Battery options for electric kart

Upperfoot

100 W
Joined
Jan 7, 2018
Messages
123
Hello guys,

Just posting this as I've been going back and forth now for a while weighing up the pros and cons of different battery chemistries and formats to meet the needs of an electric kart.

LiPO for my use case is the best fit, its high discharge, high energy density, low sag, flat discharge curves, low weight and cycle life is good if properly cared for, but the problem of safety and high maintenance presents a problem.

Now safety I can solve by ensuring I use a reputable BMS and charging system to prevent over-discharge, under and over voltage and to have a protective shell around the cells.

The problem of maintenance comes in with storage voltage, LiPO suffers from degradation if stored at 100% and cycle life and performance over the lifetime of the cells decreases as a result, is there something I can do to combat this but keeping the simplicity of charging/storage?

Now I've looked at alternatives, 18650 lithium cells such as the Sony VTC5A and Samsung 20S look good except the steep discharge curve and voltage drop (Samsung 20S drops from 4.2 to 3.3v over the lifetime of the discharge) which is a major issue for performance and is something I can't seem to get around other than requiring them to be topped up constantly to prevent it during race events.

LifePo4 is another alternative but the extra weight required because of the lower voltage is an issue for a sport in which weight is a major component, but is a much safer chemistry.

Looking forward to hearing your opinions and help!
 
What are your power requirements, and physical limitations?
So far i dont see a question/problem to solve without having actual usable numbers.
 
To be specific LiPo is an issue because of maintenance with storage voltage, LifePo4 is an issue because of extra weight and cost, 18650 has an issue with the steep discharge curve.

I'm looking for ways to mitigate those problems so I can evaluate my options with a better understanding.

If you want numbers

750 amps peak discharge
100v plus
32-40AH
36KG weight target

Now I achieved that with 24S6P configuration of 6S Turnigy Graphene at 48AH, I'm looking at alternatives that are easier to maintain through a BMS
 
what are your controller limits in voltage?

getting the maximum voltage means you notice less amp losses and the amps are lower in general.

you can also use a controller that is "power aware" like a sevcon so you can enact limits based on kW instead of amps (or both) that should even out the performance during discharge. but using a sevcon is "cost prohibitive" to say the least.
but cranking the voltage should be priority one, so how high can you go or how high are you willing to go? have you already bought the controller and/or motor?
 
flippy said:
what are your controller limits in voltage?

getting the maximum voltage means you notice less amp losses and the amps are lower in general.

you can also use a controller that is "power aware" like a sevcon so you can enact limits based on kW instead of amps (or both) that should even out the performance during discharge. but using a sevcon is "cost prohibitive" to say the least.
but cranking the voltage should be priority one, so how high can you go or how high are you willing to go? have you already bought the controller and/or motor?

I already have sevcon gen4 size6, and my kart already runs on it, which I've already raced.
 
Do one better, interview about it at 8 minutes, racing in the wet

[youtube]40a3hsxC16E[/youtube]
 
that is a realy nice toy.

what part number sevcon do you have? that makes it more clear in what your limits are.
i have a couple size 4 110V here but that can do "only" 300A.

if you are looking into replacing the pouches for something more robust and simper to charge its fairly easy to do, do you want to keep the same battery box footprint you have now? if so, what is the size of the box?
 
flippy said:
that is a realy nice toy.

what part number sevcon do you have? that makes it more clear in what your limits are.
i have a couple size 4 110V here but that can do "only" 300A.

if you are looking into replacing the pouches for something more robust and simper to charge its fairly easy to do, do you want to keep the same battery box footprint you have now? if so, what is the size of the box?

Sevcon Gen 4 Size 6, 80v good for 120v with the new firmware update, 10s burst is 660a. The motor is an ME1616.

The size of the box is 595mm x 265mm x 240mm and with all 24 batteries weighs 36KG

To get back to specifics, what I am looking for ideally is an upgrade to what I have now to include simplicty and longevity.

I have plenty of power from the batteries, I'm looking at simplifying the charging system but cannot get past having to discharge the LiPo's to storage voltage, so I'm looking at either swapping the chemistry to something were that isnt a concern, or incorporate a system to discharge the battery to storage voltage.

Also longevity is a concern.
 
first i would crank it up it 28S, so 117V fully charged.

for a 30kg battery made from 18650 cells (generously 6kg for the box) you can have about 650~700 cells (roughly) taking 6kg for the housing.
for the current rating you need 30A per cell for the peak bursts of 660 amps to give you 80kW (~110bhp).

in order to hit 30A realiably and get reasonable lifespan you need lower capacity and higher current rating. so you go into the samsung 25E for example, still 2500mAh but it can take 15A continous discharge and higher peaks without much effort. they are also used in a lot of power tools but they do sag a lot under 30A loads.
that would give you a 62Ah battery at 100V, so call it 6.2kWh of total capacity versus your current 4.1kWh

i would recommend using a bigger/more heavy/larger battery or reconfigure the kart to possibly hold a second battery to keep the weight distributed right.
doing so will make a considerably longer living battery and one that will not sag under such heavy loads. personally i would like to stay under 15A peak per cell and under 10A sustained if you want good lifespan. so either more cells or ones that have less capacity and more current rating. smaller capacity cells also means they taper off faster and you are still stuck with the current problem of low power when empty. having a bigger battery means you can stay at the top voltage/current longer and you can recharge faster/more power between sessions as a bigger battery can take much higher charging currents.

once you have the battery sorted it is very simple to have a "discharge box" to get the battery down to a storage voltage of 100V or have a storage charger.
charging such a battery is also very simple and is only limited by what power you can get from a plug, if you so desire you can get a 3 phase 5kW fast charger if they have that on the track so i would not worry about that yet.

first is the question of what you see as reasonable lifespan and what actual power you need during a race. the higher capaciy i mentioned is not for actual more capacity but to give more power for longer.

i think that a lot of issues can be fixed with more capacity and reasonable cells like the 25R instead of using smaller capacity cells with more current rating.

if you allow for a bit more weight so you can get 840 cells for a 28S30P battery. that would give you 75Ah (7.5kWh, nearly double what you have now), wich would give you a maximum charge rating of 12kW. so you are truly limted by what charger you are preparing to use between races. discharge currents would be within spec at 20A or so peaks. so the cells will not go out of spec under peak loads.
this way you only need 1 proper bms to control it all and read it from a lcd to inform you of voltages, temps and all that good stuff.
and next to that 1 or 2 chargers for either normal charging before races and one quick charger for between races as that will shorten the lifespan but gives maximum performance.
and then 1 discharge box that drains the battery down to 100V after a race for storage.
such a battery would last many years and give a much flatter discharge curve compared to anything you have now with those bomb-bags you are using now.
 
Just don't charge it after running it. Only charge just before you go out. You should be able to keep the cells at a good storage level that way.
 
Stay with the lipo, it is the best combo of power, capacity, weight, size, ...and cost.
......The problem of maintenance comes in with storage voltage, LiPO suffers from degradation if stored at 100% and cycle life and performance over the lifetime of the cells decreases as a result, is there something I can do to combat this but keeping the simplicity of charging/storage?
As Fechter said... just dont store it straight after charging.
Give it a lap or two to burn off the cells voltage below 4.0 v before putting it back in the trailer, and you will be ok.
Make sure there is no parasitic drain from any ancilaries or the bms etc, during storage.
 
Upperfoot said:
Hello guys,

Just posting this as I've been going back and forth now for a while weighing up the pros and cons of different battery chemistries and formats to meet the needs of an electric kart.

LiPO for my use case is the best fit, its high discharge, high energy density, low sag, flat discharge curves, low weight and cycle life is good if properly cared for, but the problem of safety and high maintenance presents a problem.

Now safety I can solve by ensuring I use a reputable BMS and charging system to prevent over-discharge, under and over voltage and to have a protective shell around the cells.

The problem of maintenance comes in with storage voltage, LiPO suffers from degradation if stored at 100% and cycle life and performance over the lifetime of the cells decreases as a result, is there something I can do to combat this but keeping the simplicity of charging/storage?

Now I've looked at alternatives, 18650 lithium cells such as the Sony VTC5A and Samsung 20S look good except the steep discharge curve and voltage drop (Samsung 20S drops from 4.2 to 3.3v over the lifetime of the discharge) which is a major issue for performance and is something I can't seem to get around other than requiring them to be topped up constantly to prevent it during race events.

LifePo4 is another alternative but the extra weight required because of the lower voltage is an issue for a sport in which weight is a major component, but is a much safer chemistry.

Looking forward to hearing your opinions and help!
I usually I don't recommend LiPoly, especially for use in a daily ride, but for a cart. it seems a good fit.
It's not for safety reasons, I use low C rate, hence, low volatility, but for the time it requires. You touched on it, the need to first charge it to storage value, and then, right before going out, put the top charge on .
I've been using it on ebikes for about 8 years and I've gone/going thru different charging techniques. They are;
1)Break the string and use common (cheap) RC balance chargers.
2)Leave the string together and bulk charge using a good CC/CV power supply and ck/balance w/ a passive device(Battery Medics).
3)Break the string and use a quality RC balance charger.
The main problems w/ each are;
1)Toy chargers fail, usually w/in a year. Plus they only go to 6S, limiting a 2 brick series string to 12S.
2)Not bad. The Mean Well HGL series of LED Power supplies go up to 60 Volts and they are reasonable in cost, are bullet-proof, and can be had with some good power. The problem comes when the LiPoly has a year on it and the cells are starting to stray during the top charge. It wouldn't be so bad if it happened during the storage charge, when you have all day, but it only seems to happen in the last 4.00V to 4.20V range when you want to go. And Battery Medics are sloooooow.
3)The best once you get over the shock of buying a Icharger Duo and PS. But remember how low the cost of LiPoly is compared to other chemistries(especially when bought on one of Hobbyking's increasingly rare sales).
In general, here are the things I've learned about using LiPoly on E-bikes;
1) Stay at 1P/2S or 2P/2S configuration and use the biggest bricks you need(can get) to get where you want to be. By keeping the pack simple, the easier it is to break it to charge. More than 4 bricks at a time gets tedious w/ connectors, balance leads, etc. And there are just too many damn wires. And since the Duo goes to 8S on each side, the pack can be 16S.
I top charge to 3.90V, or if I know I'm going out on the morrow, 3.95V. Putting on the top charge doesn't take that long and since I don't have to fuss w/ the passive balancers, I can gather up my gear and ck the air in the tires and shock while they are charging.
 
Thanks guys for the input, lots to re-evaluate, if I'm selling these packs on I think high rate 18650 is the best choice for my own piece of mind as they are arguably a safer format.

A few comparisons against would be;

  • Turnigy Graphene 8AH 6S 15C - 120 amps cont - 1.1kg - £90 (pack I use now for comparison)
  • 6S4P Samsung 20S 18650 (48g, 2AH, 30a) 8AH - 120 amps cont - 1.1kg - £60
  • 7S2P LifePo4 Vapcell IFR26650 (90g, 2.6AH, 55a) 5.2AH - 110 amps cont - 1.2kg - £100
  • 7S3P A123 LifePo4 26650 (78g, 2.5AH, 40a) - 7.5AH - 120 amps cont - 1.6kg - £90

I will be using the Orion JR bms on the pack, and using an Eltek charger alongside it (they both communicate together via CAN)

The idea is two packs connected in series both with their own bms
 
Just had a thought, I could just have the charger and bms as one unit, less weight and less complex battery system and all I need is a connector that will integrate all 16 bms leads and main power.

My controller handles under and over voltage and current so I don't exactly need to double up on it, and it stops the parasitic draw from the bms being part of the battery :)
 
One of the biggest safety risk with any pack is in the quality of the assembly, connections, wiring , etc (The human factor) together with poor charging / maintenance practice.
A wiring short on any pack can result in a major incident..even 18650/26650 and LiFepo4 cells.
One advantage of the Turnigy lipo packs, is ease of connection and reduced number of connections to make up a large pack,...reducing the human error possibility
Somewhere there is a post with videos of tests for various cells and chemistry’s (i will try to find it)
It shows a huge variation between even lipo pouch packs..some almost exploding, others doing no more than smoking.
I understand your thought train, but i suspect you are overestimating the risk with lipo.
PS...remember if you do change to those 18650 cells, whilst they can deliver the Amps, you will see a lot more voltage sag, and voltage drop during discharge.
 
they sold integrated bms/chargers combo's with massive plugs on electric scooters and stuff. and they are crap. every company that sold them either went out of business or went bankrupt due to warrany claims.

considering the amp ratings i would recommend getting someone fairly close to you to build the battery for you as mistakes would get "exciting". welding .2 nickel the proper way so it survives a kart track is not something i recommend doing as a beginner.
a proper sized battery like this will basically never drain due to bms parasitic drains. it would take decades to do that.
even if you go with a bigger pouch pack and hook a bms to that and a single charger you need to make proper connections to make it safe and reliable.
a bms weighs less then 200 grams, so that is not really a thing/reason while lugging around a block of semi-solid copper behind you. the plug neede to make a reliable connection weighs more then the bms itself.
 
We use two separate 14S/20P modules of LG HG2 cells ("18650") connected in series for 51.8Vs nominal. Paired with an AC-9 Alternating Current motor and a Curtis 1236SE controller, this makes for a fun ride. Check it out.... https://www.facebook.com/VenomKarts/
 
that setup is good for 300A peak, how do you deal with the heat from those cells under high loads?
 
Upperfoot said:
I will be using the Orion JR bms on the pack, and using an Eltek charger alongside it (they both communicate together via CAN)The idea is two packs connected in series both with their own bms
please dont use canbus for this. its totally not needed for the charger to talk to the bms in applications like this.
the bms will stop the charging automagically if something goes out of spec with the mosfets it already has on board.
adding a extra complication that is canbus on this makes it too complex for such a simple process. it also utterly kills any way to change stuff or get the charger going when you have canbus issues or change your bms or charger. its also adds cost.
 
flippy said:
Upperfoot said:
I will be using the Orion JR bms on the pack, and using an Eltek charger alongside it (they both communicate together via CAN)The idea is two packs connected in series both with their own bms
please dont use canbus for this. its totally not needed for the charger to talk to the bms in applications like this.
the bms will stop the charging automagically if something goes out of spec with the mosfets it already has on board.
adding a extra complication that is canbus on this makes it too complex for such a simple process. it also utterly kills any way to change stuff or get the charger going when you have canbus issues or change your bms or charger. its also adds cost.

Its an extra layer of connectivity, nothing to be afraid of.
 
Upperfoot said:
Its an extra layer of connectivity, nothing to be afraid of.
i am not afraid, but it is a extral layer of complications and cost that is simply not needed. there is no need for a matched system to have the bms and charger to talk to eachother in the first place. they dont have anything to say to each other.
if you have different batteries with different voltages it is a great solution. but not for a setup like this.
 
Back
Top