the post office until AFTER the deadline for getting it sent.  OK, I could live with that, of course it would go out the next day, even though it was a Saturday.  NOPE.  Sunday?  NOPE  OK, Monday is a Federal Holiday, so NOPE.  It turned out it took six days to get here.  In the meantime, I did manage to keep the cell charged a bit, as I couldn't turn on the XW6048 charger, the other cells were still charged.  So, with the BMS able to eliminate the small amount of energy for those cells, I managed to charge at dusk and dawn, turning on and off individual strings as the amperage allowed.

I did manage to get the Powerstream charger, and got the cell charged up to at least a minimum acceptable level.

Now, as Simon mentioned, the BMS from Clean Power Auto (Dimitri) did have one module that stayed in the "On" position.  I've had a thermostat stuck in the "Off" position and subsequently cracked a head on an engine, so I understand that on occasion a part can fail.  I was able to correspond with Dimitri, I have only excellent things to say.  He was and is very helpful and took a fair amount of time to explain things to a non-expert such as me.  Also, as mentioned, the BMS does not send the voltage readings anywhere.  I had purchased a system to monitor the individual cell readings, but I never got around to installing it.  I really should have been checking cell voltage on a more regular basis, but everything seemed to be working, so I didn't.  I think Dimitri's system is excellent, and high quality, and were I needing a system I would not hesitate to recommend Clean Power Auto.

I would also advise anybody with a Li battery bank to get an individual Li cell charger from the people at Powerstream too.  I could have avoided spending money for priority shipping, which was of course wasted by the USPS incompetence out in Utah.

Now, the one problem with the "only one cell damaged" is that these cells, purchased from Balqon, are old Thundersky batteries, either stored or used for who knows how long, substituted for my order of Winston Li batteries.  Balqon of course is belly-up, as I tried to warn Winston long ago what a poor representative they are for Winston.  There is no chance to get another cell.  

Raj174 had an interesting idea, to remove the one cell and make it a 15 cell pack, but I'm unsure about doing that.  I'm not absolutely sure either how the + and - series connections would be with removing a cell, as then it seems the main negative cable would have to be put together with a negative intercell connection, which doesn't seem as nice as how it is now with the 16 cell pack.

I'm not sure about the entire pack capacity, as I don't want to stress any of the cells, and especially the one cell that had a problem.  At this point I'm just thankful that the XW6048 can work with them in float mode, and have somewhat had to abandon any load shifting from peak to off-peak that I was doing.  The batteries were theoretically supposed to last for 10 years, Mr. Samra gave his word and guarantee that they'd last for five or he'd replace them, I'll hold my breath for that.  Now, they're at about 2 years and 4 months, so not even halfway to five.  I'm not sure how much longer I can get them to last.

Raj174

Hi sub3marathonman,
A few questions.
Can you post a picture of your battery bank?
What are the charger settings? 
What is the size of the PV array?
What is the amperage of the LFP charger that you bought? I'm assuming it's a single cell charger.
Does the BMS have a readout of the individual cell voltages? If so, what is the highest and lowest at float?
In what way does the deficient cell alter the way you would normally use the system?
 
Small world, I live in Florida too.
Thanks,
Rick 

mnittler

I did manage to find a Tesla Battery warrantee online and it looks like the batteries should maintain 80% of capacity for 10 years or will be repaired or replaced at Tesla's discretion.

karrak

sub3marathonman said:

I am thinking it was about 2.12V or so, for several days. 

I did manage to get the Powerstream charger, and got the cell charged up to at least a minimum acceptable level.

What is the model number of your Powerstream charger?

As Rick has stated it would be good to get some pictures and specs of your system. I am especially interested to know what other equipment the Clean Power BMS and battery is connected to to protect the battery.

Ideally I think you should disassemble your battery and check the capacity of each cell by discharging it, then recharging it while monitoring the amount of charge going into each cell. This is not a trivial operation with your 700Ah cells.

An easier approach would be to disconnect the battery and charge each cell individually with your Powerstream charger. This can be done without disconnecting the individual cells from each other. When all 16 cells have been charged individually your whole pack will be top balanced. I would then connect the battery up to your loads and drain the battery while monitoring the individual voltages and the Ah being drawn. The first cell that gets to 2.8V is the weakest in the battery and will determine the capacity of the whole battery.

I had purchased a system to monitor the individual cell readings, but I never got around to installing it.

I am interested to know what monitoring system you purchased. I am yet to find a good commercial system. That is the reason I designed my own.

I would also advise anybody with a Li battery bank to get an individual Li cell charger from the people at Powerstream too.

You can also get a variable supply which gives you more flexibility but does require more care when using it.

Now, the one problem with the "only one cell damaged" is that these cells, purchased from Balqon, are old Thundersky batteries, either stored or used for who knows how long, substituted for my order of Winston Li batteries.  Balqon of course is belly-up, as I tried to warn Winston long ago what a poor representative they are for Winston.  There is no chance to get another cell.  

The cell doesn't need to be from the same manufacturer. As long as its specifications are equal to or better than the remaining 15 cells you will not have any problems as long as you have individual cell monitoring. There is one LFP based offgrid system system I know of in Australia where one cell had lost around 25% of its capacity and the owner replaced the cell with a cell from a different manufacturer. If you did change the cell I would change it with either a Winston cell which are now sold in the US by Voltronix or a CALB cell as these have similar characteristics.

I'm not sure about the entire pack capacity, as I don't want to stress any of the cells, and especially the one cell that had a problem.  At this point I'm just thankful that the XW6048 can work with them in float mode, and have somewhat had to abandon any load shifting from peak to off-peak that I was doing.  The batteries were theoretically supposed to last for 10 years.

The other 15 cells should be just fine if the only fault has been the 1 cell being discharged by the faulty BMS.

A word of warning about floating LFP batteries. Unless they are being used continuously in say an off grid system that has appliances drawing load on a continuous basis you should not float the battery. If you are load shifting I would program your charge controller to terminate the charge and not float. If you are not going to be using the battery for a while you should discharge it down to around 50%SOC.

Simon

Dave Angelini

mnittler said:

I did manage to find a Tesla Battery warrantee online and it looks like the batteries should maintain 80% of capacity for 10 years or will be repaired or replaced at Tesla's discretion.

I think if you are really interested in the warranty you will need to look farther than what you have posted. The one I saw had 10 pages and the 10 years (simplifying here) is based on cycles, DOD, ambient temp, and over current which is logged.

karrak

Photowhit said:

Cost vs Capacity is an on going theme, many would not dare have an off grid system without alternative charging. But for those of us who dare. We need the extra capacity! It won't be used often. A 16Kwh forklift battery can be drawn down to 20% SOC even if you normally don't use more than 2Kwh's a day. It can NOT be replaced with a 8 or even 10 Kwh Lithium battery bank without the added expense of a generator!

I think the subject of how much reserve battery capacity to have, especially if like you or I you one doesn't have a backup generator is quite complex.

I agree that if you only have enough solar panels to cater for average power usage over a period of time that the lithium battery needs to be around the same size as a lead acid battery. You can get by with a smaller solar array with a lithium battery because of the greater charge efficiency and much reduced absorb times.

On the other hand if you oversize the solar array you can get by with a smaller lithium battery. This is exactly what I have done with a second LFP (LiFePO4) based system that I have installed for a friend. Both of our systems have ~10kWh of battery storage. When both of our systems were installed our daily power requirements in winter were around ~3kWh/day. About three months ago my friend's two parents moved into their 'granny flat'. The parents will be staying there for about a year. Because of this their energy requirements have nearly doubled. To cater for this we doubled the size of their solar array while keeping the rest of their off grid system the same.

This makes economic sense now as solar panels have become so cheap, especially if you buy second hand panels. We have the crazy situation in Australia that because of the energy credits one gets when installing new equipment that if you want to upgrade an existing system it is cheaper to buy new panels than use the old ones. This has meant that the price of second hand 4-5 year old solar panels is around A$0.20-A$0.25 (US$0.15-US$0.20) per watt!

Simon

karrak

mike95490 said:

Lithium has ZERO forgiveness.   Overcharge them once (like forgetting to disable temperature compensation) ruins them, deeply discharging them once, ruins them.

Forgetting to disable temperature compensation will not ruin an LFP battery. The standard temperature compensation coefficient for lead acid batteries is ~-3mV/cell/degree C. For a 6 cell 12V LA battery this equates to -18mV/degree C. Charging a 4 cell 12V LFP battery at the recommended charge voltage of 13.8V (3.45V/cell) at 0 C and forgetting to turn the temperature compensation off would increase the charge voltage by 0.45V (25*0.018) to 14.25V which equates to 3.56V/cell.

This is below the 3.65V/cell that most LFP manufacturers recommend as the maximum voltage to charge an LFP battery at and way below the threshold voltage of ~4.4V/cell (17.6V for a 12V battery) above which the electrolyte starts to rapidly break down causing serious damage to the battery.

On the discharge side you would have to take a LFP cell below ~2 volts (8V for a 12V battery) before you would get any rapid damage to the cell.

Simon

Dave Angelini

The other thing that is often left out with LFP is the surge ability and the effect it has on life. I have not seen any yet that I can get that will support more than 5KW continuous and a 7KW surge. My large customers need this and I have had to stack inverter/chargers to get the surge and capacity. 

Estragon

What happens to a LFP battery charged at <0°C?

As I understand it, they can be discharged at freezing temps, but not recharged. I don't know what happens to them though.

westbranch

Another part of that question is 'do LFP cells warm up like FLA cells do when discharging?' 
If they warm up, how does that impact the lowest ambient temperature they will charge at?

Dave Angelini

I really have no idea how the individual cells do in a home brew system like Simon's. He and Rick have done a great job here but
for a system like the LG RESU or Tesla system #2 the BMS can log just about anything that could be bad for warranty. If systems are in places that are not around the temps that humans like, one should not expect long life. I have heard that Tesla will has similar specs but is not ready until June. Not sure about Tesla, is anyone?

Operating Temperature is -10 to 45 Degrees Celsius

One important piece of information missing from the above tables is its operating temperature.  RESUs can function at ambient temperatures from -10 to 45 degrees Celsius.  Ambient simply means the air temperature in the immediate surroundings.  It is possible that some day we may know why battery manufacturers don’t just say, “local air temperature”.

The lower limit is not likely to be a problem for Australians, but it is possible for it to shut down during a severe heatwave.  To prevent it conking out from heat exhaustion, installing RESUs out of direct sunlight is a very good idea.  LG Chem gives its recommended operating temperature as 15-30 degrees, so placing it indoors is probably a good idea and will help improve its efficiency and extend its battery life.

It’s Quiet

The RESUs have no cooling fan and rely on natural convection from its metal case for cooling.  This means they are very quiet, which is useful for indoor installation.

The 48 volt RESUs’ warranty promises they will still have at least 60% of their nominal capacity after 10 years or until they have stored a total amount of electricity just under 2,500 kilowatt-hours for every kilowatt-hour of nominal capacity.  This comes to:

• RESU3.3  8,200 kilowatt-hours
• RESU6.5  16,100 kilowatt-hours
• RESU10  24,300 kilowatt-hours.

If your RESU cannot keep its capacity up, then LG will repair or replace it.  If they no longer produce RESUS they have the option of compensating you with cash.  Unfortunately, the amount of money they offer can be pretty insignificant and depends upon the age of your system:

• 0-2 years:  100% of purchase price
• 2-3 years:  72% of purchase price
• 3-4 years:  58% of purchase price
• 4-5 years: 44% of purchase price
• 5-6 years:  30% of purchase price
• 6-7 years:  16% of purchase price
• 7-8 years:  6% of purchase price
• 8-9 years:  4% of purchase price
• 9-10 years:  2% of purchase price

westbranch

Well in my case we did not go to the cabin  for > 3 weeks as the external temps were down to - 30* C , when we got out when it warmed up to ~ - 20*C in the day, the batt temps were ~ + 4*C based on the temps seen via the Local App.  I'm not positive just what was keeping the basement at that 'warmer  temp' but I suspect that as the skies were clear , the 2 1800# batts were getting warmed daily by an extended End Amps setting EA =1.0.... not sure how long the heating would have gone on for if the temps headed to - 40*C....
I do know that the R25 Logix ICF insulated basement walls plus R28 8" walls and an R40 roof with large hi tech S facing Triple E windows  have an impact,  ie outside -15*C, inside +5*C when the sun was shining with no heat on for 3 weeks...

Estragon

Assuming there were no loads and the bank would be full, would the absorb still heat the batteries up much?

The ground at basement floor level will stay pretty constant year-round. In my insulated but unheated crawlspace, the ground moderates extreme temps quite well. Your well insulated building with good heat gain would help a lot too.

westbranch

Yes, I had thought that I would need some Hydronic heating and with solar water heaters, piping in the concrete since yr 1...  objective was to keep the temps above 5*C, in the basement, but after this winters experience my 'cold battery' fears are a bit out of proportion, I hope...

I have noticed the batteries temp rising during the long absorbs , especially when the days were getting longer before the Mar. 21 solstice, this with the fridge running 27/7 while there, after Mar 22 fridge has been on constantly..

Raj174

I found this bit of information:

From this article titled "Nonaqueous Liquid Electrolytes for Lithium-Based Rechargeable Batteries" by Kang Xu from the Electrochemistry Branch, Sensor and Electron Devices Directorate, U.S. Army Research Laboratory, Adelphi, Maryland 20783-1197 published in the Chem. Rev. 2004, 104, pp4303-4417.

"Thus, at temperatures lower than the liquidus temperature (usually above -20 °C for most electrolyte compositions),50e,159,160 EC precipitates and drastically reduces the conductivity of lithium ions both in the bulk electrolyte and through the interfacial films in the system. During discharge, this increase of cell impedance at low temperature leads to lower capacity utilization, which is normally recoverable when the temperature rises. However, permanent damage occurs if the cell is being charged at low temperatures because lithium deposition occurs, caused by the high interfacial impedance, and results in irreversible loss of lithium ions. An even worse possibility is the safety hazard if the lithium deposition continues to accumulate on the carbonaceous surface.

"At temperatures higher than 60 °C, various decompositions occur among the electrolyte components, electrode materials, and SEI or surface layers, while LiPF6 acts as a major initiator or catalyst for most of these reactions.152,310,332,333 The damage caused by high-temperature operation is permanent. Because gaseous products accumulate, a safety hazard is also likely. Therefore, the specified temperature range for the normal operation of most commercial lithium ion cells is -20 °C to +50 °C. While sufficient for most consumer purposes, the above range severely restricts the applications of lithium ion technology for special areas such as military, space, and vehicle traction uses."

Winston battery company also makes a LFP cell with the element Yttrium in the electrolyte and claims that this reduces the charge discharge temperature to -45C. Link to their battery specs below.

http://en.winston-battery.com/index.php/products/power-battery/item/wb-lyp400aha?category_id=176

Rick

westbranch

Very interesting, and illuminating facts about LI chemistry batteries..
Apparently from my contact there is a new version of the Lithium Manganese Oxide (LiMn2O4)... or LMO, it has a lower temp limits than LFP, less cycles but higher energy density than LFP in cold but lower output in higher temps...

mike95490

karrak said:

mike95490 said:

Lithium has ZERO forgiveness.   Overcharge them once (like forgetting to disable temperature compensation) ruins them, deeply discharging them once, ruins them.

Forgetting to disable temperature compensation will not ruin an LFP battery. ..........

Simon

  I'm stating this because of neophytes that are not used to Li behavior. Flooded batteries are very forgiving, AGM, a little less, Li, much less, and they violently burn when abused.
How would you feel, if someone doing research, saw your post, and based their decisions on it, and paid an expensive price later. So I error on the side of abundant caution, and I sleep well at night.
 YOU may make the decision to run to the limits, but then you are erasing all margin for error.

BB.

And, with Li-Ion batteries, chemistry details matter... There are many different types of cells and each has their own min/max voltages, temperatures, and operating conditions. Some will catch fire if mishandled. Others Li Chemistries are supposed to be more forgiving.

-Bill

karrak

mike95490 said:

karrak said:

mike95490 said:

Lithium has ZERO forgiveness.   Overcharge them once (like forgetting to disable temperature compensation) ruins them, deeply discharging them once, ruins them.

Forgetting to disable temperature compensation will not ruin an LFP battery. ..........

Simon

 
 YOU may make the decision to run to the limits, but then you are erasing all margin for error.

Please provide more details as to what you mean by  "run to the limits"? What are your recommendations on charging and discharging an LFP battery that avoid the limits?

Simon

Dave Angelini

I would never speak for Mike  
My answer to your question Simon is that I do not want any of my clients ever having to do what you and Rick are doing. I want a pre-made battery bank with a BMS built in that communicates with the offgrid power system and I do not want any set-points or adjustments.
There will be a bunch of makes of LFP batteries that do this with Outback and Schneider very soon. I really like the results I am seeing  with the Schneider Bridge in my garage and the Outback Skybridge is moving along.

I think it is great that you guys are doing this.

Estragon

@Dave Angelini

Tesla and presumably other EV makers seem to be incorporating battery heaters in their charging to get around the low temp charging issues, and to extend low temp range.

Do you know if the new stationary systems will do this too? Some garages get pretty cold in these parts.

Photowhit

Good question, I've been thinking about where I would put a couple units here inside, if I made the switch. They are light and small enough I might mount them on the wall above my freezer, but I would have to reinforce the wall and move the freezer out a few more inches.

westbranch

I thunk similarly to Dave except that I want to stay away from the outer limits, Hi and Low. Ie run in the 20% to 80% +-zone  There has to be a setting to program, if not. that is not the system for me. Reason is if a SHTF happens you still have some 'emergency' power to keep the light on so to speak while you work at fixing the problem... last resort start Gen set...

Dave Angelini

There is a Troubleshooting mode built into "one" system that allows certain parameters to be changed to a lower range.
The standard for all of this is CANBUS and the programming is wide open. I am not sure it will survive the Beta though.

The longest warranted life is for temperatures that people like and it is logged for warranty. This could all change with newer models. 

The other thing that is love hate about what I am seeing with Tesla and LG is the 5KW continuous max and 7KW surge plus other limits to protect the battery. Even a second Tesla or a second 10KWH LG bank has these limits. The current models use only one battery at a time. Large systems are not going to be here as soon as smaller ones. If you have issues with SHTF a second battery bank or the same thing you would do now, start a generator. You can always add a second bank later this way as they will not interfere, each bank does not care about the age of the other.

The good news is the battery has firmware also, this all can be massaged down the line as these systems mature.

I really like that I can load a 165 pound battery with a 10KWH capacity in my truck and hook-up two 3 foot battery cables and a canbus 2 wire connector. Just check or update the firmware, plug the controller into the bus, go home and watch the system on the web. I like that I don't need the web also

mike95490

karrak said:

Please provide more details as to what you mean by  "run to the limits"? What are your recommendations on charging and discharging an LFP battery that avoid the limits?

Simon

I mean that I would treat Li batteries very carefully,   I would be sure to disable Temp Comp in the charger, I would stay in the middle of the charge curve, well below the empty and full knees where things change rapidly and where the BAD stuff happens. The limits vary with each install and bank. You disagree with me, and that's fine.  I prefer to give conservative advice, and apparently, you like to push the limits and assume all the protection gear will always work perfectly and save the battery bank. 
 This weekend, I hope to get pics of a large "protected" Li bank that burned an outbuilding down.  the owner hauled them out, but the building was a loss.  The Fire Dept arrived after it was all out.  He's  been running on generator for 2 months now and really wants to rebuild but with a forklift battery and not Li.

karrak

Dave Angelini said:

I would never speak for Mike  
My answer to your question Simon is that I do not want any of my clients ever having to do what you and Rick are doing. I want a pre-made battery bank with a BMS built in that communicates with the offgrid power system and I do not want any set-points or adjustments.
There will be a bunch of makes of LFP batteries that do this with Outback and Schneider very soon. I really like the results I am seeing  with the Schneider Bridge in my garage and the Outback Skybridge is moving along.

I think it is great that you guys are doing this.

I agree that for the vast majority of people what you suggest is the best way to go  but there are a small percentage of people who want to DIY for whatever reasons.

IMO it is better to give anyone who is interested in a DIY lithium ion battery well researched information and experience about the dangers and how to safely operate such a battery to the best of ones knowledge and ability and be prepared to back ones information with evidence. One should also be prepared to look at other peoples evidence and points of view and back down when it is obvious that the information one is providing is wrong.  Scaremongering with half truths or downright false information or providing information on how to operate a lithium ion battery that is based on hearsay or ones own untested ideas is counterproductive.

Simon

karrak

mike95490 said:

I mean that I would treat Li batteries very carefully,   I would be sure to disable Temp Comp in the charger, I would stay in the middle of the charge curve, well below the empty and full knees where things change rapidly and where the BAD stuff happens. The limits vary with each install and bank. You disagree with me, and that's fine.  I prefer to give conservative advice, and apparently, you like to push the limits and assume all the protection gear will always work perfectly and save the battery bank. 

So how do you keep an LFP battery within the empty and full knees, lets say 90% and 10%, or would you pick 80% and 20% when charging with solar power?

I do not push the limits.

• For LFP batteries the absolute voltage limits before things really bad start happening quickly is ~4.4V/cell to ~2.0V/cell. Normal voltage operating limits are 3.65V/cell to 2.5V/cell. Unless I am doing some tests my maximum operating range is 3.50V/cell to 2.8V/cell.
• Maximum charge current for the large cheaper Chinese prismatics is around ~1C and maximum discharge current is around ~2C, my maximum charge current is ~0.1C and maximum discharge current is ~0.6C.
• A conservative operating temperature range would be  between +45C and 0C. The temperature range my battery operates in is ~+40C to ~+10C.

My protection gear is only there in case of a fault or stuff-up. The only time it has been used in anger is when my battery was first installed and was badly out of balance (around 30% out) due to a stuff-up from the battery supplier and me assuming that the battery had been balanced. For the past four years monitoring, the solar controller charge parameters and an occasional manual balance are all that has been required to keep my LFP battery and my friend's battery in their safe operating zone.

 This weekend, I hope to get pics of a large "protected" Li bank that burned an outbuilding down.  the owner hauled them out, but the building was a loss.  The Fire Dept arrived after it was all out.  He's  been running on generator for 2 months now and really wants to rebuild but with a forklift battery and not Li.

What maybe as or more useful than pictures would be a description of what protection equipment there was protecting the battery, what type of lithium ion battery it was, what the charge controller charging parameters were and if possible a description of what happened and possible causes. With this information maybe we could all learn what mistakes were made so we do not repeat them.

Simon

Dave Angelini

bill von novak

Dave Angelini said:

SKYBOX

Anyone have any details on that thing yet?  DC intermediate voltage?  400 or 48 volts?   "Charge controller" (MPPT tracker) integrated?  Max DC voltage in?  Power?

jonr

Sounds like a combined grid-tie inverter, off-grid inverter and charge controller.  I found 600V, 6000 watts (2x 3000) input.  Batteries are optional.  Can work with lead acid, so perhaps 48V.