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

[Maine Sail]

Originally Posted by Jd1

Could you please clarify this statement for me? It seems that you are saying that Balmar wants you to cut B+ but cutting B+ will vaporize the diodes.
?????????????????

This is why I specifically bolded REGULATOR B+.

On a Balmar regulator this is the red positive wire that powers the REGULATOR (not the red v-sense wire).

It is not the alternators output wire or ALT B+ it is REG B+, and that is a big difference.

Cutting the REG B+ will do nothing to the alt diodes but cutting the ALT B+ can vaporize them....

[senormechanico]

MainSail,
I know what you mean about the steps for alternator and belt preservation, but we have plenty of solar.
The alternator rarely gets any kind of load for more than a couple of minutes.

[deckofficer]

Results of cycle testing.

Capacity test cycle (#38) 2.75 hours @ 41 amp draw 112.75 a-hr from a 100 a-hr pack.

Discharge 11.61 pack voltage. 2.96, 2.91, 2.75, 2.99


Capacity test cycle (#100) 2.75 hours @ 41 amp draw 112.75 a-hr from a 100 a-hr pack. (Same)
Discharge 11.57 pack voltage. 3.00, 2.94, 2.60, 3.03

That #3 cell is proving to be the weakest but solid overall performance that shows almost zero capacity loss after 100 cycles.

[goboatingnow]

Originally Posted by deckofficer

Results of cycle testing.

Capacity test cycle (#38) 2.75 hours @ 41 amp draw 112.75 a-hr from a 100 a-hr pack.

Discharge 11.61 pack voltage. 2.96, 2.91, 2.75, 2.99

Capacity test cycle (#100) 2.75 hours @ 41 amp draw 112.75 a-hr from a 100 a-hr pack. (Same)
Discharge 11.57 pack voltage. 3.00, 2.94, 2.60, 3.03

That #3 cell is proving to be the weakest but solid overall performance that shows almost zero capacity loss after 100 cycles.

Can I ask , in all seriousness, why a technology capable anything from 800 cycles to 2000 , would give you any concerns at 100 cycles.

Dave

[deckofficer]

Originally Posted by goboatingnow

Can I ask , in all seriousness, why a technology capable anything from 800 cycles to 2000 , would give you any concerns at 100 cycles.

Dave

100 enroute to 2000. I was just curious if at 100 cycles if there was any noticeable lack of capacity and except for pack voltage being .04 a volt less, it still did the same work.

[Maine Sail]

Originally Posted by deckofficer

Results of cycle testing.

Capacity test cycle (#38) 2.75 hours @ 41 amp draw 112.75 a-hr from a 100 a-hr pack.

Discharge 11.61 pack voltage. 2.96, 2.91, 2.75, 2.99


Capacity test cycle (#100) 2.75 hours @ 41 amp draw 112.75 a-hr from a 100 a-hr pack. (Same)
Discharge 11.57 pack voltage. 3.00, 2.94, 2.60, 3.03

That #3 cell is proving to be the weakest but solid overall performance that shows almost zero capacity loss after 100 cycles.

Bob,

I really think you would be wise to trigger shut down based on cell#3 or any cells voltage and not total pack voltage. 2.60V is a tad low. Personally I do not like to see mine dip below 2.8V, ever. Even during capacity tests I cut at 2.9V for the lowest cell. The total capacity of the bank IMHO is determined by your lowest CELL not by total pack voltage because as you can see you have one cell that is really dipping into the danger zone.........

You really want an Ah counter so you can cut your cycles at 80Ah's consumed or 80%... You are doing full blown capacity tests to 100% DOD and I would not expect these cells to get anywhere near claimed 2000 cycle capacity because those claims are made based on 80% DOD...

[deckofficer]

Originally Posted by Maine Sail

Bob,

I really think you would be wise to trigger shut down based on cell#3 or any cells voltage and not total pack voltage. 2.60V is a tad low. Personally I do not like to see mine dip below 2.8V, ever. Even during capacity tests I cut at 2.9V for the lowest cell. The total capacity of the bank IMHO is determined by your lowest CELL not by total pack voltage because as you can see you have one cell that is really dipping into the danger zone.........

You really want an Ah counter so you can cut your cycles at 80Ah's consumed or 80%... You are doing full blown capacity tests to 100% DOD and I would not expect these cells to get anywhere near claimed 2000 cycle capacity because those claims are made based on 80% DOD...

Your correct, I am pulling cell # 3 too low on the capacity test. On a regular cycle they are all above 3.00 volts. A regular cycle is 41 amps for 2 hours, a capacity test is 41 amps for 2 hours and 45 minutes. I will end future capacity tests earlier so that cell # 3 is at or above 2.9 volts.

[durundal]

So having tried to get as much info as I can and having followed this entire thread, I can't recall a case of it happening, but has anyone had LiFePO4 issues that resulted in damage to anything outside of the cells or associated electrical stuff? No cases of fires or venting that ignited or sudden bus polarity reversals or anything else unlikely and weird? Going round and round tracking millivolts is likely to scare off more people worried about complexity than being able to say "Try it, it can't burn your boat down and worst case the only thing you're out is money if you toast the cells by trying real hard to do so."

[goboatingnow]

Originally Posted by durundal

So having tried to get as much info as I can and having followed this entire thread, I can't recall a case of it happening, but has anyone had LiFePO4 issues that resulted in damage to anything outside of the cells or associated electrical stuff? No cases of fires or venting that ignited or sudden bus polarity reversals or anything else unlikely and weird? Going round and round tracking millivolts is likely to scare off more people worried about complexity than being able to say "Try it, it can't burn your boat down and worst case the only thing you're out is money if you toast the cells by trying real hard to do so."

The answer is no and a long shrug and sheeeeesh

What is the basis for your question ? What prompts you to ask it.

Dave

[Maine Sail]

Musings on my charging voltage:

I was going through some notes I took early late last year when I was doing the top balance to my pack. I had forgotten one of the factors that played a role in my decision to use 13.8V or sub 14.0V charging..

I drained my 400Ah series bank to about 70% SOC after charging it in series until the highest cell hit 3.6VPC. I then stopped charging and removed 120Ah's or approx 30% of the banks capacity.

Voltage levels were still pretty close, but not as perfect as I wanted.

I then removed the series jumpers and wired the cells in parallel and let them sit for a few days to self balance..

Next I set my bench top power supply for 20A (30A model) and 3.50V. I then began to charge the now parallel 1600AH 3.2V bank... It charged for a looooong time......

As the voltage neared 3.456 VPC it was still climbing extremely slowly, as in barely moving. I then looked over and saw it at 3.461V (a big change compared to how it had been behaving) and was surprised because I had not seen it move off of 3.452 in a long while. Unfortunately I missed exactly where the climbing started, in the thousandths of a volt range, but it was already going on at 3.456V....

At 3.456V the voltage started climbing very fast, a thousandths of a volt every few seconds.. In a few minutes (sadly I did not write it down) it had hit 3.5V and began voltage limiting and current began dropping. I then bumped the voltage up to 3.7VPC, sent it back into bulk, and it took what I recall as about 20 minutes to get to 3.7V.

It had taken approx 24 hours of constant bulk charging at 20A to go from a resting voltage of 3.336V to 3.456V..!

I then tweaked the voltage up to 3.8V as this was my top balance voltage. All told it took approx 30 minutes in total to go from 3.456V to 3.800 V. It then sat for a while at exactly 3.80V and current declined until I ceased top balance charging at 5A / 3.8VPC

This is one of the factors I used to determine my top on-board charging voltage of sub 14.0V charging.

The cells were essentially "full" at 3.456VPC or 13.824V as a 13.2V pack. Voltage began shooting up very, very quickly after 3.456V / 13.824V.. Keep in mind this was with only a 20A charge source on a 1600Ah bank at 3.XXX volts..

I decided to leave those last few Ah's between 3.450VPC / 13.8V and 3.6VPC / 14.4V on the table as a safety factor and decided on not pushing deeply into that upper knee.

Had not read my notes in a while and found that one interesting..

[goboatingnow]

Originally Posted by Maine Sail

Musings on my charging voltage:

I was going through some notes I took early late last year when I was doing the top balance to my pack. I had forgotten one of the factors that played a role in my decision to use 13.8V or sub 14.0V charging..

I drained my 400Ah series bank to about 70% SOC after charging it in series until the highest cell hit 3.6VPC. I then stopped charging and removed 120Ah's or approx 30% of the banks capacity.

Voltage levels were still pretty close, but not as perfect as I wanted.

I then removed the series jumpers and wired the cells in parallel and let them sit for a few days to self balance..

Next I set my bench top power supply for 20A (30A model) and 3.50V. I then began to charge the now parallel 1600AH 3.2V bank... It charged for a looooong time......

As the voltage neared 3.456 VPC it was still climbing extremely slowly, as in barely moving. I then looked over and saw it at 3.461V (a big change compared to how it had been behaving) and was surprised because I had not seen it move off of 3.452 in a long while. Unfortunately I missed exactly where the climbing started, in the thousandths of a volt range, but it was already going on at 3.456V....

At 3.456V the voltage started climbing very fast, a thousandths of a volt every few seconds.. In a few minutes (sadly I did not write it down) it had hit 3.5V and began voltage limiting and current began dropping. I then bumped the voltage up to 3.7VPC, sent it back into bulk, and it took what I recall as about 20 minutes to get to 3.7V.

It had taken approx 24 hours of constant bulk charging at 20A to go from a resting voltage of 3.336V to 3.456V..!

I then tweaked the voltage up to 3.8V as this was my top balance voltage. All told it took approx 30 minutes in total to go from 3.456V to 3.800 V. It then sat for a while at exactly 3.80V and current declined until I ceased top balance charging at 5A / 3.8VPC

This is one of the factors I used to determine my top on-board charging voltage of sub 14.0V charging.

The cells were essentially "full" at 3.456VPC or 13.824V as a 13.2V pack. Voltage began shooting up very, very quickly after 3.456V / 13.824V.. Keep in mind this was with only a 20A charge source on a 1600Ah bank at 3.XXX volts..

I decided to leave those last few Ah's between 3.450VPC / 13.8V and 3.6VPC / 14.4V on the table as a safety factor and decided on not pushing deeply into that upper knee.

Had not read my notes in a while and found that one interesting..

Very interesting

, I use a multi channel data logging multimeter to handle the lack of note taking , dump it into excel. Interesting observations , I have seen however some quite long tail charging at high Cs even with the voltage coming up fast as been necessarily to achieve capacity and it varies from manufacturer to manufacturer ( caveat cylindric LiFecells ) . In 2014 I will be free of an NDA around such cell research and I will be doing my own stuff and can publish the resulting graphs.

Dave

[Maine Sail]

Originally Posted by goboatingnow

Very interesting

, I use a multi channel data logging multimeter to handle the lack of note taking , dump it into excel. Interesting observations , I have seen however some quite long tail charging at high Cs even with the voltage coming up fast as been necessarily to achieve capacity and it varies from manufacturer to manufacturer ( caveat cylindric LiFecells ) . In 2014 I will be free of an NDA around such cell research and I will be doing my own stuff and can publish the resulting graphs.

Dave

Dave,

I current limit my alt to about 120A-130A or so and I have about a 35 +/- minute taper to 5A of acceptance at 13.8V... Compared to LA batteries this is AMAZING...... My first voltage target with the alt is 13.9V then it drops to 13.8V I then use a low float voltage, below resting, to essentially terminate alt charging.......

I have a Cell Log 8, a Pentametric and a regular battery monitor but I fall down in Excell and recording it, naming the files etc..... I need a good data logging MM..... Martin Lorton uses one that I really like but it is $$$$$$..

[goboatingnow]

Originally Posted by Maine Sail

Dave,

I current limit my alt to about 120A-130A or so and I have about a 35 +/- minute taper to 5A of acceptance at 13.8V... Compared to LA batteries this is AMAZING...... My first voltage target with the alt is 13.9V then it drops to 13.8V I then use a low float voltage, below resting, to essentially terminate alt charging.......

I have a Cell Log 8, a Pentametric and a regular battery monitor but I fall down in Excell and recording it, naming the files etc..... I need a good data logging MM..... Martin Lorton uses one that I really like but it is $$$$$$..

Have you seen the DataQ www.dataq.com , voltage loggers , the lower spec ones are very good value

Dave

[Blue Skye]

Question - Why are deckofficers cells not maintaining balance? It appears everyone else is reporting the cells voltage are staying very tightly balanced. Is it because they were not properly balanced or he is pushing them too hard? No offense intended by asking, just curious as to how this could happen.

And yes, I have read through 90% of this thread, including the old cells he received and the multiple attempts at trying to bottom and top balance them. As posted earlier by myself, thanks to all of you who have shared your experience and money in order to educate the rest of us

[diugo]

The correct peak charge voltage is really a function of the charge current, because of the "lift" superimposed on the cell terminal voltage by IR---the current times the internal resistance.

So when people say "I charge to 3.75 volts per cell", I always ask myself "but at what current?" And when spec sheets claim the max voltage is 4.0 volts per cell, that's only when charging at really high current, like C or even higher. This is done only to make the cells appealing to EVers who want to recharge as quickly as possible.

Not well stated is that at much lower charge currents, the peak voltage is way closer to 3.5---maybe even 3.45.

With my Balqon Specials I strive to keep them no higher than 80% SoC, in an experiment to see if doing so will result in several years of additional longevity. The only practical way to do this is by counting ampere-hours.

But when I do my annual Peukert tests (every January) and must charge to full capacity to do so, I go until the highest cell reaches 3.50+IR, and then discharge until the lowest cell reaches 2.80-IR.