Ways to re-charge a lipo battery after it drops below 2.7v

[RedLine19K]

Ok but, in a thread about safe limits, 4.3v statements muddy the waters. As the other thread suggests, every +.1v halves the cycle life. Lipos purposely only have .985 ratio to leave the slightest protection against lithium plating, which driving to 4.3v surely does. And for what? a surface charge that lasts less than a second, and far less than 1% capacity, while reducing cycle life by 10x. But that's all just speculation so... how bout some 4.35 discharge graphs?

For racing, it might be worth it. Some people use a pack for only 1 race. But it's a disservice to post "Oooh, 4.3=MOAR POWER" without detailing the risks. Much in the same way saying 2.7v is a minimum without noting the variance in chemistry, current, components and conditions that make it safe.

I guess what I'm saying is if anyone posts general absolutes, they should be based on the most conservative and common conditions. Knowledgeable experts can push the limits and share the results in detail.

 

[liveforphysics]

Ok but, in a thread about safe limits, 4.3v statements muddy the waters. As the other thread suggests, every +.1v halves the cycle life. Lipos purposely only have .985 ratio to leave the slightest protection against lithium plating, which driving to 4.3v surely does. And for what? a surface charge that lasts less than a second, and far less than 1% capacity, while reducing cycle life by 10x. But that's all just speculation so... how bout some 4.35 discharge graphs?

For racing, it might be worth it. Some people use a pack for only 1 race. But it's a disservice to post "Oooh, 4.3=MOAR POWER" without detailing the risks. Much in the same way saying 2.7v is a minimum without noting the variance in chemistry, current, components and conditions that make it safe.

I guess what I'm saying is if anyone posts general absolutes, they should be based on the most conservative and common conditions. Knowledgeable experts can push the limits and share the results in detail.

On the 'compact' lipo's they have less margin in anode capacity to accept overcharge. On a nano-tech (at least the 45-90C formula), you can insert >6Ah into the 5Ah pretty easily for 20% more energy storage, which can make the difference between completing a race or DNF'ing. However, yes it comes at the cost of doing real permanent damage to the cells, so it's obviously not something you want to do with a pack that you're not willing to sacrifice for an event/contest/record etc. It's also obviously got more dangers involved as well.

Not something you want to do for a pack you're expecting to last more than a few dozen cycles, and not something you want to do if you don't have the skills to monitor properly, and not something you want to do if you're not willing to accept it bursting into flames, and not something you would ever want to do and then store the battery for any period of time, as it's breaking down your electrolyte into gasses each second it's sitting at those potentials.

Basically, it's not something you ever want to do for any reason unless you're an insane EV racer willing to gamble the whole vehicle and battery to give yourself a slight edge in a race. However, at least a small handful of members do fall into that category on occasion, which is the only reason I mention it.

It's the choice of the individual person to decide how much risk vs battery weight savings they want to have, but IMHO, noobs shouldn't be using RC lipo for commuter vehicles anyways, it's uniquely awesome power density advantages are simply not required for low power commuter type applications.

4.1-4.15v to 3.5-3.6v is the window to operate if you want the packs to last a bit, and you still get ~80% of the useful capacity.

 

[RedLine19K]

Now that is interesting.. 20% more capacity is surprising. LiFePO4 will spike from 3.6 to 4 while adding no real capacity, I thought most would behave the same. I've seen some high temp tests from NASA, but most overvoltage deals with failure, not performance increases.

I've often wondered in the quest for ever higher imaginary C ratings, why not just derate the capacity. A 6ah @ 20C magically becomes a 5ah @ 24. But i'm sure the internals are more like deep cycle vs starter for lead acid.

The Zippy compacts happen to be my favorite batteries, but I never push them too hard. The extra energy density means more capaciy forthe same weight=even lower efffective C rate. I never bothered with the nano 45-90s because the reports of puffing are so common.

Yep, if 1C(3C) drain meets the needs (who doesn't want an EV to at least last an hour?) then LiFePO4 is a safer way to go, or even 18650s. I've used LiPO in RC for 10 years and only had one mishap. Of course, it was the first battery I ever had. Learning curve looks a lot like the lipo cliff....pretty steep.

 

[Ykick]

4.1-4.15v to 3.5-3.6v is the window to operate if you want the packs to last a bit, and you still get ~80% of the useful capacity.

This EXCELLENT advice has served me well beyond 3 years almost daily cycle use. Thank you Luke!

2.7V/cell is basically impossible to observe in REAL WORLD application IMO and it would appear to be especially bad for RC Lipo anyway.

Only "wisdom" I can share is that RC Lipo chemistry ain't for everybody. If you can't make the commitment to understand it and treat it properly - save yourself a lot of grief and danger, get out while you still can...

 

[omegagamer89]

I do want to thank everyone for their help in this thread. I've since gotten everything set right, and took the bike out for another test ride today. This time I rode till about 4.02v per cell, and then turned around. By the time I got home, I was down to about 3.82v, so this gives me a much better idea of the distance I can realistically go without running down the battery.

That being said, that distance is a bit shorter than I was hoping for. I realize I could have gone a bit farther before turning around, and still not harmed the batteries, but Im hoping to go a lot farther.

So, I have a new question. Im fairly certain I already know the answer, but I figure it cant hurt to double check it with you guys.

As I said earlier, I have 3 of the turnigy 5000mah hardcase packs in series to serve as my battery pack. I know that if I were to get 3 more batteries, make a second, idential pack, and wire THAT in parallel with the first pack, it would double the amps the pack is putting out.

However, that wouldnt double the amp HOURS. Correct? Basically, I would get the same distance out of the pack, but I would get more torque/more powerful acceleration.

Is this about right?

I realize that if I were to make a second pack, identical to the first one, I could always just carry them both and swap one out for the other when it gets low, but it's more convenient to not have to swap battery packs in the middle of a ride, yknow?
Also, if Im correct and wiring two pack in parallel wont result in longer battery life, what are some things I can do that WILL result in longer battery life and longer rides? (Aside from getting better batteries)

 

[wesnewell]

If you parallel in another 5ah pack, then your pack would be 10ah and it would double your range. It would also be less stress on the battery pack. A 5ah 12s 20C pack is rated for 100A. A 10ah 12s 20C pack is rated for 200A. You would also enjoy less battery sag, so a little more power under heavy load.

 

[omegagamer89]

If you parallel in another 5ah pack, then your pack would be 10ah and it would double your range. It would also be less stress on the battery pack. A 5ah 12s 20C pack is rated for 100A. A 10ah 12s 20C pack is rated for 200A. You would also enjoy less battery sag, so a little more power under heavy load.

I was actually just about to come back and edit my post. XD I did some research, and from what I saw it would appear I was wrong, and wiring the two individual pack in parallel will increase the amp hours. Thank you, wesnewell, for confirming what I thought.

Im going to do a bit more research on a few things, but I think I might just order 3 more batteries and set up 2 packs in parellel. Thanks again for the help, everyone.

 

[dogman dan]

Keep asking questions even if you think you know the answer, then sort out which of us has the opinion you trust most. Learn all you can.

You don't have to follow all of any persons advice or opinion. Use your brain to sort it out, and do what you feel makes sense to you, for your house, for your bike. Part of the joy of using a home assembled battery is you don't get limited by what others think. You can do what you want with discharge and charging. Bend the rules for a race, or practice a more conservative method than a store bought bms does. Your choice.

Even if your controller is a moderate 20 amps one, it will help to carry 10 ah instead of 5. It will sag less, giving you a better idea as you ride of what your voltage is at rest. It will have a lower discharge rate, which all batteries however powerful will like. And on the shorter rides, the lower depth of discharge may also help extend lifespan, or at least help keep the pack more balanced.

I like to parallel first then make the series connection. But do what fits your needs best. In my case I have 15 ah of 48v, made from 20c lico from hobby king. I maximized my flexibility by having it in two packs. One is 5 ah for short rides where I want light weight. The other is 10 ah for longer rides. And I can carry all 15 ah when I need to.

You don't have to stop at 3.8v if you aren't home. But if you are not a block or two away, you need to slow way down and mostly pedal home once you get below 3.75 or so. Once you pass 3.5v, you literally have only about a block left.

 

[omegagamer89]

Thanks again for all the info and advice.

I do have one question; is there any particular reason why you have yours paralleled, and THEN in series, vs the other way around?
What I mean is, is there actually some benefit to having, say....
3 'packs' each made of 2 5000mah units paralleled, and then having those 3 'packs' wired in series
VS
2 'packs' each made of 3 5000mah units in series, and then having those 2 'packs' paralleled?

Is there any advantage to one arrangement over the other, or is it just down to preference?

 

[Ykick]

Thanks again for all the info and advice.

I do have one question; is there any particular reason why you have yours paralleled, and THEN in series, vs the other way around?
What I mean is, is there actually some benefit to having, say....
3 'packs' each made of 2 5000mah units paralleled, and then having those 3 'packs' wired in series
VS
2 'packs' each made of 3 5000mah units in series, and then having those 2 'packs' paralleled?

Is there any advantage to one arrangement over the other, or is it just down to preference?

Good luck finding full agreement about this issue. I don't parallel before series wiring for 2 reasons:

1) I sometimes change my pack capacity configuration from time to time which is easier without parallel balance connectors either permanent or through paraboards.

2) I also have fear of not being able to detect dying cells because they're paralleled to other cells. My way is checking individual cells via CellLog 8S logging function and/or simple IR load testing and keeping a record of cell IR. Parallel cell connections would seem to skew those results.

Some will argue that parallel wiring cells will allow weaker cells to be "assisted" by stronger cells. However, a user just recently reported how a shorted cell took out 2-3 good cells paralleled with the bad cell.

 

[omegagamer89]

Good luck finding full agreement about this issue. I don't parallel before series wiring for 2 reasons:

1) I sometimes change my pack capacity configuration from time to time which is easier without parallel balance connectors either permanent or through paraboards.

2) I also have fear of not being able to detect dying cells because they're paralleled to other cells. My way is checking individual cells via CellLog 8S logging function and/or simple IR load testing and keeping a record of cell IR. Parallel cell connections would seem to skew those results.

Some will argue that parallel wiring cells will allow weaker cells to be "assisted" by stronger cells. However, a user just recently reported how a shorted cell took out 2-3 good cells paralleled with the bad cell.

Mmm... Thats a good point to consider. After some looking around, I think at this point Im gonna go with making 2 'packs' of 3 units in series each, and then wire those 2 'packs' in parallel, partly because that will make it easier to add the additonal batteries onto my current setup.

 

[Ykick]

Mmm... Thats a good point to consider. After some looking around, I think at this point Im gonna go with making 2 'packs' of 3 units in series each, and then wire those 2 'packs' in parallel, partly because that will make it easier to add the additonal batteries onto my current setup.

If a cell shorts in this configuration the other series string in parallel can obviously overcharge the series string with a shorted cell. That's the risk of series>parallel connections. But, the "insurance" is to not waste the easy and clear ability to continually test individual cell condition.

 

[RedLine19K]

Would something like this prevent a low pack from pulling the other down?
http://www.ebay.com/itm/310578003621
I think the drop is only .3-.4v so a 1% hit for protection isn't too bad? Any EEs want to explain the yes/no?

 

[fechter]

Would something like this prevent a low pack from pulling the other down?
http://www.ebay.com/itm/310578003621
I think the drop is only .3-.4v so a 1% hit for protection isn't too bad? Any EEs want to explain the yes/no?

Nice diodes!

Under high load, the voltage drop will be more. Up to 1v depending on load. The diodes can handle this fine, but the heat they give off will be a lot, so will require a very large heat sink. If you have a spot on the frame or other large metal surface, you might be able to heat sink to that.

For sure the diodes would prevent one pack feeding back into the other one.

The other downside to that arrangement is you need a separate BMS for each string (if you're using a BMS). If you parallel the packs at the cell level, with interconnections between each cell, then a single BMS can serve both and you don't need a diode.

 

[RedLine19K]

here's the datasheet..
400a100v
maybe .6 at 40A? Is that 24w heat? Thats one i had saved for EV drag racer... would this be better?
This is the one i was thinking of with .4v drop, but only 15v
100A 15v

I didn't think he was using a bms, in which case i think i'd favor the 5s2p x 3s instead of 3s5s x 2p... but 3 or 6 of these:
8s voltage alarm

adjustable voltage alarm for each cell/pack... not exactly convenient - leave them connected would drain a pack in a day/week? but.. they will scream before batteries are cooked. Plus they are handy for checking balance.. especially before paralleling packs