LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[poiu]

Originally Posted by svAricia

Finally got my order from Balqon, originally ordered 12 400ah cells back in December with their shipping promo. They still don't have 400ah cells in stock, been emailing back and forth for few month until just recently I told them I couldn't wait any longer so they upgraded me 700ah cells Good thing I have enough space and 2100ah should be plenty.
Not sure how old the cells are but they all have Balqon logos on them, checked their voltage and it ranges from 3.303 to 3.293

I am having great difficulty getting hold of Balqon, let alone getting a delivery.

I bought a pack of 8 x 700Ah batteries a year ago and one has failed recently. Its voltage doesn't go beyond 3.36v on charge and it discharges down to 2.5v when the others are at 3.2v. The capacity is about 2/3 the capacity of a full battery and so my pack capacity is therefore reduced to that amount. Mr Samra agreed it was faulty and was going to send a replacement under warranty. Since then, for 3 weeks there has been no communication by email and no one answering the phone and no responses to phone messages. Total silence. Rather worrying. Are they going under? How do you get hold of people there?

[Maine Sail]

Just a quick musing regarding the Cell Log 8S that so many, myself included, use as a tool to monitor voltage of an LFP bank.

Last week while reconnecting and re-installing my LFP bank I realized the Cell Log 8S takes its power from just ONE CELL........ Can't believe I totally over looked this before.

Connect the neg lead and then the #2 position and the unit powers up. While this draw is very, very small the Cell Log 8S could potentially cause a cell imbalance if left permanently connected.. I can't see any way it would not, if left connected long enough.. Would be nice if you could set it up to pull voltage in a 4S config for 12V power but it boots off just one cell..

I leave mine OFF when I am not looking at it, but still a device intended to keep track of cell balance can actually cause a cell imbalance.... D'oh....

Just something to consider.....

[campr]

There is a 'hack' to make it take it's power from all attached cells apparently the VRM supports it:

Endless-sphere.com • View topic - CellLog 8 hacking

hope it helps you if you are really concerned about it.

[NahanniV]

Originally Posted by Maine Sail

Just a quick musing regarding the Cell Log 8S that so many, myself included, use as a tool to monitor voltage of an LFP bank.

Last week while reconnecting and re-installing my LFP bank I realized the Cell Log 8S takes its power from just ONE CELL........ Can't believe I totally over looked this before.

Connect the neg lead and then the #2 position and the unit powers up. While this draw is very, very small the Cell Log 8S could potentially cause a cell imbalance if left permanently connected.. I can't see any way it would not, if left connected long enough.. Would be nice if you could set it up to pull voltage in a 4S config for 12V power but it boots off just one cell..

I leave mine OFF when I am not looking at it, but still a device intended to keep track of cell balance can actually cause a cell imbalance.... D'oh....

Just something to consider.....

The problem only seems to be if you have more than 6 cells in series; In that case no power is drawn from cells 7 & 8; This is probably a leftover from the PowerLog6 design which I think came first. Have a look at the diagram referenced above, the device will power itself from the highest voltage pin (up to pin 6) and should draw evenly from all the cells. I added the cells and painted the reverse biased diodes red and forward biased ones green.

[Hkalan]

Hello,

I am under the impression from the manual and HobbyKing, that it will first take power from the USB side, then if no power there. It pulls from the cells to energize itself.

I installed a 12v cigarette lighter receptacle, to plug in one of those adapters that have the 2 USB (to charge your phone in a car) so I can power the cellLog8 from the complete bank. I was worried it would pull from 1 or 2 cells to power itself, creating a difference between the other cells.

Am I mistaken ?

Alan

[NahanniV]

Originally Posted by Hkalan

Hello,

I am under the impression from the manual and HobbyKing, that it will first take power from the USB side, then if no power there. It pulls from the cells to energize itself.

I installed a 12v cigarette lighter receptacle, to plug in one of those adapters that have the 2 USB (to charge your phone in a car) so I can power the cellLog8 from the complete bank. I was worried it would pull from 1 or 2 cells to power itself, creating a difference between the other cells.

Am I mistaken ?

Alan

You are not mistaken. The USB power is used first. Always powering the USB 5V obviates the problem.

My understanding is that there is only a problem powering from the cells with a 24V system (8 Series cells); In that case cells 7 & 8 would not be powering the CellLog. So, 13 to 20mA would be drawn from cell 1-6 but not 7 &8 (unless you make the mod outlined above).

My PowerLog6 was connected to my 400AH 4S bank all winter and did not cause the cells to get out of balance. It was powered from the cells for the majority of 6months.

Cheers,
JM.

[T1 Terry]

Originally Posted by Maine Sail

Just a quick musing regarding the Cell Log 8S that so many, myself included, use as a tool to monitor voltage of an LFP bank.

Last week while reconnecting and re-installing my LFP bank I realized the Cell Log 8S takes its power from just ONE CELL........ Can't believe I totally over looked this before.

Connect the neg lead and then the #2 position and the unit powers up. While this draw is very, very small the Cell Log 8S could potentially cause a cell imbalance if left permanently connected.. I can't see any way it would not, if left connected long enough.. Would be nice if you could set it up to pull voltage in a 4S config for 12V power but it boots off just one cell..

I leave mine OFF when I am not looking at it, but still a device intended to keep track of cell balance can actually cause a cell imbalance.... D'oh....

Just something to consider.....

over 3 yrs continuous now on 2 different 12v battery banks, no out of balance, it is not an issue with 4 cells in series battery banks. Even some of the 8 cells in series systems seem to have random cells go high voltage atthe top end of their charge cycle, not always cells 7 & 8. If the power drawing only from cells 1 to 6 and not from 7 & 8 was really an issue, 7 & 8 cells would always be the run away cells.
Having said that, in many systems 7 & 8 are regularly the run aways, but further investigation finds there is more than 1 logger on the battery at the same time, so the draw from 2 or more loggers will cause a problem, a very minor one, but a problem all the same.

As far as the reports such as those on the Endless-sphere forum, they are using 10ah cells, you would see an issue on cells with a capacity that small, but not house system capacity batteries.
The powering via the USB port is the ultimate fix, but not via the computer if you want to do logging on the unit itself, once it makes a comunication contact with the computer and the logger program, all information is sent to the computer, not stored on the logger, so a file must be made on the computer and recorded there if you need it.
Log files @ 30 sec sampling as great for reviewing just how the pack is travelling. Poor cell connections stand out instantly, they are the low cells under load yet high cells when charging.

T1 Terry

[mcarling]

I've yet to read, see, or hear anything which convinces me (or even tends to convince me) that LiFePO4 cells need cell-level monitoring even as much as (let alone more than) lead-acid batteries need cell-level monitoring. Everything I've seen, heard, and read here and elsewhere evidences that a balanced set of LiFePO4 cells that is not abused will stay in reasonable balance for years.

I would verify the balance or do a rebalance before a transoceanic passage, but I would otherwise not give a the ongoing balance of a LiFePO4 bank any thought. If I could have any BMS of my choice for free, I would decline and keep it simple.

[CharlieJ]

mcarling-
Yes, but…

A very simple BMS such as the House Power BMS referenced in this thread many times, provides a level of protection for very, very low cost.

I, too, understand that balancing (top or bottom) on a regular basis for the fractional C charge and discharge profile that marine house banks operate in is not necessary. However, having a BMS that monitors parallel strings (cell level monitoring) as well as pack level voltages for incipient HVE and LVE is, in my opinion, worth the minimal incremental cost increase and good engineering.

I will not deploy an LFP bank without a BMS.

[mcarling]

CharlieJ,
The same could be said about lead-acid batteries -- and it would be even more true for lead-acid batteries.
I think a fair case can be made for HVE and LVE monitoring and disconnect, but the advantage of doing that at the cell level rather than at the bank level is negligible and, in my opinion, does not exceed the additional complexity and risk. Of all the potential failures which could cause an imbalance among cells, failure of a cell-level BMS may be the most likely.

[CharlieJ]

To each his own…for the negligible cost increase there is a significant increase in safety.

I am currently on my fourth LFP system design…I will continue to use cell level monitoring as part of the BMS unless hard engineering facts come to light that show that this is not the proper design strategy.

[mcarling]

Originally Posted by CharlieJ

To each his own.

Here I agree 100%. To each his own.

It's not the money that's an issue for me. As I wrote above, if I were offered a BMS of my choice for free, I would decline the offer.

Until some hard engineering facts come to light that cell-level monitoring provides a benefit that exceeds the additional risk, I'll continue with my BMS-free system. I have yet to even find a plausible theory (with or without empirical support) for why cell-level monitoring should be more important for LiFePO4 than for lead-acid. Introducing additional complexity and failure modes without either theory or empirical evidence just isn't something I'm willing to do.

In my opinion, these are perfectly suitable for a small house bank (despite the absence of cell-level monitoring):
http://en.winston-battery.com/index....ory/lp-battery

[T1 Terry]

Here we go :lol: Macarling, do you have an Li battery system up and running? If so, at what voltage do you stop charging?

The simple facts are, lead acid cells boil their electrolyte if severely over votage charged, but they become electrolysis while charging and at low voltage charging, I think that is fairly well known, what happens when the same thing occurs when using Li cells?

T1 Terry

[mcarling]

Originally Posted by T1 Terry

Here we go :lol: Macarling, do you have an Li battery system up and running? If so, at what voltage do you stop charging?

I have a nominal 36V LiFePO4 system up and running as part of a home-brew electric bicycle. I charge at 42 (12 x 3.5) volts. Using a voltmeter, I have never detected an imbalance of more than a few thousandths of a volt.

Originally Posted by T1 Terry

The simple facts are, lead acid cells boil their electrolyte if severely over votage charged, but they become electrolysis while charging and at low voltage charging, I think that is fairly well known, what happens when the same thing occurs when using Li cells?

I don't know what you mean by "become electrolysis". Electrolysis is a process, not a thing which something can become. Therefore I'm not sure I understand your question exactly. The details of what happens during charging is different for different Li chemistries. The failure modes for LiFePO4 are much more benign than for lead-acid. A severely overcharged LiFePO4 prismatic cell will increase in temperature and pressure and could permanently deform and eventually rupture its outer plastic case, though the individual cell packets inside should remain intact (though permanently deformed). It's an extremely rare and not especially exciting failure. It would just require replacement.

Again, I have not argued against HVE and LVE monitoring and automatic disconnect, however, I would implement them on a battery bank level rather than on an individual cell level. When was the last time you saw HVE and LVE monitoring for individual lead-acid cells?

[T1 Terry]

I don't know what you mean by "become electrolysis".

The bugga of letting spell checker do it's thing without rereading it :lol: It should have read, "the lead acid cell becomes an electrolyser"
If you have a 36v series string battery then connecting 2 cell loggers across the 12 cells, watch the cell votage as you approach the 42v mark, are all the cells at exactly 3.5v at the same time the 42v mark is reached? Keep a log over multiple discharge/recharge cycles ad watch what happens just at the end of the 42v charging cycle, then you will understand the need for individual cell voltage monitoring.

T1 Terry