LiPoly evolves while LiFePO4 is stagnant?

methods said:
Why dont you want to balance them?

These cells just don't need balancing is all. The Makita/Bosch packs these cells come from don't do any balancing at all... they only do over current protection, LVC protection etc. Before my my CBA II broke I did 42 konion cell discharges and the lines were pretty much identical.. like I couldn't see anything but one line despite that there were 42 on the graph... haha... you can differentiate only at the very, very end of the discharge. Simple is beautiful.
 
wow. . .
As I sit here banging my head against the desk playing with different BMS ideas that sounds real good.

-Patrick
 
Regarding the Konions, which are LiMn-based, I think one thing that helps them stay so well balanced is that because they are smaller, you need to use more in parallel for a given setup, so they tend to help each other. I haven't read where anyone using these have drained them down close to their minimum level (3.0V?), under typical bike use, and then seen how well balanced they are. Doc's megapack hasn't been close to fully discharged, I think, and I just haven't seen many other reports. Nonetheless, I don't think there is anything inherent in the chemistry that would make them any less susseptable to becoming unbalanced. Most of my a123 packs stay very well balanced, as long as I don't run them down to LVC cutoff. Typical high-C LiPo packs also stay very well balanced, but they too can eventually drift apart, so balancing, or individual cell charging occasionally is always a good thing.

I also don't believe that Konions can bypass any current on their own when the cell is full, like an SLA can, so I still think if you do have a low block of cells, and you are bulk-charging the pack, it will never get a complete charge as the higher level cells will get full first and essentially cut off the current. When an SLA cell gets full, it can absorb a bit of "extra" current, which allows the lower cells to get at least that much current so that it will eventually get full. Not so with any Lithium-based chemistry that I'm aware of.

-- Gary
 
For the folks who aren't running more than 12s packs, or groups of 12s packs when charging, this little guy is pretty cool.

http://www.helipross.com/mega-power-lcb-12s-deluxe-lithium-cell-balancer.html

He functions as an always running stand-alone balancer, or just kicks into balance mode when cells are above 4v or where ever you set it to kick-in. Thats the beauty of the thing, it's completely programmable to work with any cell chemistry. I have the charger that programs them, and i would me more than happy to configure/program them for anyone who needs it done.

It has a fully programmable HVC that you can set anywhere between 2s to 12s total pack voltage, AND it also shuts down charging if any single cell goes above the threshold you set (maybe 4.25v).

It was made to work with a little 10amp charger, but I've done some heavy beefing up on my mine, and it's taking 26.5 amps through it like a pro. The FETs to control the charger current are IRF2807's, which are rated to 75v and 82amps. It runs two in series with little heat sinks on them. I bought 10 more of this FET, and I plan to bump it up to be able to handle my 53amp charger current safely. I also ran some 12awg wire from point to point on all the high current areas of the board, though the board was all ready pretty beefy looking.

I'm building mine right into my pack, along with a display that gives me a digital display of the voltage in every cell all all times.
Both of these things together don't even weigh half a pound, and I'm enjoying the piece of mind I will have from knowing my charge current is instantly stopped if any cell gets above 4.2xV, and knowing that my pack will be in a constantly state of balance as I ride.

It's really quite a lot of bang for the buck towards preventing your lipo packs from making their own bangs :)
 
I actually have (4) 12S balancers all taken apart on my desk right now.
I reverse engineered the LBA10 12S balancer, found the opto couplers, traced to the chip, and identified the communication format as UART.
That is about as far as I got before I got bored.

Tonight I am building my new cable that will allow me to run the HVC / LVC alerts on the bike then plug in at home with a DB25 to charge with the dual 12S bridged balancer

I had not seen the above 12S balancer, thanks for the link.
Here is the 12S balancer that I have been using:
http://www.atsrcplanes.com/lba10_cell_balancer.htm

I was going to leverage its cutoff circuitry for a different task but I have decided that I hate this balancer for 3 reasons:

1) You MUST apply "pack voltage" to the red wire
2) There are 2 fets that control the black wire and you cant use it for anything that is not referenced to the position 0 cell (lame)
3) There is no indication of how many cells you have hooked up when you turn it on. I.E. if you have a broken wire it wont tell you
4) To expand past 12S it either needs a bridge balancer or a uC to intercept the uart signals. Not worth the trouble
5) Crappy user interface, floppy wires, semi expensive.
6) Does *not* force-balance cells that are over 4.2V. A good balancer will discharge any cell over 4.2V

I have modified 2 of mine by removing all the wires and internally soldering up the power.

If a balancer has "force balance" for cells that reach 4.2V then you can use multiple balancers without tieing them together.
For instance, you could use 4 6S balancers to balance a 24S pack with a power supply. Any cell that reached 4.2V would be held there until the rest caught up.

With crappy balancers like the LBA10 if two cells are at 4.26 and 4.26V it will just let them sit there like that. Irritates me to no end.

-methods
 
The mega power unit does force balance, but it also won't balance until it has the outside leads from the pack plugged into it. However, since all it does it sense total voltage with those leads, I'm sure you could just run a wire on the board from the 1st and last balance pins over to the spot on the board that it wants to see pack voltage, as they would be the same.

It has 12 bi-color LEDs, and it does off, solid green, blinking green, solid red, or blinking red to indicate what it's doing to each cell.

It does have a major downside though. It only discharges at 70mA to balance... That means it's not going to be able to hold any cells anywhere during high current charging. It's more like something that will maintain your cells in balanced before charging, and then shut down charging if one goes above 4.2x. It shows a pretty simple transistor/resistor layout repeated 12 times on the board, I'm sure I could setup different FETs and 10w resistors and make it have 2amp balancing ability, but I think I'm just going ot run it as it is for a while.
 
For the LBA10 you only have to connect the red wire, not the black wire.
I have desoldered all 4 wires on the board and ran jumpers from the +V the #7 pin already as you eluded to.
(one step ahead of you son ;)

The LBA10 balances at around 450ma at 4.2V so that is good

Are you sure yours forces balance at 4.20V? You should measure it. I bet it is like most of mine where it does not force balance until 4.25V or 4.30V

Currently I am soldering up little proto-boards that have a bunch of sockets. There is one proto board for each 6S pack.
Each board has:

7pin JST XH for the battery balance input (Y cable attaches that ties multiple parallel packs together)
7soc JST XH for the Balancer
7soc JST XH for the Monitor
7soc JST XH for additional packs

So now all this crap will be on board and all I have to do is plug in.
While charging I will get audio alerts for any cell over 4.22V and I will have full 24S synch. balancing.

While riding I get alerts for LVC at 3.0V and HVC at 4.22V
Always scares the crap out of me when I hit Regen and my alarm goes off :shock:

-methods
 
That's the cool thing about these megapower 12s units. You program them to do what you want. You trigger force balancing to start clear down at 3.7v lipo, and 3.0v LiFePO4, or you can set it as high as 4.35v ( I wouldn't recomend that). It has a pretty slick microcontroller in there that is totally programable to change all the balance and cut-off points to be where ever you want.

But, it sounds like the hyperion balance unit has a much better discharge rate. 70mA is kinda laughable when trying to balance 20Ah cells, or when charging at 26.4amps now, and soon to be charging at 53amps. I'm just going to see how well my cells stay in balance for a while, and if it needs to be modded to balance at higher current, I have the parts on hand to make it balance at a couple of amps.

Gotta love LiPo 500-600amps of fury on tap, and re-charge your pack from drained to fresh in under 30mins or your pizza is free :) This 0.2C charging stuff is for kids.
 
methods said:
10Ah * 15C gets you 150A rated output
You could dump 200A burst no problem

Does this mean there would be negligible voltage sag? Because my Lifepo4 sags to 46V from 50V when i accelerate.
 
All depends on the internal resistance.
The simple way to think about C rating is like this:

(please forgive if you have heard this before)

Take a battery that is sitting at its rated voltage. Lets say 50V.
Now apply as much load as you can while maintaining a constant temperature and not dropping below LVC. That would be your constant C rating.
Then apply as much load as you possibly can without crossing LVC without regard for temperature. That would be your Burst C rating. It is burst because if you continue at that rate the pack will overheat.

That is not exactly how it works but it is a good way to think of it. Luke can probably give a more specific definition of how a battery is rated off the top of his head.

So the higher the C rating the stiffer the voltage will be.

If the C rating says you should be able to pull 150A and you are pulling 20A the pack wont sag at all.
If you are pulling 60A it is definitely going to sag.

You will most definitely see 4V or more of sag under heavy load. The more cells you run the more sag you get.

Think of V = I * R Ohms Law

V is the voltage of your pack.
R = R1 + R2
R1 = resistance of your pack
R2 = resistance of your motor.
I = Lets make I constant for this example

Power = I^2 * R (current squared times resistance)

I^2 * R1 = heat in the pack
I^2 * R2 = power to your motor

The C rating is inversely proportional to R1 so to make your C rating go up we need to minimize R1
This too will minimize the voltage developed over R1 (which is your sag voltage).

Maybe that explanation will help you visualize the voltage sag.
Dont forget that both R's go up with temperature.

-methods



-methods
 
Quality LiPo has very little voltage sag. When you dump 100+ amps continous from a single 5ah lipo battery, and that battery stays cold to the touch, you know you aren't getting diddly squat for voltage drop, because battery voltage drop is power lost resistively heating the pack.

Im not a LiPo sales rep or something, it doesn't make a lick of difference to me what battery people choose to use, but I don't think you will find anything out there that can compare if you are looking for high performance, and can accept the potential risks that come with LiPo.

Methods can simply velco giant LiPo packs all over his bikes, and then regularly crash and wreck into things, short them out into things, and charge at crazy rates with crazy proto-type chargers and not have a single LiPo problem. That should tell you something about how much safer modern LiPos have become.

It's really not an unreasonable set of rules to follow for a battery.

Don't get a cell over 4.25v
Don't discharge below 2.5v
Don't stab, slash, or crush your cells.
Run a fuse because ish-happens. This is just a darn good idea for any performance bike and any performance battery.
 
I remember a similar set of rules given for a fluffy little guy that went by the name Gizmo :mrgreen:

No bright lights
Dont get him wet
and most importantly, never, ever feed him after midnight.

-methods :wink:
 
Wonder how they compare as far as connection robustness?

The biggest consistent problem I have had is my 12 ah green rectangular pack's terminal connections get loose from road vibration. Same pack Ypedal has (we really need a name for these). Run them in series with a 10 ah Lifebatt pack.

Run mostly to and from work (in the summer) without a LVC or BMS. I just never run them above 8 ah. My Flintstone charger seems to keep the rectangular pack balanced.

DK
 
Internal resistance on the black wrappings LiPo packs sold on fleabay as 12c is something like 15-20mOhm for the 5ah cell.

Internal resistance on the quality but still cheap zippy lipo 5ah cells is 4.5mOhm.
Top shelf stuff like polyquest and thunderpower are around 3.5mOhm for 5ah cells.

A 4P pack of good zippy cells is going to be 1.25mOhm, and a good pack of TPs or PQs is going to be around 0.9mOhm.

So, for some quicky calculations:

Fleabay crap black shrinkwrap oldschool LiPos 4p 20Ah @ 100amps is going to drop 0.5v/cell (and heat the battery with 50w of heat!!!, 12.5w of heat per battery)
Hobby city zippy 5ah cells 4p to make 20Ah @ 100amps is going to drop 0.125v/cell. (and heat the batterys with 12.5w of heat, 3.1w per battery)
TP cells 5ah 4p to make 20Ah @ 100 amps is going to drop 0.09v/cell, (and heat the batterys with 9w of heat, 2.25w per battery)

So, if you were running a 50v 20ah pack at 100 amps, you would see 6.5v of drop with fleaghey crap lipo, 1.62v drop with zippy, and 1.2v drop with top shelf stuff.

*Edited to correct error*
 
Deepkimchi- No terminals or screw terminals to come loose on these things. Everything in the packs is a soldered robust connection, and on the good cells, it's a very robust wide copper tab soldered to a piece of ultra fine strand 10awg or 8awg silicone insulated wire leading out to a deans ultra connector. RC guys are not into loose connections. Loose connections means a +2,000 plane or heli falling out of the sky like a stone.
 
liveforphysics said:
TP cells 5ah 4p to make 20Ah @ 100 amps is going to drop 0.9v/cell, (and heat the batterys with 9w of heat, 2.25w per battery)

So, if you were running a 50v 20ah pack at 100 amps, you would see 6.5v of drop with fleaghey crap lipo, 1.62v drop with zippy, and 1.2v drop with top shelf stuff.

uh, for those that it's not obvious u misplaced the decimal, it should be 0.09V/cell.
which i calc to a 1.26V drop, would that be correct?

i'll assume this was deliberate to c if anyone's paying attention. :wink:
 
Whoops!!! Thanks for catching my mistake! :p

I will edit the post to correct the numbers, and I will try to be more careful and double check next time!


-Luke
 
Good info, Luke. :)

One other point about high-C/low IR cells is that becuase the voltage sag under ebike-typical loads (100A, or less..) is so low, the cells stay balanced better. I've seen the same thing with a123-based packs. If I don't run a pack down to LVC cutoff, the cells will stay well balanced. Running down to cutoff, however, will cause one, or more, cell blocks to end up lower than most of the rest. This is due to slight capaciaty and IR differences that develop over time. Healthy cells will bounce right back, however, and the pack stays well balanced, through multiple simple bulk charges, until the net time I run to LVC cutoff. This is with individual cell LVC circuits, not a controller/pack level LVC.

What will be interesting to see how these lower-priced Zippy LiPo packs hold up after 500-600 cycles. Given yours and Patrick's proclivity for extremes, I'm guessing we won't have to wait long to get data on tortured packs with this many cycles. That, and just seeing how well they do after a year, or so, even sitting on the shelf. The big factor will not be how high a C-rating the packs have, but the level of quality control enforced during the manufacturing process. If the QC is tightly controlled, the cells will resist the intrusions of moisture and air that ends up causing lesser quality cells to "puff", just sitting on a shelf for a few months. I still have some "gen 1" Thunder Power packs that still hold the fully charged surface charge, and I haven't touched these packs for about 1-1/2 years. On the other hand, I long ago got rid of any of the problematic "gen 2" cells that would puff within weeks sometimes.

Hopefully, Patrick won't fry his 'nads in a crash, having all that unprotected LiCo mix between his legs, but I hope those opting to take advantage of the great pricing of these Zippy packs will install the packs in a way that offers good protection for the cells. :roll:

-- Gary
 
I am always willing to set an example for my peers. :roll:
I promise that if I fry my nads off I will take lots of close up pictures :wink:

I have been listening to what you say Garry...
You probably know that I did have a bad over-the-bars crash with one of my Lipo packs already so it is not an abstract concept.
I am thinking of taking some of your advice and creating a nice lipo-wrap snuggle bag for them.

A good nylon like the type used for backpacks is strong, puncture resistant, tear resistant, light, water resistant, and easy to sew.
I am considering having my wife make up little jump suits for my lipo packs.
I could have built in velcro and straps. Maybe thin hard plastic walls.

I dont buy into the "contain the fire" argument but I definitely agree with your argument for resisting puncture related fires.

-methods
 
Nylon makes a terrible sticky burning mess when it gets hot or catches fire.

Nomex and fiberglass cloth are both cheap and very well suited towards protecting you from your LiPo. Same stuff the LiPo sacks are made of.

I'm making my LiPo pack out of lexan. Light, flexible, shatter proof, abrasion resistant, see-through, and I can't puncture my 1/8th inch sheet with my hardest swing with an icepick. With heat gun, you can shape and form the stuff however you want. Won't be the best as protecting me from my LiPos, but it will be fantastic for protecting my LiPos from me.
 
LOL! :shock: :oops: :lol: 8) :eek:
 
Hi Luke,

liveforphysics said:
LiPoly design has managed to roughly tripple the C value in it's cells, and increase energy density at the same time. It has done this in just a few years, and continues to grow and improve rapidly.

So, that brings the big question, what has the development of LiFePO4 chemistry stone-walled? Unless I'm mistaken, the best performing cell you find is still the A123 2300mAh cell, which weighs the same as a 5000mAh LiPo cell, and has lower power density, and half the energy density.

I would think that due to the inherent safe properties that large amounts of RnD funds are being applied towards further improvement, but why has lipoly zoomed ahead while LiFePO4 takes a seat?

In the C race LiFePO4 has just passed lipoly and caught up with Ultracapacitors!:
http://www.technologyreview.com/energy/22280/

Ultra-High-Power Lithium-Ion Batteries

New materials from MIT could power laser weapons or give hybrid cars jackrabbit acceleration

...Test batteries based on the new electrode--developed by Gerbrand Ceder, a professor of materials science at MIT--can be discharged in 10 seconds. In comparison, the best high-power lithium-ion batteries today discharge in a minute and a half, and conventional lithium-ion batteries, such as those in laptops, can take hours to discharge. The new high rate, the researchers calculate, would allow a one-liter battery based on the material to deliver 25,000 watts, or enough power for about 20 vacuum cleaners.

This level of power output would put these batteries on par with ultracapacitors, gadgets that can rapidly discharge power but can't carry much energy for their size, says John Miller, a vice president for systems and applications at Maxwell Technologies, a manufacturer of ultracapacitors, who wasn't involved in the research. The new batteries would store nearly 10 times as much energy as an ultracapacitor of the same size. The combination of small size and extreme power could make the batteries particularly useful for race cars, he says. (Starting this year, new Formula One racing rules will allow race cars to store energy from braking to deliver very brief jolts of acceleration.)

To improve the batteries, the researchers modified an electrode material called lithium iron phosphate to allow electrons and ions to move in and out of it much more quickly. The advance is based on computer models that Ceder developed in 2004. The models suggested a way to improve conductivity by directing lithium ions toward particular faces of crystals within the material.

To exploit this, Ceder included extra lithium and phosphorus. This helps form a layer of lithium diphosphate, a material known for its high lithium-ion conductivity. He says that ions encountering the material are quickly shuttled to faces that can pull them in, allowing for very fast discharging.

The fast-discharging materials may also recharge quickly, raising the possibility of cell phones that charge in seconds, Ceder says, but this would require expensive chargers. Ric Fulop, vice president of business development at A123 Systems, a battery maker based in Watertown, MA, that has licensed Ceder's new material, says that it could be useful for hybrids or for delivering the power needed for laser weapons. (Fulop notes that A123 is not developing batteries for the latter application.)

Unfortunately energy density hasn't caught up (although the new large format A123 cells might be an improvement over their existing cells):

Other researchers have already modified lithium iron phosphate to achieve power levels high enough for power tools and for most hybrid vehicles. Indeed, iron phosphate batteries are already being sold by more than one battery maker for such applications. Ultimately, the energy capacity of lithium iron phosphate is lower than that of other lithium-ion battery materials, making Ceder's advance of limited value, says Jeff Dahn, a professor of physics at Dalhousie University, in Halifax, Nova Scotia. This battery is good for acceleration, but not as much for long range. "A real breakthrough . . . would be a new positive electrode material with quantum-leap performance specs" in energy storage, Dahn says.
 
Cool,

Hopefully this new tech will soon cause the original A123 cells to become 2nd tier. Then the ramped up production makes them plentiful and in the $1-$1.50/cell price range. To me they are good enough in all categories other than price and availability. The change I'd like to see is that they work toward no need for balancing.

John
 
Back
Top