Li-Ion: Should I top off the battery cells to 4.2v?

[tuxman]

BTW, I've tried to search for an answer before asking the forum.

My ten 3.7v cell li-ion 12ah battery pack is 41.7v full. But, I've read 4.2v per cell is the sweet spot. Should I adjust the charger to 42v? Should I bother with this slight indifference?

Will 41.7v max help to pro-long the life of the battery?

Also, when I get done with a short ride, down to 40.1v, should I top off the charge? At what voltage should I top off at?

 

[icecube57]

Id go with the 80-20% charge range. I think you will get more cycles in that range.

 

[fechter]

With lithium batteries, supposedly it's better if you don't top them off after a small discharge. The calendar life of the cells will be longer at lower voltages. Temperature probably plays a bigger role. They live longer at lower temperatures, but perform better at higher temperatures.

My guess is it won't matter much.

 

[neptronix]

For lipos, it is said that they will live longer, the closer you keep them to the middle voltage while being stored.

I am not sure if this applies to lifepo4/etc, but i would think that keeping it at mid-charge would be a good thing. No reason to top it off anyways.

 

[dogman dan]

Yes, but not to the point of obsessive compulsion about it. The generally accepted rule of thumb for most chemistries is an 80% discharge tends to result in less need for balancing than a 100% discharge. It's said that more cycles result, but I'm not clear whether you get more total watthours from a battery that way, or just get similar total watthours in the form of more short cycles.

Lead acid is the exeption, which likes less than 50% depth of discharge, and really needs immediate recharging if at all possible.

It won't hurt your lithium to go to the store several times before charging, over a period of a week or so, provided you aren't that close to 100% discharged. In practice, I never know for sure how long my next ride might be, so I often discharge 20% , and then just put it back on the charger. Then I know I can ride a long way next time if I want to. No reason for you do do this, if your next rides length is more predictable. Lots of folks do ride to work all week, and then charge friday if they have a short trip.

 

[tuxman]

Thanks guys for the quick replies. As you were replying, I was busy editing my post. You will find me doing this quite often! I double check for spelling and such, and make sure it's a good read for others.

What do you think about adjusting the charger to 42v?

And I will take the advice given, and charge up after 20% or more has been used.

 

[neptronix]

By the way, your batteries charge to 4.17v per cell for a reason.

There is very little energy above 4.1v per cell in the first place.
Don't bother charging it above 4.17v, that is just reducing their lifespan even more.

 

[tuxman]

Thanks for the advice!

 

[stanz]

BTW, I've tried to search for an answer before asking the forum.

My ten 3.7v cell li-ion 12ah battery pack is 41.7v full. But, I've read 4.2v per cell is the sweet spot. Should I adjust the charger to 42v? Should I bother with this slight indifference?

Will 41.7v max help to pro-long the life of the battery?

Also, when I get done with a short ride, down to 40.1v, should I top off the charge? At what voltage should I top off at?

What type Li-ion cells? That's very important. The military uses a variety of Lithium cell types. They extend the life of their Lico cells by only charging them to 3.95V instead of the 4.1V rated maximum.

Without knowing the cell chemistry I would not top them up after a short ride.

 

[tuxman]

It does not say on the out side of the battery pack. I will find out from the manufacture.

 

[DrkAngel]

Check individual banks of cells, make sure none are higher than 4.2v. Minimum recommended discharged voltage is 3.5v, under throttle "voltage sag" should not drop below 3v per cell-bank.

Storage voltage is variously recommended at 3.7v or 3.8v, per cell.

I charge to 4.2v, and get 8 miles before my pack drops to 4.1v.
Of course charging to only 4.17v will prolong lifespan, percentage unknown.
The question should be, "How important is an extra 4% range, to you?"

 

[neptronix]

I charge to 4.2v, and get 8 miles before my pack drops to 4.1v.
Of course charging to only 4.17v will prolong lifespan, percentage unknown.
The question should be, "How important is an extra 4% range, to you?"

Crazy, your cell chemistry must have a very flat discharge curve and your pack must have a ton of amp hours.
On my lipo, the difference between 4.15v and 4.1v is around 1 percent of my capacity.

I suppose it does vary from chemistry to chemistry.

There is a paper by nasa talking about lithium ion batteries and their cycle lives VS the voltage they were charged up to. It seems to apply to these cells which have a nominal voltage of 3.7 or 3.8v. The conclusion was that 3.9v gave many thousands of cycles, 4.2v i think provided cycles in the hundreds, 4.1v provided about a thousand cycles..

They had a brilliant graph.
For the life of me, i cannot find it, so here's what i got:

Capacity fade of various voltages:

http://www.che.sc.edu/faculty/popov/Publications/Premanand1.pdf

Here is a great thread on ES.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=15552

Shallower discharges seem to extend cycle life a lot. By how much... seems to depend on the chemistry.
That being said i am charging my lipos to 4.1 instead of 4.15 now. I am shooting for 1000 cycles.

 

[DrkAngel]

I charge to 4.2v, and get 8 miles before my pack drops to 4.1v.
Of course charging to only 4.17v will prolong lifespan, percentage unknown.
The question should be, "How important is an extra 4% range, to you?"

Crazy, your cell chemistry must have a very flat discharge curve and your pack must have a ton of amp hours.
On my lipo, the difference between 4.15v and 4.1v is around 1 percent of my capacity.

I suppose it does vary from chemistry to chemistry.

Shallower discharges seem to extend cycle life a lot. By how much... seems to depend on the chemistry.
That being said i am charging my lipos to 4.1 instead of 4.15 now. I am shooting for 1000 cycles.

Depth of discharge is the major determining factor in"usable cycles".

And, as you can see, in the graph that you previously posted, Li-ions do have a very steady discharge curve, right from 4.2v down to 3.5v!

 

[neptronix]

^---a123

^---turnigy lipo

^---thunder power lipo

^--- EIG li-mn

As you can see, there is a pretty wide variance in discharge curves for lithium batteries. Some flat, some pretty peaky.

Is there such a thing as unusable cycles? :lol:
If a battery's life doubles by charging it to 4.1v versus 4.2v, that's pretty damn good.

 

[DrkAngel]

As you can see, there is a pretty wide variance in discharge curves for lithium batteries. Some flat, some pretty peaky.

All the Li-ion graphs look consistent, with pre-load voltage of 4.2v and end of, recommended, usable at about 3.5v!

The 1st graph "A123" is obviously Lithium Iron, not Lithium-ion!
They have completely different chemistry, energy density and voltages and is invalid in the comparison you try to present!

Similar to dropping a ni-cad (1.2v) chart amongst an Alkaline (1.5v) comparison.

 

[bobc]

Lots of nice graphs - the differences 'may' be explained by how they are measured. I did a data log of a turnigy lipo discharge and noted that the "no load" cell voltage reduced linearly as the battery discharged whereas, because the cell series resistance seems to reduce as the cell discharges, the "full load" voltage was fairly fixed for an awful lot of the discharge time. here is the data log graph:

You can see the general voltage level during power-on and power-off.
That's real data captured by me ;^) pink/purple trace is amps, dark blue is volts (6s) and green is speed
the battery is a 2200mAh "flightmax" rated at 20C - being discharged at that + a bit

 

[tuxman]

You guys are great! Thanks for the charts.

It just now clicked; my 37v battery pack is 41.7v to allow for voltage sag under load. If normal voltage sag is 3-4v, would charging to 40v increase the lifespan of the pack? And still maintain 36-37v under load to the motor?

To find out my voltage sag, I can wire an xlr4 plug onto my volt meter and plug it into the charging port on my battery pack. I could note the voltage measurements under load while going up the hills, at the start, in the middle, and at the end of my commute.

 

[DrkAngel]

As you can see, there is a pretty wide variance in discharge curves for lithium batteries. Some flat, some pretty peaky.

All the Li-ion graphs look consistent, with pre-load voltage of 4.2v and end of, recommended, usable at about 3.5v!

The 1st graph "A123" is obviously Lithium Iron, not Lithium-ion!
They have completely different chemistry, energy density and voltages and is invalid in the comparison you try to present!

Similar to dropping a ni-cad (1.2v) chart amongst an Alkaline (1.5v) comparison.

Is there such a thing as unusable cycles? :lol:

"Usable cycles", is generally rated as, the point where the battery still retains 80%, sometimes 85%, of it's original output capacity.

If a battery's life doubles by charging it to 4.1v versus 4.2v, that's pretty damn good.

Do you have any documentation for that premise?


The determinant factor for Lithium Ion, Lithium Ion Polymer also, best ways to damage, cycle life - health, is:
#1. severe depth of discharge (below 3.5v - (3v is considered, an actual, "Danger" point!))
#2. overcharging - but, most chargers & PCBs have over voltage protection at 4.25v, or somewhat higher, to prevent this!

#3. high rate of discharge **
#4. high rate of charge **

** #3 & #4 are more applicable to Lipo, which, typically, lack the, built in, safety, which regulates the charge-discharge rate, found in "standard" Li-ion.
Foe example: 500 cycle rated Lipo might provide 50 usable cycles in the, high rate, RC aircraft arena.

 

[neptronix]

As you can see, there is a pretty wide variance in discharge curves for lithium batteries. Some flat, some pretty peaky.

All the Li-ion graphs look consistent, with pre-load voltage of 4.2v and end of, recommended, usable at about 3.5v!

The 1st graph "A123" is obviously Lithium Iron, not Lithium-ion!
They have completely different chemistry, energy density and voltages and is invalid in the comparison you try to present!

Similar to dropping a ni-cad (1.2v) chart amongst an Alkaline (1.5v) comparison.

Man, did you get a degree in contrarianism?

1) We weren't even talking about the beginning and end voltage ranges, we were talking about the difference of state of charge from 4.1 to 4.2v per cell..
2) If these graphs look consistent to you, you have serious problems with your vision. I suppose i should have not tried to show you data since others have failed at making a point by showing you data as well.
3) Lithium iron is a form of a lithium ion chemistry, so is lithium cobalt, so is lithium manganese, so is lithium polymer ( which is technically referred to as 'lithium ion polymer'....
4) I wasn't talking about overall energy density. I never brought it up.
5) If you went to contrarian school then you missed one thing; what was the C rate for each discharge graph?

I can tell that you are not actually listening.

 

[DrkAngel]

Lithium iron is a form of a lithium ion chemistry, so is lithium cobalt, so is lithium manganese, so is lithium polymer

Good luck trying to charge a Lithium Iron pack with a Lithium ion charger!

 

[neptronix]

If a battery's life doubles by charging it to 4.1v versus 4.2v, that's pretty damn good.

Do you have any documentation for that premise?

The determinant factor for Lithium Ion, Lithium Ion Polymer also, best ways to damage, cycle life - health, is:
#1. severe depth of discharge (below 3.5v - (3v is considered, an actual, "Danger" point!))
#2. overcharging - but, most chargers & PCBs have over voltage protection at 4.25v, or somewhat higher, to prevent this!

#3. high rate of discharge **
#4. high rate of charge **

** #3 & #4 are more applicable to Lipo, which, typically, lack the, built in, safety, which regulates the charge-discharge rate, found in "standard" Li-ion.
Foe example: 500 cycle rated Lipo might provide 50 usable cycles in the, high rate, RC aircraft arena.

I provided some links in this thread, did you not read them?
Points #1 - #4 are completely valid, but you are missing point #5. Re-read the links i posted.

#3 - #4 apply to all lithium chemistries, even lead acid, and probably the Nickel types too. High internal resistance creates heat, and cells ran over their C rating will produce heat since the battery essentially becomes a heating element. Heat destroys things.

There is a reason why all cell spec sheets that list cycle life list it at a 1C ( or below ) rate.
They never list what happens at 2C, even on a cell that is capable of producing 2C ( or much much higher )

I wish i can find that nasa study, they showed that lithium cobalt ( non-polymer ) could have it's life extended like crazy if you charged it to 3.9v/cell.. i saw it linked from here. I will start bookmarking stuff like this from now on...

 

[neptronix]

Lithium iron is a form of a lithium ion chemistry, so is lithium cobalt, so is lithium manganese, so is lithium polymer

Good luck trying to charge a Lithium Iron pack with a Lithium ion charger!

If you really believe that, you obviously have no idea how lithium chargers work. If you are just trying to play contrarian again, then.. how quaint :lol:

It's kinda sad that you came here from a forum where you were the smartest guy on it.
I hope you take a chance to learn some new things, ES has a lot to offer.

 

[neptronix]

Lots of nice graphs - the differences 'may' be explained by how they are measured. I did a data log of a turnigy lipo discharge and noted that the "no load" cell voltage reduced linearly as the battery discharged whereas, because the cell series resistance seems to reduce as the cell discharges, the "full load" voltage was fairly fixed for an awful lot of the discharge time. here is the data log graph:

Yeah, there is definitely some difference in how they were measured. The C rate will affect the curve quite a bit.

I suppose i was just trying to illustrate a point at how the difference of 4.1-4.2v varies on each battery really.

Your curve looks pretty damn flat, way flatter than what i have. I got these results with a 1.8C load on a turnigy 20C 5ah battery. What are you using there, nanotech?

http://neptronix.org/forumpics/lipotutorial/24_dischargingmechanics.gif

I suppose my graph is a bit flawed. I should re-do it as applying the 1.8C load did result in a big drop from 4.15v to 4.10v, which did not adequately show the small difference between 4.15 and 4.1v..

 

[DrkAngel]

I wish i can find that nasa study, they showed that lithium cobalt ( non-polymer ) could have it's life extended like crazy if you charged it to 3.9v/cell.. i saw it linked from here. I will start bookmarking stuff like this from now on...

Of course, if you only charge a Li-ion pack to 3.9v, and discharge to 3.6v. the battery will last many more cycles than the same battery charged at the recommended 4.2v and discharged to 3.5v!

Unfortunately, you would only get 43% of the energy output!
You would need a battery almost 2 1/2 times the size, to get the same output as the recommended voltages.
.......................................................... ? .................................
I guess I just don't appreciate the logic of paying 2 1/2 times as much, and having to carry around 2 1/2 times the weight to get the same amount of electric capacity?

That sounds like government bureaucracy logic!

 

[neptronix]

Of course, if you only charge a Li-ion pack to 3.9v, and discharge to 3.6v. the battery will last many more cycles than the same battery charged at the recommended 4.2v and discharged to 3.5v!

Unfortunately, you would only get 43% of the energy output!
You would need a battery almost 2 1/2 times the size, to get the same output as the recommended voltages.
.......................................................... ? .................................
I guess I just don't appreciate the logic of paying 2 1/2 times as much, and having to carry around 2 1/2 times the weight to get the same amount of electric capacity?

That sounds like government bureaucracy logic!

LOL, you are all over the place drkangel :lol:
I never recommended that you discharge to 3.9v, i just used it as an example, and i used 4.1v as an example of how on most chemistries you can double your cell life by charging to that voltage instead of 4.2v.

According to the data you collected, however you did it, you got 4% as the difference between 4.2 and 4.1v per cell.
Double the life, at the expense of 4%...

On my lipo batteries, at 3.9v resting you have about 65% capacity left.
It all depends on the discharge curve characteristics of your battery. But on every battery graph i have seen, a very minimal amount is stored at 4.2v VS 4.1 and even 4.0. I think turnigy nanotech is the exception, however i have yet to see a proper discharge graph for turnigy nanotech.