Bottom balancing?

I see about 3-10mv difference in the middle 75%.

At the top, i've seen 10-15mv difference over dozens of unbalanced charges now.
Below 3.6v, on good packs, it's usually 10-40mv, increasing to 300mv down into the 3.0v zone.

Now, a really good diagnostic is logging a charge and discharge with a celllog 8s. You can hook that puppy into USB and run it with logview to get the same graphs i produce with my iCharger 1010b+; pretty cool.

If a cell is constantly getting knocked out of balance, there surely is a reason for it.. either low capacity or a lower or higher internal resistance..

neptronix said:

I see about 3-10mv difference in the middle 75%.

At the top, i've seen 10-15mv difference over dozens of unbalanced charges now.
Below 3.6v, on good packs, it's usually 10-40mv, increasing to 300mv down into the 3.0v zone.

Now, a really good diagnostic is logging a charge and discharge with a celllog 8s. You can hook that puppy into USB and run it with logview to get the same graphs i produce with my iCharger 1010b+; pretty cool.

If a cell is constantly getting knocked out of balance, there surely is a reason for it.. either low capacity or a lower or higher internal resistance..

Click to expand...

Are these #s from when the pack is at rest of under load? If at rest then my packs are doing at least this good as well. I have only actually balanced my packs once in thier cell life and to me they really didn't need it. But then again we all know what a perfectly ballanced pack is but what exactly is should be considered an "out of balance" pack that we should be concerned about?

Bob

My take on it is, balance is not important unless you hit low voltage disconnect. Each cell has its own capacity, and forcing all cells to the same voltage does little good and probably some harm. If you can get 8 or 9 amp hours out of each 10 amp hour cell in the string, it doesn't matter if that comes from the top or bottom end of the capacity, and staying away from the limit at either end will reduce the gradual loss of capacity.

If you need absolute maximum capacity then top balance, but consider adding another cell to get that capacity without reducing pack life. If one cell eventually hits the LV disconnect at the end of your ride, recharge the pack normally and then add more charge to the low cell to bring it back into midrange. Top balancing will do that conveniently if the shunts can handle the current, but keeping all the other cells at max voltage just to bring up one low cell has got to reduce total lifetime.

My take on it is, balance is not important unless you hit low voltage disconnect

Click to expand...

I tend to agree. Although, I've never run a pack for a year with out balancing. It'd be interesting to see the drift in a situation where lvc/hvc are used but balancing is not.

At the top, i've seen 10-15mv difference over dozens of unbalanced charges now.

Click to expand...

I just finished top balancing my ping and their a 10mV difference between the lowest and highest cell after a night of rest.

Below 3.6v, on good packs, it's usually 10-40mv, increasing to 300mv down into the 3.0v zone.

Now, a really good diagnostic is logging a charge and discharge with a celllog 8s. You can hook that puppy into USB and run it with logview to get the same graphs i produce with my iCharger 1010b+; pretty cool.

Click to expand...

Yeah... until I get a second cell-log (I blew it up), I don't think I'm going to take it to LVC. Too much risk. I'll just do my normal 16 mile commute at 1c while logging each 8s string being monitored (twice.) Then I'll do it at 2c twice.
It kinda sucks cause lifepo4s discharge curve is so flat. All the cells are within 3mV in middle %80. You can see more deviation under loads though.

I suppose bottom balancing it and then charging up would be a safe way to see capacity though.

Keep in mind that what you really want to balance is cell capacity, not necessarily voltage. In the middle part of the charge/discharge curve, they will tend to be very closely matched, but as the cells reach full (or empty), the variations get larger. Just looking at the voltages may not tell you much about the capacity.

I barely see 0.01v difference.

fechter said:

Keep in mind that what you really want to balance is cell capacity, not necessarily voltage. In the middle part of the charge/discharge curve, they will tend to be very closely matched, but as the cells reach full (or empty), the variations get larger. Just looking at the voltages may not tell you much about the capacity.

Click to expand...

This is my experieince exactly. I have one battery that has a cell that has slightly less capacity then the 5 other cells in the pack. So I just keep that one cell at 4.15 and the rest at 4.10. This seems to keep all the capacities within an acceptable variance, as they all are equal by the time they get to lvc. When I was trying to top balance all 6 to 4.10, that one cell would arrive at the cliff much sooner than the rest and risked damage.

Top vs BOTTOM balancing.. that famous debate 8)

I found a guy ( from ElithionTV ) on youtube that explain the difefrence and it seem he is not on teh same side as Richard from EVTV :lol:

but i like the way he explain.

Major advantage of top balancing:

- total energy delivered bt the battery is higher
- voltage sag is lower due to lower internal resistance at higher soc

http://www.youtube.com/user/ElithionTV

EVTV vs ElithionTV yeah baby! :wink:

Doc

Doctorbass said:

EVTV vs ElithionTV yeah baby! :wink:

Click to expand...

I would like to see ol Jack Rickard and Luke put together Jack
had such high opions of Luke and his battery knowledge
i think it would be an excellent show, 30 seconds in Luke chokes out Jack
more action than we see in the Octogon on UFC

KiM

I am not a fan of having BMS on board all the time either and I don't try to bottom-balance either. I carry no LVC cut-off and my HVC is my eyes and a cell-log.

What I do is just pull 75%, 80% out of the pack, and then quit.

I then plug a Cell-Log into an 8s Balance tap which is on each string and bulk each 24v {8s} string with a Meanwell. I usually start off around 27v and then turn her up to 28v as the A's drop off.

There is a Watts-Up between the Meanwell and the +/- terminals of the meanwell and that gives me another focal point for monitoring.

I just watch the cells fill up on the cell-log and watch the amperage fall off on the Watts-Up. Ideally I should have strings charged fully to 28.0v with no more than 50-60mv between the cells.

As soon as the difference between the cells in the string approaches 50mv or over, I swap over onto the Turnigy 8s balancer and finish the string with a 0.5A balance charge.

As one string is balancing, the other is bulking, and so-forth.

I would like to see ol Jack Rickard and Luke put together Jack
had such high opions of Luke and his battery knowledge
i think it would be an excellent show, 30 seconds in Luke chokes out Jack
more action than we see in the Octogon on UFC

Click to expand...

thanks for that laugh HI-larious

I carry no LVC cut-off and my HVC is my eyes and a cell-log.

Click to expand...

Neither do I.

I have to say, the hvc issue, or lack of them has got me loving LIFE. Since the SOC difference between 3.5v and 3.8v is negligible, I can go ages between balances. Like literally 2-3 or three months of daily use. It's like, why even bother? Something has to go VERY VERY wrong for me to worry about overcharging, and I have rarely ever go below 50% SOC so no worries there either. Very non-temperamental batteries there. Totally great for noobs to battery management.

Honestly, I find it funny that it's never been mentioned in the endless life vs. lipo debate. The difference is SOC between 4v and 4.2v for lipo is what? 10%-20%? Honest question there. And the margin of safety is so much smaller. Most charge to 4.15v - only a .05v room for error. If you charge Life to 3.5v, you have like .4v margin and even that isn't really harmful.

Don't meet to start a debate. Totally got some lipo packs on the way. Just thought it was interesting.

auraslip said:

I carry no LVC cut-off and my HVC is my eyes and a cell-log.

Click to expand...

Neither do I.

I have to say, the hvc issue, or lack of them has got me loving LIFE. Since the SOC difference between 3.5v and 3.8v is negligible, I can go ages between balances. Like literally 2-3 or three months of daily use. It's like, why even bother? Something has to go VERY VERY wrong for me to worry about overcharging, and I have rarely ever go below 50% SOC so no worries there either. Very non-temperamental batteries there. Totally great for noobs to battery management.

Honestly, I find it funny that it's never been mentioned in the endless life vs. lipo debate. The difference is SOC between 4v and 4.2v for lipo is what? 10%-20%? Honest question there. And the margin of safety is so much smaller. Most charge to 4.15v - only a .05v room for error. If you charge Life to 3.5v, you have like .4v margin and even that isn't really harmful.

Don't meet to start a debate. Totally got some lipo packs on the way. Just thought it was interesting.

Click to expand...

I use the A123 cell which is a great cell, the cells all top back up the same way and are easy to bulk charge. If you have built a half-decent pack then they should all be pretty evenly discharged in the first place. I can see the logic of a bottom-balance but if it isn't necessary then why bother? I just bulk to 28v {8s} and even if I don't balance they effectively settle down into the same range once I burn off the surface voltage. Great cell.

auraslip said:

Most charge to 4.15v - only a .05v room for error.

Click to expand...

4.2v is no magic number. I've been to much more without serious damage. My crash test dummy 20c Zippy actually got to 4.9v before the fireball came. People should use less than 4.2v to get longer pack life, not because an unbalanced cell will blow up at 4.25 or 4.3v. Can't speak of the old school lipos, but these newer RC cells are very tough and I believe there are many who take the risk and deliberately overcharge them by 10-20% at races to get max. energy density, although it involves having to use Pb or NiMH charge mode on their RC chargers.

Sorry to wake up this thread, but there something I don't understand on curves posted on the first message of this thread.

Why the Y axis is SoC and not Time ? How is determined the 0% SoC of the pack ? It's seems to be on the global pack voltage to have approx 2.1V / cell ?

But more important question, with top or bottom balacing, there is alway one cell which will be used in it's full HVC/LVC range, it's the weakest cell, IE the cell with the lower capacity :

- After a top balanced charge, the start discharge voltage is 3.65V and the true 0% SoC of the pack is when the weakest cell reach LVC, 2.1V in the present experiment.

- After a bottom balanced charge, the start discharge voltage is also 3.65V because it's this cell which triggered first the HVC and stopped the charge, and this cell reach the 2.1V LVC at the same time than other cell.

In the first 2 curves, we can see than the weakest cell is the cell number 4 :

- On the first curve (discharge after a top balanced charge), we can see that this cell start form 3.65V and reach the 2.1V LVC before other --> this cell has the lower capacity

- On the second curve (discharge after a bottom balanced charge), we can see that only the cell 4 has its voltage at 3.65V, it's logical because it's this cell which firstly reached the HVC which stopped the charge --> this cell has the lower capacity.

This demonstration is needed to see that the full cell 4 capacity is alway during discharge in the 2 charge configurations (bottom/top balanced), it means in the 2 charge configurations this cell is alway used in this full HVC/LVC voltage range during discharge.

Thus is anyone can explain me how cell 4 curve can be different between the 2 curves in the HVC/LVC range if there is the same discharge current ? For me there is a mistake somewhere, it's not possible because in the 2 configuration the start SoC of this cell is the same (100%).

For the top/bottom balancing debate itself, I think the only useful reason to do a bottom balancing is to don't use a BMS during discharge and only cutoff discharge on global pack voltage, right ?

So these guys on EVTV just bottom balance their big LiFe prismatic packs so that all the cells are 2.75v and then rarely ever have to balance again by choosing a bulk hvc based on an average cell hvc on the low side of 3.65v. Some cells will always bulk charge up to 4.1v while others will be at 3.5v but they always hit the bottom in balance.
.
http://www.evtv.me/vidarch.html

I have serious doubts that those guys on evtv know much about batteries.

merged sendler2112's thread into the already-existing identically-named thread about the same thing (which could have easily been found in a quick search ).

Search on thread title of "Bottom balancing":
http://www.endless-sphere.com/forums/search.php?keywords=%22Bottom+Balancing%22&terms=all&author=&sc=1&sf=all&sk=t&sd=d&sr=topics&st=0&ch=300&t=0&submit=Search
brings up other relevant threads having discussed all this before.

We've done this EVTV rubbish to death here, several times. The bloke that heads up that show is technically naive (that's being polite about it) and clearly has no real understanding of how the cells he is using work and why SoC can only be determined accurately when all cells are charged to a set charge cut off voltage.

Let's not let his ignorance cause people to damage cells by following his advice, when there is a very much larger body of evidence to show that there are far better ways of assuring pack safety and long life.

amberwolf said:

(which could have easily been found in a quick search ).

Click to expand...

Sorry. I did search. "bottom balancing". And it did not come up. Good information. Thanks for finding it for me. I will have to remember to go out to Google for searching Endless-sphere where it would have probably come right up.

Bottom Balancing? (Changed the title ... to clarify.)
I refer to "Bottom Balancing" as balancing at the optimal DOD!
A nasty point of contention for many.
But, please let me present my viewpoint.
I would consider bottom balancing to be, equalizing cells at the lowest recommendable discharge point. I create capacity maps of cells and, using these, determine the precise regions of greatest usable energy density.


Bottom balancing would be the preferred method for these 18650 cells.
The majority of stored capacity is near 3.7V, I would "bottom balance" at 3.65 to take full advantage of this "capacity bulge". 3.65V seems to be the optimal shelf, just before the capacity plummets.
Since these cells have a good capacity right to 4.20V, I would "bulk charge" to approximately 4.10V per bank, allowing a large 4.00V - 4.20V leeway for peak charge on individual banks.

Top balancing would allow one weak bank to drop off this cliff or, prevent the rest of the banks from delving into the capacity rich portion of the cells.

3.7V 2600mAh 18650 LiCo cells

Also bottom balancing would easily highlight any weak banks prompting repair or simply adding cell-cells to weak banks.

Of course you would want to map you cells capacity, in order to determine a "capacity map" or graph, similar to above.
I used - mAh/100th V - discharge test

Alternate-easier? method

Works best with Li-ion Lipo (LiCo) cells
If you have a watt meter or Ah meter -
begin with discharged cell or pack
Charge till meter indicates 100mAh (or 1wh)
stop charge
pause several seconds (10?)
record voltage
repeat
create map of capacity

You don't want to take any cells off the cliff.

Bottom balancing would require waiting for 1 cell to go off the cliff then discharging the other cells until they also go off the cliff and match the lowest voltage cell.

So now you have taken the whole pack off the cliff into cell damage zone.

That is the main problem with bottom balancing.

I discharge the cells before I put them together in a pack.

That is the only time I bottom balance.

Ya'll can run your cells at any voltages you want to. They are your cells.

But if you want them to last stay away from the bottom.

etriker said:

Ya'll can run your cells at any voltages you want to. They are your cells.

But if you want them to last stay away from the bottom.

Click to expand...

I would rate the "cliff" at just above 3.6V, so I balance at 3.65V where there seems to be a solid shoulder of stable capacity.

The 3.65V point is where my empty would be set.
Taking full advantage of the 3.70V capacity "bulge".
Of course, I never intend on reaching "empty" ...
With 30-40Ah packs ... I don't need to.

Unless you balance off the cliff or when the cells spike up then very small differences in voltage will equal quite a bit of difference in capacity.

The accuracy of the balancer then comes into play big time.

If it is off just a few thousands and the pack is balanced not on the upper spike or lower cliff then the balancer can throw the pack way out of balance.