LiPo's for dummies

Desertprep

1 kW
Joined
Oct 27, 2007
Messages
356
Location
United States
I have read a few very excellent and detailed posts about Lipo batteries. I would like to have a crash course now, or Lipo's for dummies, to help me sort the detail that I have accumulated. In order to use Lipo batteries for my ebike I have to:

1. Buy a series of packs that will provide me with the voltage and ah that I need.
2. Buy a bms
3. Buy a charger

And of course, take care good care of these so that I don't end up on the front page of the China Daily after they explode! It seems to me, as I was reading through the posts, that there is other hardware that is necessary.

Can anyone help?
 
I probably shouldn't even comment as I am a total Lipo Noob myself, I have just built my first Lipo pack, and am about 5 cycles into it. But in some ways me being a total Noob might be helpful as I recently faced the same questions you are. In addition to what you listed hardware wise, obviously you will want to decide what connection system between the batteries you want to use (ie, Anderson connectors/bullets/solder etc..). This will also include things like whether you will parrallel the balance wires, in which case you will need to get JST extensions etc.. When you say BMS, do you mean something like Gary's/Ggoodrum LVC board? If you aren't getting that, then another hardware item is potentially low voltage buzzers. Other hardware to consider is a battery medic(s) and cellog(s).

There are two Lipo building methods, the "fixed" battery method, where you build the packs into a fixed configuration, and potentially put in place fixed charging and discharging harness/arrangements, or what i would call the Lego type approach, where you put the battery into different configurations as you go. I do the latter, where each of my battery packs has split positive and negative discharge leads with Anderson connectors (I based my method on Dogman's thread, "parralelling Lipo, dogman style" http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=21363&p=311650&hilit=dogman+style#p311650). This means i can build up packs in different configuration of voltage and amphour everytime I ride. This approach gives me flexibility with volts and amphour, but it is way more dangerous than the "fixed" pack method, as with the non-fixed or Lego method you can easily short packs when you are rebuilding packs with all the connections required, also, it is much easier for a few connections to come lose and create problems.

My suggestion is that before you start put good thought into your desired voltage and amphour, so you can make an arrangement that doesn't change much between charge and discharge. For example, i have a 20ah 36 volt pack, being 10s4p (i think that is right nomenclature), basically i have two x 4 parralel strings of 5S, that i then series together with one series connectionhttp://www.endless-sphere.com/forums/viewtopic.php?f=14&t=21600. This means that to go from discharge to charge, I just disconnect the one series connection, and then i have two 5S 18.5v 20ah packs, which i charge as two seperate 5S 20ah strings. THen all i need to do is reconnect the one series connection between the two packs for discharge. Knowing how you are going to charge will also help you decide how best to spend your money on a charger.

But I agree with you that it would be great if the experienced Lipo guys on here did a practical for dummies guide in setting up a first pack (I know there is the Lipo Care and Maintenance thread, but that is more a guide on not setting yourself on fire, it doesn't actually set out required hardware and suggested configurations). A step by step guide on what hardware is needed (or recommended), and how best to put it all together. Maybe there is such a thread but I have missed it if there is.

Best of luck with it all. As a Noob myself with only 5 cycles in I assure you that Lipo is definately worth it. The flexibility I now have with pack weight and voltage, and the lack of voltage sag has definately made all of it worthwhile. Cheers Phil.
 
I am by no means an expert on the subject of LiPos, however I do have a tiny bit of experience with which to share. In brief, we can study the problem as follows:

Battery Chemistry:
Before selecting I made a quick-comparison of different chemistries and their merits; LiPo, LiFePO4, NiCd, etc. For my money - LiPo has the highest power density at the best price-point by virtue of mass-production for the RC market and by access to inexpensive suppliers - such as HobbyKing. Once you have sources to a good consistent supply of batteries then we evaluate what can accomplish with the available stock.

Distance:
Next, evaluated how far we need to go for daily commuting (or other significant goal). It is a good idea to pad this figure by 20% for peace of mind. I had a parallel goal of long-distance cross-country travel, so whatever battery I chose – it would have to serve in both roles, and it had to be reliable.

Voltage:
It is important to understand what voltage you ultimately need to have. Most people start out using 36 or 48 V controllers which is very common. I already had a 36V controller that was part of a kit and came with a LiFePO4 battery pack, therefore upgrading to 37V (42-30V LiPo range) was not an issue.

Cell Count:
It so happens that 37 / 2 = 18.5 => 18.5 / 5 = 3.7 V …or the nominal voltage of the LiPo cell. Therefore I chose 5S1P as my “base unit configuration” and stringing 2 of these together in series gets me to 37 V. Easy math.

Generally, the voltage matrix goes as follows for common configurations:
Code:
Volts / Cell-Count / Qty units required in series
37   / 10 / 2 * 5S
44.4 / 12 / 2 * 6S
44.4 / 12 / 3 * 4S
55.5 / 15 / 3 * 5S
59.2 / 16 / 2 * 8S
59.2 / 16 / 4 * 4S
66.7 / 18 / 3 * 6S
74   / 20 / 4 * 5S
74   / 20 / 5 * 4S

Amp-Hours:
Now we go back to the issue of distance required. I need to travel about 22 miles/day minimum; let’s pad this up to 30 miles. Through previous testing 1 Ah @ 37 V will take me 2 miles in ideal conditions. Therefore 30 miles / 2 = 15 Ah.

C-Rating:
This is the value at which rate the battery can discharge or charge. It is typically denoted for example as 15/2 or 20/2. The first part of the figure means that the battery can discharge at 15X or 20X the Ah value in one hour, and the second part means how fast the battery can recharge, at 2X in one hour. The typical discharge range is 15 to 45. For most applications 15 to 20C is plenty good, with higher ratings costing quite a bit more.

Find a Match:
I have experience with HobbyKing (HK), and they sell a 5S1P battery in 5000 mAh (5 Ah) for ~$40 and it weighs 1.27 lbs each. If I need 15 Ah minimum, 15 / 5 = 3 units in parallel. We already calculated that 2 units were needed in series to create 37 V, therefore the complete battery assembly is (2 * 5S) * 3 … which is written as 10S3P, or 6 * 5S1P. The weight would be 6 * 1.27 = 7.62 lbs. The cost is 6 * $40 = $240 USD.

Pop Quiz:
What other unit configurations are possible for 37 V given 4S, 5S, 6S, and 8S units?
  • A: 4S + 6S = 10S in series - which would become the Base Unit to parallel for increasing Ah.

More-Advanced Analysis:

LiPoSpreadsheet.png


Now if you really want to get tricky, I created a Spreadsheet that tracks
  • Brand
    Ah
    Cell-count
    Total Voltage
    C-Rating
    Weight in Grams
    Unit Cost
    Dimensions
    Supplier
From this I extract:
  • Cost/Cell
    Cost * gm
    Cost/Watt
    Watt/Weight
    Cost/Ah/Cell
    Weight in Pounds
    Total Unit Watts
    Ah / C rating
    L, W, H
Once the table is built I populate it with real data from suppliers, and apply conditional formatting with color scales: Green = Optimum (target), Yellow = Average (median), Red = Extreme (edge-case).

The battery with the most Green wins. In truth – there are no clear winners; you must decide which features are most important. I chose by maximum power density: Most voltage and Ah per gram for a given price - which may vary from month to month. I also chose alternate battery packages in case my primary choice was unavailable or discontinued. Caveat: Generally there is value in selecting one base unit and staying with that brand; don’t mix chemistries or configurations, and Keep It Simple Silly (K.I.S.S.)

Conclusion:
Sorting out batteries in this manner can be fun; it’s like a puzzle – with each piece having something to offer. My advice is: Take your time, be picky, ask questions, chose wisely.

I did not use a BMS for many months; however I have been diligent about keeping batteries in balance. I would not mix a BMS conversation in this thread other than point to references on the subject.

I built a custom-charger assembly to suit my needs out of modified Meanwell units. Again, I’d defer to other threads on the specific subject.

Best of luck, KF
 
Let me start by saying that this has been really helpful, coming from someone who is just starting out in this whole EV thing, this rocks :)
Kingfish, I have a question, would the values you used in your calculations vary from motor to motor, specifically that 1 Ah at 37 volts will take you approx. 2 miles. I assume this would change based on how many Amps your motor draws. I am looking at powering my build with a motor that draws approx. 75 amps peak at 37 volts. What would this change in terms of battery duration?

thanks,
M
 
Marcbot said:
Let me start by saying that this has been really helpful, coming from someone who is just starting out in this whole EV thing, this rocks :)
Kingfish, I have a question, would the values you used in your calculations vary from motor to motor, specifically that 1 Ah at 37 volts will take you approx. 2 miles. I assume this would change based on how many Amps your motor draws. I am looking at powering my build with a motor that draws approx. 75 amps peak at 37 volts. What would this change in terms of battery duration?

thanks,
M

There's a massive variation between bikes, motors, terrain etc in terms of Wh per mile (Ah per mile is fairly unhelpful measure when making comparisons, Wh per mile allows comparison between different voltage packs more easily.

For example, my electric recumbent used to average around 10 to 11 Wh per mile at about 20mph, yet the lower powered folding ebike I've just built seems to work out at close on double this, around 20Wh per mile.

Your example of 37V, 1Ah is 37Wh, so that would have taken my recumbent nearly 4 miles, but would only take the folder a bit under 2 miles. My guess is that for some of the more powerful builds on here 37Wh might struggle to take them 1 mile.

Jeremy
 
I've lost the hard drive on the computer that has all my links to threads. I assume you've mabye already seen my lipo noob links thread? Somewhere in this section there is a thread with lots of links to basic lipo info threads.

Step one is deciding how you want to charge. Do you want to make a big pack and charge it all at once? This is simpler to charge but more complicated to construct. Really smart people are at work on providing us the bms to do this. Again, the links to it I have lost. Or do you want to use regular lipo chargers? The problem there is to simplify the breaking down for charging and reasembling of the power leads to run the bike. The chargers tend to be 6s at a time, so you end up breaking a 48v pack into as many as 3 pieces to charge. But it may not be as big a deal as that. There are 8s and 10s chargers avaliable.

The bms is not so required if you are monitoring the pack with a lipo charger on the charge side, and using low voltage warnings on the riding side , and preferably using a wattmeter to keep yourself from ever needing a lv warning in the first place. Ride till 80% discharged max, and you don't need a bms to stop you from killing a battery.

Even with really high c rate lipo, consider 10 ah to be the smallest size pack you consider. You can run 5 ah for awhile to experiment, but get some more asap. It just gives you the abilty to ride a decent distance. Ordering lipo 4 blocks at a time is good for the things coming with more padding around them, so you can always just add 5 ah at a time till you have the range you need.

Don't worry about exploding. Just follow the golden rules in those lipo noob threads. Never charge a pack that has been overdischarged inside! Just buy a new one.

But DO worry about short circuits, and be carefull where a wire with thin insulation might rub. Armor all the wiring that might get pinched in a crash with extra insulation, and put something around the packs to protect them if the bike crashes or simply falls over. A good short may not explode these days, but it will always be able to set a room on fire. So don't build a bike with a battery that shorts out and burns down the house just because it got knocked over by the cat.
 
Hmm... my application for these batteries would be to power a mountainboard, so a relatively low weight... I think that I really don't have to go overboard, I would rather start small and upgrade later :wink:
So 4 x 5s1p packs, 2 wired in parallel and 2 units in series wired together... would this give me what I want?

M
 
Marcbot said:
Hmm... my application for these batteries would be to power a mountainboard, so a relatively low weight... I think that I really don't have to go overboard, I would rather start small and upgrade later :wink:
So 4 x 5s1p packs, 2 wired in parallel and 2 units in series wired together... would this give me what I want?

M

Thanky for the kind comments :)

Q: What's a "mountainboard" ? :?

BTW - Jeremy is exactly correct: Use Watts to calculate distance and efficiency (W/mi).

Cheers, KF
 
A mountainboard os essentially a snowboard with wheels, the sport was created by snowboarders who had nothing to do in the summer. It is an off road downhill sport, they also have a really low center of gravity, making them more stable than their cousin the skateboard. Basically, by strapping a motor to one, you get a street rocket :twisted:
Thanks for the input everyone, upgrading by 5 Ah bricks is definitely a smart idea, but where can I find 10 Ah bricks, or would I just make my base unit by two paralleled 5 Ah bricks?
I plan to take them apart for charging, and have low voltage buzzers rigged to each brick, so hopefully nothing will go too bad :D
Are there certain chargers that are more reliable, it seems that this is a very important aspect of battery upkeep...
 
Kingfish said:
BTW - Jeremy is exactly correct: Use Watts to calculate distance and efficiency (W/mi).

Cheers, KF

Thanks, I was correct in suggesting the use of Wh, but I'm afraid you've missed out the all-important time element from your comment above suggesting the use of Watts. You need to use Watt Hours per mile, as Watts per mile is meaningless. The watt is a unit of power, when what you're looking for here is a useful unit of energy consumed (power per unit time) over a given unit distance (miles). Strictly speaking we should be using Joules (the SI unit of energy) or maybe Ergs, BTUs or even HP Hours, but none of these really fit our needs as well as using the Wh, plus there is a precedent for using the Wh for electrical energy measurement - it's the norm with utility companies (or at least its big cousin the kWh is).

Sorry to be a bit pedantic, as a recently retired senior principal scientist I can get a bit worked up about inappropriate measurement units.................

Jeremy
 
Haha, I'm just staring Chem/Phys :D Ive got a few more years to go, but this is really helpful in understanding the real life use of many of the units and concepts that we learn...
 
Marcbot said:
Haha, I'm just staring Chem/Phys :D Ive got a few more years to go, but this is really helpful in understanding the real life use of many of the units and concepts that we learn...


Good luck with your education, really glad you've chosen science subjects, I doubt you'll regret the decision.

I started life as a chemist, working on compounds using radioactive isotopes of carbon, hydrogen and iodine for medical diagnosis and treatment. I then got deeply involved in developing the specialised analysis and measurement techniques I needed in the lab, as there was nothing off-the-shelf that would do what we wanted. The need to design and develop new equipment to meet our needs gave me a strong interest in electronics, so went and did a second degree in EE (sponsored by my employer - what luck!). This interest in test and measurement then led me to combine my hobby (flying) with work, when I switched careers and became a flight test scientist working for the Government. This really was the best of both worlds; fascinating measurement challenges in a severe environment and flying brand new military aeroplanes, with the added bonus that I was being paid for it! After 38 years working in science and technology I consider myself to be very, very lucky to have had so much fun at work.

BTW, there is another chap here with an electric mountainboard, maybe you've seen his thread down in the stand-up scooters section?

Jeremy
 
Jeremy Harris said:
Kingfish said:
BTW - Jeremy is exactly correct: Use Watts to calculate distance and efficiency (W/mi).

Cheers, KF

Thanks, I was correct in suggesting the use of Wh, but I'm afraid you've missed out the all-important time element from your comment above suggesting the use of Watts. You need to use Watt Hours per mile, as Watts per mile is meaningless. The watt is a unit of power, when what you're looking for here is a useful unit of energy consumed (power per unit time) over a given unit distance (miles). Strictly speaking we should be using Joules (the SI unit of energy) or maybe Ergs, BTUs or even HP Hours, but none of these really fit our needs as well as using the Wh, plus there is a precedent for using the Wh for electrical energy measurement - it's the norm with utility companies (or at least its big cousin the kWh is).

Sorry to be a bit pedantic, as a recently retired senior principal scientist I can get a bit worked up about inappropriate measurement units.................

Jeremy

My bad ~ and I should know better as I've been corrected about this same error before :oops:
Ah well... I can write a program that makes fish go across the screen :)

~KF
 
If only everyone could have fun and get paid at the same time...
Yes, I have checked out other mountain boards and they have been really helpful to understanding the different steps and parts that go into building one, however, for batteries I figured it was the most important part of the project so I better REALLY know what I'm doing :D
 
Ok so let's see if this thread can educate a dummy (me), and I'll draw from both Kingfish's excellent breakdown and dogman's advice on where to start:

I want to build a bike (hypothetical at this point, but hopefully real in the near future) that will use at least 48V so that I can use a 9C motor to do 30ish. I also understand that it's possible to have a battery pack that, when fully charged, has a nominal voltage somewhat greater than this number so long as the controller I have on my bike is able to handle the voltage.

I also want to have the batteries configured so that the weight is spread out over the bike. To do this I would like to have two packs, one on the front of the bike (like a burrito I've seen elsewhere on ES) and one in the frame of the bike. This helps with the weight balance because the motor becomes the heavy component in the rear, but it should also serve to make the charging of these packs simple, if they can simply be disconnected from one another and charged separately, but left pieced together as units of smaller packs.

From the first simple chart that Kingfish created, it looks like my best option would be to use a 59.2V 16S2P configuration, because it is an easy division between two places. Right? This then allows me to have the batteries split into two groups. If I'm right, and I could be way off, then if I used 8 of these:
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=8580
Divided into 2 packs, I would have a 59.2V bike with 20Ah capacity. Right? Or do I need half as many?

This is where the electrical engineering begins to get the best of me, but I think that LiPo seems to be the way I'd like to go. As such, I'd like to get a handle on this. Thanks guys. Great start to this thread.
 
8 of those packs, arranged as two packs of 8S 2P connected together in series, would give you a 16S 2P pack with a nominal voltage of around 59V, a hot off the charger voltage of around 67V and a capacity of 10Ah. There's no good reason not to divide the pack up like this, as long as the pack interconnections are made with suitably hefty cable.

Jeremy
 
I would suggest evaluating the price difference between a single 8S1P and Qty-2 of 4S1P. In my studies I found the lower cell-counts were less expensive. And there are more physical sizes to toy with! :)

Divided between two packs: OK, sure. Suggest using 10 AWG stranded wire for connecting the batteries to the controller and you should be fine.

Best, KF
 
16s lipo will fry most regular 48v controllers, that have 63v caps in them. 16s lifepo4 is ideal for a 48v controller.

You can get away with running 15s lipo, but it's right on the edge, with the battery putting out 63v hot off the charger. I advise the use of 14s lipo, which is just about the same voltage as 16s lifepo4.

So a 14s pack of 10 ah size would have three pairs of 5000 mah batteries. Two 5s 5ah paralell connected, two 5s 5ah paralell connected, and two 4s 5ah paralell connected. Then you series connect the pairs into 14s. Each pair needs a low voltage warning, or a voltage monitor displaying the voltage.

With three chargers, you can charge it all at the same time, or charge slower by charging each pair one at a time with one charger.
A pair of batteries can easily be carried on each side of the frame, and burrito the third pair. But even 4 pairs can be carried on the frame easily, even without any triangle space.

Another convenient option is 12s lipo. This won't be so fast, but you can get 6s lipo blocks and have just two sets of batteries to charge instead of three. I haven't seen 7s lipo, but you could also do an 8s pack and a 6s pack to get 14s.

Lastly, getting a bigger controller is a good plan. Then you'd be good to go for anything from 15s to 20s . You are headed there anyway.
 
Kingfish said:
I would suggest evaluating the price difference between a single 8S1P and Qty-2 of 4S1P. In my studies I found the lower cell-counts were less expensive. And there are more physical sizes to toy with! :)

Divided between two packs: OK, sure. Suggest using 10 AWG stranded wire for connecting the batteries to the controller and you should be fine.

Best, KF

His link was to the 4S, 1P packs, not the 8S packs.................

It looks like he wants 16S, 2P from 8 off 4S, 1P packs, which sounds reasonable to me and is probably around the best bang for the buck, too.

Jeremy
 
Hunting the right price at HK is a chore for sure. Right now the bargain to me is 20c zippy's. 5 ah 5s is about $41. I paid $63 for the same size turnigy 30c for my racing pack. I just bought the zippys for the heat races where I only need to finish decent to get into the main event race.

Other convenient size stuff is out there, like the 10s packs that are pretty close to 36v lifepo4. But then you look at 10s charger prices, and it's not so cheap. All the real bargains on lipo chargers seem to be 6s types, and prices go up for 8s.
 
dogman said:
Hunting the right price at HK is a chore for sure.

I'll say. Pretty much why I'm keeping this as a hypothetical for right now.

I'm trying to decide between two bikes, the new Surly Troll or the Pugsley (fanboy I know, they just really know how to make steel frames and my cross check is like an extension of my body), but considering that the Pugsley is AT LEAST $1500 before it's even electric, I think the Troll will win out. I want to avoid suspension, so big tires are the answer, maybe, and I mean maybe, a suspension fork on the Troll.

Depending on the final build, I think that 48V or thereabouts will be plenty, and that will help keep the cost of the controller down. I also think that range may be a contributing factor to what battery configuration I decide on. Lots of speed seems great now, but I know once I get going at or near 30mph, it will seem crazy fast. I do like the ability of LiPo configurations to evolve over time, and it's nuts to think that 24 hours ago I thought LiFePo4 was the ultimate in battery tech. I've been learning for 3 weeks, and I'm still changing my perspectives entirely every 3-4 days as I read more.

I think that 15Ah is where I'd like to start, simply because my routes and destinations vary from day to day, and I don't want to be discouraged from wandering just because I don't have sufficient battery to let my bike carry me there.

Back on topic: I think that making LiPo packs from smaller units is a great idea, because at the end of the day, you can make a smaller capacity pack with a higher voltage to experiment with different levels of power to your setup (so long as all your components will allow it). It could also be an interesting way to change from long haul to short but quick rides if there's enough flexibility to change up from day to day.
 
I'm really enjoying having a nice big pile of lipo, since I have several bikes. So I can run 48v on one, 72 on another, and so on.

But for the mundane daily ride, where I charge twice a day, I still prefer the charging convenience of my 48v 15 ah pingbattery.

For now at least, charging the lipo is a lot more time consuming and inconvenient. But if my daily ride was shorter, I could easily charge lipo once a day.
 
Back
Top