Mini Tesla continued

[mr kart]

I started this project about a year ago but had no time and money. I am now restarting the project and here's what I'm aiming for:
30mph
Li-ion batteries
Easy to maintain
Not too expensive.

So here's the parts I'm thinking:
2000w 48v brushless motor
2000w 48v brushless ebike controller
Panasonic ncr18650b 3100mah 13s8p pack
Sprint kart (for now)
13s 45a BMS

The Panasonic cells are 6.2a so 8 would be 48a. I'm using the vruzend kit to connect all the cells. I will start with the power train and then put it on something.
Help would be appreciated as this is my first build but I do have some prior knowledge of batteries and motors.
Will keep you updated

 

[Functional Artist]

I started this project about a year ago but had no time and money. I am now restarting the project and here's what I'm aiming for:
30mph
Li-ion batteries
Easy to maintain
Not too expensive.

So here's the parts I'm thinking:
2000w 48v brushless motor
2000w 48v brushless ebike controller
Panasonic ncr18650b 3100mah 13s8p pack
Sprint kart (for now)
13s 45a BMS

The Panasonic cells are 6.2a so 8 would be 48a. I'm using the vruzend kit to connect all the cells. I will start with the power train and then put it on something.
Help would be appreciated as this is my first build but I do have some prior knowledge of batteries and motors.
Will keep you updated

We'll help if we can

What is the "rated" speed (RPM) of your 48V 2,000W motor?
What gear ratio do you plan on using?
What size rear tires?

 

[AustinPeek]

Vruzend Kit Version

Excited to hear about the rest of this build. Did you get your hands on a V2 Vruzend kit or are you planning on pulling all that power through the V1.5 kits?

 

[mr kart]

I haven't got the motor yet, but I think it's about 5400rpm. I've go 11" tyres on the kart and I think a 50-60t sprocket.so gear ratio will be 5-6:1 but I might have to increase that.
I will be using the vruzend v1.5 kit as the batts can only give 6A and I will be using 5A max. If I have a problem I will upgrade but I think it's ok for now. Also, can you replace the stainless steel with nickel strip on the kit? Or are the caps the limiting factor? Thanks for your imput

 

[itsid]

stay AWAY from the panasonics!
they are great cells, for LOW Amp applications, reliable .. great..
but for high amps.. don't even TRY!

Think about getting high performance Sonys (Vtc6?) they are capable of 30 Amps for a few seconds and IIRC 18 until they're depleated.

Frankly 18650 are plenty, but the wrong type of cell anyways.
(38120S are a better size for this application tapped ends allows you to pair them on copper bus bars.. which is sooooo nice and cool and reliable and safe )

NEVER buy cells of ebay until you trust the seller completely..
most 18650 on ebay are fake! be aware!!
And you are up for some serious money in batteries;
one thousand bucks for batteries is something you want to be sure about!

IYAM you should buy 12V motorcycle batteries (LiFePo if you want the weight saving)
and pair these up. (will not be much more expensive but much more reliable, especially since there's no tab welding involved)

vruzend.. never tested.. good idea, but still not a fan in this type of application tbh...

'sid

PS e-bike controllers usually do not have reverse...
So maybe you want to swap that against a kart controller just for convenience sake

 

[AustinPeek]

Response for Tesla

I believe the caps between the V1.5 and V2 kits are the same. It's the bus bars that are the difference. The V1.5 are 7A peak per cell. The V2 are 25A peak per cell.

I do love the concept of LEGO 18650 Kits (plus 18650 would make Elon proud). If you choose the 18650s, I would suggest the LG HG2 cells. I have bought them personally in bulk from China for $2.13 a piece. Excellent Value. I've ran tests on 44 so far and they appear to be authentic.

Though, I would suggest listening to Sid. He is the God of Electric Projects forum.

-Austin

 

[itsid]

authentic new LGs for 2+odd is AWESOME!

But I've seen resleeved (lower rated than original) cells that were likely sent back for disposal,
I've seen the 'famous' 8800mAh cells that are basically empty, and a crap ton of fakes that just
have a LG/SONY/SAMSUNG sticker on it and are essentially crap

IDK, a trusted source is great.. (in china as well as in any other country)
untrusted sources and lithium cells don't work well together...(not in germany, not in the USA and certainly not in china)

Anywho, my concerns about the vruzend kit in this type of application is NOT the interconnection between the cells, it's the caps themselves..
I can't tell how well they can maintain contact under vibration..
and if a cap or two rattles loose, in worst case that means your parallel pack is missing a column,
and the additional amps are drawn from whatever number of columns is still "available"...
you can't see such issue, only when your cells are starting to smolder and cook
(which usually is shortly before they burn in a bright unextinguishable fire)

So a friction fit cap is nothing I'd trust on a kart
(on a much slower road bike...with much fewer cells... perhaps..
on a 104cell kart battery... surely NOT!)

Again, I haven't tested that kit, barely noticed it mentioned before to be honest.
But with 38120s and a piece of solid copper and bolts it's as easy to make a solid connection without spotwelding..
one that's certainly more durable than a friction fitted plastic caps IMHO
just saying
(as a bonus you need much fewer cells since they're way bigger and hold about three times the capacity)
Still expensive, still a ton of work that can easily go wrong (say you don't prelevell the cells properly)
So again.. LiPo motorbike batteries are easy to get hold of and comparable in price,
And if you haven't build a few dozen smaller (say RC sized) batteries from scratch, a full blown multi amp hour kart batpack is certainly not a good idea to start with anyways.

'sid

 

[mr kart]

hi all,
I received the 110 Panasonics that I ordered (sorry Sid but id already bought them before you replied) but i think i can make this work. these cells cost £1.72 ($2.28) each and where taken out of a big li-ion battery for a project but the seller never got round to it.
so i will be using the vruzend kit as it has been working for other people and i will but foam on it to stop vibrations. but i think i might have to use the v2 kit to be safe but ill wait until they're not on kickstarter as i might be cheaper.
should i use cell level fuses? or will that just be annoying and unnecessary?

and also ebike vs scooter controllers, what can they do in terms of regen and reverse-foward ect?

 

[itsid]

Oh my furry shirt.. USED cells??

Don't!
stop now.. !!!
it's nearly impossible to level used cells out,
it's hardly possible you can properly match them up,
and no BMS could fix that either ...
temp fuses are an absolute MUST in your scenario (and you will be throwing money away)
how many? at the very least 25 (one per 4 cell square) in your case..

get rid of the panasonics with as little loss as possible, otherwise it'll be a total loss!

The most dangerous thing in the world will happen eventually;
(it'll work on your initial test) and you will feel confident about the such cobbled battery..
which essentially is nothing but the most scary time bomb you can imagine.
it's is the exact opposite of a safe approach!

Let's put it this way:
I wouldn't even allow you to put such monstrosity in my garage over night...
it may hold a day, maybe a month.. maybe 'till the end of this year...
it WILL BURN eventually, in a 110 cell lithium fire as hot and bright and unextinguishable as a metal fire burns.
And it will take you garage with it, your house in case you store the battery in the basement.

Sorry, if that reads like "panic mode" much..
but it is exactly as save as it would be to store a bucket of gasoline next to your lit fireplace
while a two year old is running around the house with a kitchen knife in his hands unattended.

Nothing will happen at first.. but someone is gonna get hurt badly eventually

'sid

 

[mr kart]

ok, first of all they are not used batteries. They were taken out of a brand new pack and some where tested to prove that. I will further test their capacities myself to confirm.

Second, what do you mean by levelled out?
If they are used batteries or not, they are all the same. As long as the parallels are all the same voltage and preferably the series' to keep everything nice and even though the BMS will keep all series cells at the same voltage.
yes the overall capacity of the cells should be the same when in series (which they are). It doesn't matter in parallel because the cells will be discharged relative to there capacity.

and if you could just explain to me HOW the cells will explode? i know lithium batteries are dangerous but if treated as they should be, they're fine

 

[Functional Artist]

ok, first of all they are not used batteries. They were taken out of a brand new pack and some where tested to prove that. I will further test their capacities myself to confirm.

Second, what do you mean by levelled out?
If they are used batteries or not, they are all the same. As long as the parallels are all the same voltage and preferably the series' to keep everything nice and even though the BMS will keep all series cells at the same voltage.
yes the overall capacity of the cells should be the same when in series (which they are). It doesn't matter in parallel because the cells will be discharged relative to there capacity.

and if you could just explain to me HOW the cells will explode? i know lithium batteries are dangerous but if treated as they should be, they're fine

Probably talkin' "out of my hat here"
...but, I'm a gonna throw my "uneducated" 2-cents in (anyways)

First, their always out of a project that the seller "never got around to"
...check their capacities myself? (x 100) how?

What about the resistances of each cell? (I think that's most important)

Think of each cell as a 2" piece of tube
…& the resistance, of each cell, is the diameter of the tube

Kinda like this
1 Ohm = 1/4" x 2" tube
10 Ohm = 5/16" x 2" tube
50 Ohm = 3/8" x 2" tube
100 Ohm = 7/16" x 2" tube
150 Ohm = 1/2" x 2" tube
200 Ohm = 9/16" x 2" tube
250 Ohm = 5/8" x 2" tube ….etc.


So, if cell #1 has lets say a 10 Ohm resistance (2" x 5/16" tube)
…cell #2 has a 150 Ohm resistance (2" x 2" tube)
...cell #3 has a 1 Ohm resistance (2" x 1/4" tube)
...cell #4 is another one with a 10 Ohm resistance (2" x 5/16" tube)
...cell #5 has a 200 Ohm resistance (2" x 9/16" tube)
...ect...ect...ect. (x 100)

Then, the connectors (wires or bus bars) & even the connections (welds or friction clips) have their own resistances

See the situation?

In this analogy, if someone connected (100) tubes together of varying diameters
…with a bunch of different sized connectors

How well would "water" flow thru it?

Think of a battery pack as (1) unit kinda like a radiator
...you want the water to flow "nice-n-smooth"
...from the beginning spout (positive terminal) all of the way to the outlet spout (negative terminal)

* From what I understand, to explode or "gass out" is usually when charging
...if a few (or even 1) of the cells are "out of balance" the charger can over charge a few (or even 1) of the cells to a point where it "lets go"

Then, adjacent cells start joinin' in & then you got a "real" mess

Here is a video where a Tesla's battery pack is "gassing out"

https://www.youtube.com/watch?v=9VtbBKnG97c&feature=youtu.be

 

[itsid]

All values of a cell are variables.. voltage, capacity and resistance;
that's a bad situation, kinda like Functional Artist's analogy...

So you need to match cells
(meaning at the exact same voltage (ideally 3.2 Volts) they have the exact same internal resistance)
And only such identical cells are to be put into a single battery.

In fact yes..in parallel mode they discharge based on their current level of charge..
And that's your problem..
say you have a bad cell in there and one that's better than most of the others
(slightly less resistance than any other in that column @ the same voltage)
THAT cell will not only be charged first, it's also the one that gets discharged first by the motor.
it sees more load than any other cell.. thus the laod is not spread across the column evenly, the cell will overheat and (if you're lucky gas out; if not ignite)
Now this column is odd and depending on your configuration will cause either the next column to be overloaded or worse it itself is allowed to spread the load across all remaining cells,
not to mention that all cells close to the now dead cell saw some damaging heat from it passing away... and thus be likely affected (and prone to go next)

And that can take as little as one charge cycle!
That's why you need to prelevel all cells!
(should be straight forward with the vruzend kit hopefully)
charge each cell individually to 30% of it's capacity (usually 3.5 Volts or such)
as closely as possible.
now connect ideally three cells (fully interconnected triangles on both ends)
no more than five per if you don't use massive copper bus bars

and let those packs sit for at least 24h.
Then interconnect the sets (again 3 to 5 packs always parallel) and again allow them to level out for no less than 24h.
repeat until you can connect all such packs to one single parallel monstrosity ,
and again allow that to level out for 24h.

Once that's done, disconnect the packs in reverse order and voltage test each set after each step.
if you see an offset of 0.1V between any two packs at any time..
you need to repeat the levelling process

(if any set dropped to less than 3.2Volts you have at least one faulty cell in there)

If you cannot level the cells after the second try, the impedance of one cell is too far off
(can happen with fresh cells from different batches of the same mfg)

any load (charge or discharge) will need to be perfectly spread, or you will
stress the weakest cells the most, and they will 'thank you' violently!

Again, the most dangerous part is, that it will work on initial testing..
long enough to make you feel comfortable and safe even.
untill you find yourself screaming in agony and pain at best just fear.

I'm not keen to be part of that tbh.. so allow me to just step back and watch from now on.

I'm fairly certain you are not eager to change your mind about those panasonics
bad cell for this application, from almost the worst possible source and a connector kit
I personally doubt is up for a task like that;
and frankly I'm rather certain you will not add nearly enough temp fuses or any kind of BMS...

So yeah, no... I'd rather stay off this topic from now on.

'sid

 

[Functional Artist]

All values of a cell are variables.. voltage, capacity and resistance;
that's a bad situation, kinda like Functional Artist's analogy...


(meaning at the exact same voltage (ideally 3.2 Volts) they have the exact same internal resistance

So you need to match cells


...say you have a bad cell in there and one that's better than most of the others
(slightly less resistance than any other in that column @ the same voltage)
THAT cell will not only be charged first, it's also the one that gets discharged first by the motor.

'sid

You didn't like my analogy?

Of course the values are variable
...I mean, the cells must be @ the same value (resistance), when their @ the same value (voltage)

Match cells, Yup!, that's what I said

I guess, my analogy, was more about if many cells of varying resistances (@ the same voltages) were connected
...how it would affect the "flow" thru the battery
.

 

[itsid]

You misunderstood;
the variability of the values is a bad situation.. kinda like you described in your analogy.. [better now?]
it's not perfectly accurate
since the diameter of any tube won't change during the day
current charge and voltage of the cells however WILL
(neither at the same amount nor at the same rate, not even at the same time really)!

but it's still easier to imagine than what happens in battery cells, so it kinda works here.

'sid

 

[mr kart]

I do appreciate you sharing your knowledge and opinions.
I will get a 18650 charger/tester to charge all the cells and test their resistance/capacity. if the cells are all the same in terms of resistance and capacity and are near their 'new' rated values, is it safe to say they are all the same mostly unused cells?
the charger I'm getting is this
https://www.banggood.com/LiitoKala-...6.html?rmmds=search&ID=48182&cur_warehouse=CN

its a cheap but decent analyser that has good reviews. i obviously wont be using it once the battery is complete.

 

[itsid]

Frankly I wouldn't use that one at all..

An Imax B6 is even cheaper (aboutish 15 bucks) and takes care of up to six cells at a time
and while it's meant for RC style batterypacks (XT60 connectors and such)
you can make that work (especially if you already have the vruzend handy you can just use the alligator clips )

Again single cell is HIGHLY recommended..

I can only repeat the Panasonics are NOT suitable for such application!
their low max discharge current is a huge issue,
paired with a relatively low lifecycle (of 500 charging cycles according to Panasonic)
https://na.industrial.panasonic.com/sites/default/pidsa/files/ncr18650a.pdf

They are not even considered SAFE in perfect condition because of that.
you want a maxed out security margin in discharge capability on all unprotected cells...
(so that in case of a quickly resolved short you do not loose any cell or worse health points)
Reason being:
the higher the discharge capability the lower the cell temperature at your desired discharge.

50 AMps are NOT going to be sufficient at all..
Take a look at FAs 48V 1000W motor test for example..
stalling speeds (maxed out laod) it pulls 80 Amps
80!! and it has half the nominal power than your motor..
so extrapolating you might run into a 160Amp situation at one point in time..
(unlikely, but not impossible)

WAY past what the panasonics are able to deliver without overheating!

NO, NO.. NOOOO and you guessed it: NO!

'sid

 

[mr kart]

I have an IMAX B6 as I am in to building my own drones. The reason I'm not using it is because I want to test the cells so I know they are all good.
And did you read FA's post? He pulled 80A without a controller. When he did have a controller max current was 25A. Also that is a burst current, once you've accelerated it's not as big.

 

[itsid]

apparently you didn't otherwise you wouldn't have asked
I know why he drew such amps
(I don't see how the motor survived that other than serious degradation of magnetic field strength)
but I know that a worst case is still within arms reach of Murphy's Law.
I've seen a controller burning since someone swapped two leads (shorting the battery during installation)
I've seen people using subpar connectors shorting batteries against the chassis..

you build your own drones? cool.. GAB707 should be a familiar name then (DRL competitor)
he burned his apaprtement down because of a hiccup in his batpack..

He now stores all (professionally made and perfectly levelled) batpacks in a steel container buried in a 3 feet hole in his backyard
which to me sounds a bit paranoid tbh.. but is not completely unreasonable considering
a single cell overheated and caused him to loose almost all his belongings forcing him to move.

Know what?
Do as you please,
don't come whining when the house burned down because of that battery.

I got it.. you want it cheap quick and easy...
disregarding all safety precautions just to save time and a few bucks.

I said all I had to say;
but keep in mind.. YOU are the only one responsible,
and you have been warned more than once.

Good luck!

'sid

 

[mr kart]

ok, I have got rid of the panasonic batteries (with not much loss).

so what are my options?

I could use:

• lead acid
• lithium ion
• lithium iron phosphate
• lithium polymer

Am I right in saying LiFePo4 is the most expensive then li ion, lipo and then lead acid. I guess this depends on what and where your buying from. I want to try get it as cheap as possible but I dont know how the best way to do that is.
So my current plan is to use the vruzend kit to make a small pack to start with (like 2/3p). I was thinking of trying to rent/borrow a spot welder to make the pack but if I'm going to add more cells in parallel then that would be a pain.
i have found more batteries on ebay they are NEW samsung 30Q 3000mah 15a cells for 2.75 each.
https://www.ebay.co.uk/itm/100-GENU...var=462043987432&_trksid=p2055119.m1438.l2649
let me know what you think.

 

[Functional Artist]

Check out these "monsters"

The VL41M are http://www.houseofbatteries.com/documents/VL41M.pdf
...3.6V 41AH
...~2" diameter & ~9" tall
...even has bolt on connectors (no welding required)

I have seen them go for ~$125.00 for packs of (8)

I don't know anything about them
...just that their out there


or

I have just gotten a 2Kwh section of a Chevy Volt battery for El Moto, my electric motorcycle

…~48V 47AH
…~10" x 10" x 10"
…& only weighs ~45 lbs.
...$400.00 delivered