Building a123 packs for dummies.....

Hi

So i've recently purchased a set of a123 cells; and gotten a free used pack. I've in the middest of testing the cells on the road managed to drop 2 cells per pack below 2.5v; i think it was something to do with my original setup; i had 2 packs in parrell 36v/4.8hr. I was using 1 bms & spliced all the balance leads to accomidate charging 2 36v packs at the same time; While riding not even 1 amp hour on my c/a i managed to bring 2 cells per pack below the 2.5v mark damaging the cell, the cells still functional but now only charge to 3.34v mark. I was bypassing the bms & obtaining the current directly from the cells. The cells i managed to drop below the 2.5 mark where the 2 cells on each pack used for both + & - lead. i've managed to rebuild one of the packs & put all near matching cells that fully charged showing 3.42 v, the other pack thats in pieces have cells with voltages all over the place; some 3.6 some at 3.34 fresh of the charger. I'm really upset that i've managed to mess of the packs.

Could you please advise how i can match the cells to try & rebuild the second pack. i've read

Match by internal resistance ? what do i use thou ohms, kohms mega ohms?

voltage - i don't agree with this part; even thou some cells after charge are at 3.6v & another at 3.34v once on load they all start to drop around 3.35 - 3.30v on a stead load with 3 100 v bulbs.

capacity- How could i measure this? what tools would i need?.

any info or fyi you could give would be much appriciated

please advise
thanks In advance

I'm a dummy, does that qualify me to answer

One of the simplest ways of measuring internal resistance is to measure the voltage drop at two different discharge rates:
http://www.buchmann.ca/Chap9-page2.asp
Don't know how hard it is in practice, as I've never tried it myself.

One thing I have noticed is that position in a pack can make a surprising difference in voltage. Charging a 10p pack last night I found a 0.10V drop between the first cell and the last cell in the parallel string. That was using short connecting leads of 12AWG cable.

I'm suprised this hasn't come up already. I want to upgrade my current
ebike/motorcycle system which is ((2X12v 20Ah series) + (2X12v 20Ah series)) series = total 48v 20 Ah battery system connected to a 300amp alltrax controller, the controller pulls alot of amperage at peak loads.

What sort of tools, batteries, circuit boards and chargers do I need to build a battery pack of similar size as this but in Li-ion. What are the configuration options of all these famous 18650 and A123 cells? how are they charged? what kind of PCB and protection is required> My goal is to build minimum 2 24vX20Ah safe LI-ion packs and then connect them in series to get 48v. If I could increase Ah somehow without too much increase in size, this is a plus aswell. Ofcourse I'm after the big pack for more range and more power!

Thanks! look forward to replies.

its good to know there are others out there that are lost like me lol ..

It takes time for it all to make sense.. but bear with it and keep reading.. don't get stuck on specifics..

300 amp alltrax controller = BIG lithium pack ! lol..

Depending on alot of factors, ie: bike weight, rider weight, terrain, and expectations of performance.. there are many ways to do this.

Expensive as hell, but if proven worthy a 4 paralell pack of Lifebatts come to mind !! ( 10 amp/hour per cell = 40ah, x 10C = 400 amps available on command !! )

one with with expensive lithium is you want to build it right for the application.. trial and error gets reallllly expensive with lithium. so you don't want to go there.

If you trash 100 $ worth of lead acid.. well.. woopee.. no big deal.. but fry 3000 $ of lithium and things start to suck.. big time !

GreatScott! said:

My goal is to build minimum 2 24vX20Ah safe LI-ion packs and then connect them in series to get 48v

Click to expand...

Like Ypedal says, the simplest (and possibly most expensive way) would be to buy LifeBatt cells and battery management system (BMS).
The cells are 3.7V/10Ah each, so to replace your existing pack you would need a 16s2p pack, i.e. pairs of cells connected in parallel to give 20Ah, and all 16 pairs then wired in series to give (roughly) 48V. You could split this pack up into however many subpacks you need to give a convenient arrangement on your bike. The LifeBatt cells are still relatively unproven, so you would be something of a guinea pig.
There are also a number of prebuilt lithium iron packs available that people here are just starting to test (Yesa springs to mind), but they tend to have lower discharge rates, so might not be the best choice for you.
The A123 cells have a proven track record and can deliver very high currents (used to power KillaCycle to sub 8-second quarter mile), but at present are only available in 2.3 Ah cells. So you would need to build a 16s8p pack = 128 cells. There is no off-the-shelf BMS available, but Gary Goodrum and Bob Mcree here are working on that right now...

One other thing: If you replace lead with lithium you will lose a considerable amount of weight from your bike, and you'll be able to get most of the rated capacity from the batteries (unlike lead). The combined effect should be a healthy increase in range over the same Ah lead acid pack.

Can you give us some more details about your bike?

yes i'd agree with the above - I own a yesa pack, and i'm on the limit of how much you'd want to draw from it - for your controller, you'd def want either A123 or lifebatt, you need their high discharge ability.

A123 in custom packs is basically limited to a custom bms style. there are ones that trigger the brake line of the crystalye controller. I don't konw much about your controller, but i'm assuming you have a similar input - that would be the best way to go. I'd probably reccommend that at this point. LifeBATT don't have much in the way of inderpendant testing and are not proven in my opinion (not open for debate in this thread).

Another option I almost forgot are the bigger Thundersky cells that Electric Motorsports and Thunderstruck Motors were selling. These would also need a custom BMS.

The TS cells are not going to work at high C rates from what i've read about them.

Something i have been wanting to do for a while now is build a large pack of RC Kokam Lithium Cobalt cells.. ggoodrum could shed some more light on this..

An E-tek with an alltrax controller mated to a large ah Kokam pack would be very interesting=Fun !!!!

Ypedal said:

The TS cells are not going to work at high C rates from what i've read about them.

Something i have been wanting to do for a while now is build a large pack of RC Kokam Lithium Cobalt cells

Click to expand...

Just checked out the Electric Motorsport site and they're only selling the Thundersky cells to developers now, which suggests they have some reservations themselves.

Don't know much about the Kokam cells, what's special about them - are they lighter than the LiFe cells?

http://www.maxamps.com/Lipo-5000-111-Pack.htm

150 $ canadian x 12 ( 4 series 3 paralell = 12 packs ) = 1800 $

That's 44v 15ah = 9 lbs !

Set the alltrax for 200 amps and hold on to your socks..

----

That's just one of the multiple options out there.

mmm that would be perfect for a motobike style setup

welcome to lethal voltage / currents though :S

hey guys,

big THanks for all the information, yeah really is a learning experience with Li-ion. I am trying to absorb it all in and I am learning tons by reading threads on this site. It's something I am cautious to dable with yet however I know I will be able to with a little more reading and time. I don’t mind doing the technical stuff.

So when building a Li-ion packs, is it as simple as putting a PCB in for every larger module you make? like how is the BMS structured, I would imagine it depends on cell pack size, then module size then overall packs size and number? SO as I understand it, normal li-on packs wouldn’t draw enough current, what symbols should I look for when choosing a pack to draw current for a 400+amp controller? Is C rating? Continuous? Does the C rating get larger as you add more modules or cells?

Another thing, how is a multiple system pack charged, I keep on seeing these little 1-4 chargers everywhere that let monitor and charge small 4 cell packs etc. but this would be impossible lets say for a 16P8S super pack, you would need way too many chargers? is there a scalable charger to let you charge packs and modules as you build them larger. I don't mind using maybe 4 small chargers max on a super pack. I use a quad charger now on my SLA setup and it does each battery independently in under 2 hours, its super small and light too.

Ok, so under and over voltage is apparently a big deal with Li-ion, is this during use or after they have been used? I can see how an individual cells voltage could be monitored but what about large complex cell / module / packs with many cells connected in series and modules in parallel etc. How is this ever important aspect monitored? For instance, if you put 4 3.6v 1.8Ah (rated at 4.2 max V and 2.5 min V) Lion cells in series to get a larger V, 14.4v 1.8Ah battery module, would you simply average the voltages over 4 to make sure everything was fine? Would this be a good technique? Ex. You would make sure the charger voltage readout doesn’t go over 4.2X4= 16.8v or under 2.5X4=10v? If this is correct, how is this done as larger packs are added?


Ypedals link, maxamp.com. Absolutely amazing! WOW, I’m blown away……..

When I built my Emoto prototype and tested etc. I realized that there is huge potential in the concept of E-transportation however, just the other day, I was mentioning to a friend how the one big, huge, crucial aspect holding this whole industry and culture from exploding…no pun intended….is Battery technology. I was thinking, If only I could get minimum 2-3 times the range on my prototype with no extra weight gain or size, then I would have a serious machine on my hands with true market potential, as right now using SLA technology I can’t see it moving past a fun little backyard invention, the range is just not far practical enough . So then I stumble onto this site and start to read about Li-ion and get linked to sites like http://www.maxamp.com and I am blown away. If the specs are accurate and I do the math, with new generation Li-ion chemistries, the lightweight, triple range elusive magic power plant battery pack could be built….I’m simply stoked. Is it hype? Why haven’t I heard more about Li-ion is main stream news?

Ehem, ofcourse, however looking at sites like http://www.maxamp.com, I might refine my statement that not so much the batteries are holding back the EV phenomenon but more massproduction,…so the COST at this point.  Sure, the technology is there to build a pack that will be 1/3 the weight, ¾ the size and triple the runtime and torque of my current battery setup……………………..it will just cost me a house mortgage to build… So for now, I think I will be sticking with SLA. Doh!

Ypedals maxamp battery setup would be pretty sweet, technically how would it be done, What sort of BMS or PCB’s would be needed. Would it become larger in size because of this, would it need to be cooled? Etc. I will keep it mind for future.

12 (4p3s) 44v 15Ah, not enough, haha, I would double the thickness of the pack, 24 (4p3s) ie. 2 12(4p3s)layered for 44v 30Ah pack for what? Only 20lbs and 3600$ plus shipping. haha. Sick sick sick…why stop there, to equivalently match my current SLA size and weight, I could push it to 72 or 6 layers of 12(4p3s) given how thin they are for a total of 90Ah!!! At 44v and only 54 pounds and 10800$!! Haha. Now that’s more like it!! If my math correct, 90Ah/20Ah(current SLA pack) = 4.5 longer duration….ridiculous, I will call the bank tomorrow for the loan...Jokes.*


As far as bike goes, would love to post pix of my prototype, however I’m researching legal things first before I post publicly. I haven’t done anything super special technology wise, its just form factor and application that I’m looking into patenting. If I can get my monster to go minimum 2-3 times duration with consistent torque, then I’m going for it. This Li-ion stuff looks promising. Right now, SLA gets me around but torque dies off rather quickly if you giv’r like you would a real MX bike.

Ok, sorry for the rant, lotta questions, a lot of learning to do.!!!

cheers

http://www.electricmotorsport.com/new/index.php

I hope you have read and seen all this before you go too far !!

http://thekneeslider.com/archives/2007/08/07/zero-electric-motorcycle/

Ypedal..

Funny enough, Zero, formerly electricross is what inspired me to build my prototype bike. No, if I were to patent anything about my EV it would be industrial design or form factor claims and just make a company that way. As anyone can build an electric motocycle. Duh.

Warning...spoiler ahead****








Yeah, the thing is with the zero, the blade etc. they are somewhat misleading. I built mine for way cheaper, obviously avoided all the R&D markeup, in my offroad bike kit, I've used exactly the same controller, SLA battery setup etc. even with a few extra goodies to increase range which shall remain undescribed..*winks.......I don't get nearly their originally stated 20+mile range, When Zero was using SLA batteries 4 SLA in series 48v, they said their bikes would do 20+miles, same with blade. TOTAL BS!!!! Now they use Li-ion and they say they do 40+. Probably not accurate, either. Think about it, if they did this, Neal Saiki would be the next Bill Gates, he would instantly rich, It would be all over the news etc. 40-45miles on a charge is really far, maybe, maybe on road but only under ideal conditions no stop and go, real world scenario it will get a lot less take it off road and another story. But again I can’t offer to much insight on because I’ve never tested li-ion.

Here is the thing, they state their statistics for ideal conditions @ 20-25km(sorry I'm canadian..lol) hour on flat paved roads with no stops and little gradation in road. This is not exactly real world conditions. Common knowledge and testing my prototype show that the two biggest things that are going to kill your battery are starting from stopped position, ie, eisnteins laws of physics, an object at rest tends to want to stay at rest and it takes alot of energy to get it moving again....zappppp, there goes your batt. now add loose, thick dirt or soil in off-road condition and it get exponentially worse. The next thing is gravity or hill climbing. The last thing could maybe be the peanut butter sandwhich which you forgot about that’s now lodged in your motor venting…don’t know..anyway….

With the identical to zero's 48v, 20Ah SLA setup with Etek motor and 400amp controller set to 300Amps, plus the extra range increasing goodies,……… I get max 10-13km traveling on average 20km on relatively flat road paved surfaces with few stops and riding smart like using gravity for coasting on hills etc. That is pushing it, torque dies after first 5-7km. This is no where near zero’s stated 20miles for SLA. I cannot vouch for their new lithium packs but make the inference. It still doesn’t check out. Now this is on ideal road conditions, take this setup off road on some trails and ride it hard, stop and go like you would on a 250mx and you'll be lucky to do 1km before torque is zapped, 1km round trip. I do it all the time on my emoto, its hella fun but she doesn't last long. Put me on their bike and I will kill it. For 6900$ no way! get yourself a real 250 and pay the court fines when the cops chase you down the street! you'll still come out ahead.

Furthermore, Zero uses the Etek, as does mine, which they say apparently puts out 20hp= 250cc? nope, not at 48v they don't, it is 9Hp. 72v would do it but at the same batt configuration would be lucky to get out of your garage before the batt goes down. http://www.electricvehiclesusa.com/product_p/mo-et-2401.htm

I think that I would be happy if the Zero or even my similar emoto could get the original stated 20mile range(for SLA setup) off-road with hard riding, a good 15-30 minutes. I would imagine that would mean the pack would have put out atleast 2-3 times the Ah rating of the current SLA setups, which from what I read is sounds possible with current Li-on, but 40-45km, get out! If they misstated their stats for SLA setups, then 40-45km is probably exaggerated too. I can’t wait until we see packs that will put out an honest 40-45km range with no loss of torque for the whole time, in off-road conditions with reasonable weight. This will truly be a new age of locomotion.

Warning, ….spoiler over****

However, all is not lost, on road, if ridden conservatively, my emoto as well as zero’s, blades and other similar setups can be a hoot and can get pretty descent range ( not near what is claimed but...) making sacrifices for speed and elevation changes. So yeah, glad I build my emoto, no regrets, just waiting for the elusive super pack to get that puppy moving far and fast!

Anyway, someone find me a superbattery and I will partner you in my company, what, we'll say 70/30 split? sound good? Hehe.

Cheers.
Jon-

You are suffering the " Peukert " effect with SLA.. your 20ah is likely delivering anything from 5 to 8ah of it's rated 20ah before going flat at high-rates. = your problem.

Lithium will change things drastically... that 15ah pack from maxamps will deliver almost 15ah.. unlike your sla.. so double your range.. realistically.

Pushed as hard as you describe you will not get mucho cycles from the packs..50.. 100.. 200 maybe.. ? RC guys ???? so your cost per cycle goes way up...

A big enough LiFe pack ( Once they magically begin to work.... trust me... you want to wait a bit more !!! ) would work hard and deliver 1000 cycles in theory.


The maxamps cells are Lithium-Cobalt, they are light, have high-power but need to be treated with respect... not something i would be happy taking a crash with..... Major care would have to be taken to protect the packs !!! with electrics to litterally stop you when any limit is reached. These electronics are not designed to handle the power you want to run so this adds to the expense greatly.

Lithium-Manganese is another option, Milwaukee and Sony etc.. safer than Li Cobalt, but still not going to offer high-cycle life when pushed hard.

Lithium-Iron-Phosphate is what we all hope to see work out for this type of stuff !!!!

Interesting stuff…..thank you.

50 to 100 cycles, what would that be, maybe a few months use, 1800$ down the tubes every few months…….not unless I was sponsored by maxamps. 

Ok, so LIFEBatts look good, they still haven’t been fully tested?

What is Iron phosphate all about, is it prototype stage or can it be bought?

what about 4 of these in parallel?http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=3889

Jon-

Kokam Lithium stuff looks amazing? anyone have a BMS for this ?

Earlier, Malcom mentioned that a 16s8p pack could be built using A123's, What about Batt management system? who should I be talking to for this? Gary Goodrum and Bob Mcree ? I would be willing to send the raw materials over and pay a small fee for someone to build this type of pack with all the safety checks.

Lotta good questions.

Perhaps I'm oversimplifying, but you can get what you want, but it won't be dirt cheap. People here will gladly help you and a few of them provide goods and services for reasonable fees.

New lithium chemistries are safer and should last long if you get enough to keep your DoD (depth of discharge) reasonable.

Even NiCd batteries can last thousands of cycles with reasonable DoD... they are heavier than Li batts, but half the weight of lead and half the cost of Li batts.

LiPo (lithium cobalt) is for suckers and people who love playing with voltmeters.


You might check out the battery info in the EV basics section, if you haven't already:
http://endless-sphere.com/forums/viewtopic.php?t=1403

LiCo for suckers? don't follow, this mean they are not good for my use?


Ok, well my needs and demands are out there if anyone has read all the previous posts I think my battery demands are clear, now just a matter of arrangment and safing them with BMS. so keep posting!


Cheers
Jon-

GreatScott! said:

LiCo for suckers? don't follow, this mean they are not good for my use?

Click to expand...

I mean only a fool would use LiCo, unless they will know the status of their pack at all times.

A lot of folks will be buying LiCo believing it is now safe like LiMn or LiFe... or trusting the label on the pack and never verifying the contents are the safer chemistry.

The ES members using LiCo are experienced and vigilant to the point of obsession... not suprisingly, since overcharge or overdischarge of LiCo can burn down a nice home.

Ah, good to know.

NiCd have good discharge rates? havn't been able to find any past 5C.