Tesla owners, squeezing in a 21700 battery pack, possible?

[RagingGrandpa]

13 hours ago, Nomis said:

With my 5a fast charger I got from wheel riders store it seemed to balance charge for a couple of hours (this was after the light on the charger itself turned green) until a steady non-declining watt consumption of about 4 watts. Then I plugged in the standard charger and noticed that it consumed 10 watts of power declining down to 2 watts within an hour or so.

Is this an indication that my fast charger doesn't balance as much/all the way and that my stock charger does?

Not really...

No two chargers will be trimmed to the exact same voltage, so if you saturate your pack using one charger, then immediately change to another charger that has a 0.05V higher output, you'll observe a different current. No cause for concern.

[Mantraguy]

I just posted this on a different thread but thought it might be more in-line with the content of this thread...

I ordered a Tesla V2 1480Wh almost exactly a month ago from the Wheel Riders Store and just got a message from them this morning that it has finally shipped. However, they have confirmed that this is a new update and uses the LG M50T battery which is now 1500Wh instead of 1480. They included these photos:

 

I just wondered if anyone had any experience with these batteries compared to the ones that were shipped with the 1480Wh version. I'm new to EUCs and am looking forward to receiving it but just want to make sure it will be ok with a 5.0A "fast" charger and that these are reliable batteries. Am I correct in assuming that there are likely two, 20S2P packs in this wheel? Thanks in advance for any help you might be able to provide!

[Nomis]

On 9/10/2020 at 7:28 PM, Mantraguy said:

I just posted this on a different thread but thought it might be more in-line with the content of this thread...

I ordered a Tesla V2 1480Wh almost exactly a month ago from the Wheel Riders Store and just got a message from them this morning that it has finally shipped. However, they have confirmed that this is a new update and uses the LG M50T battery which is now 1500Wh instead of 1480. They included these photos:

 

I just wondered if anyone had any experience with these batteries compared to the ones that were shipped with the 1480Wh version. I'm new to EUCs and am looking forward to receiving it but just want to make sure it will be ok with a 5.0A "fast" charger and that these are reliable batteries. Am I correct in assuming that there are likely two, 20S2P packs in this wheel? Thanks in advance for any help you might be able to provide!

Good for you. Now I'm getting worried once again about my "tesla" cells. But they have behaved flawless after 500 km of use with 5A of charging and topping up with the stock charger regularly to hopefully keep it balanced.

 

The M50T cells can safely be charged at 3.4A so in the 20s4p config (yes 20s2p * 2) it can handle 13.6A, thus 5A is not a problem (and the bottleneck is actually the charge port and maybe the BMS to charge at higher than 5A)

[Murdomeek]

Maybe it is just precaution due to the recent gotway nikola fires that have the panasonic 21700a (tesla) batteries

 

The Nikolas have also been updated to have the LG M50t cells

[OneLeg]

Just an FYI, I have configured a MCM5 with 48 x Sany 20700B (4250mah 20A), the side panel is not big enough to take 32

[LaserEdge]

My only concern with the LG M50T 21700 batteries is battery sag. The discharge rating on them is 7.3A. Substantially lower than 10A. It might translate to increased chance for cutoffs.

[Murdomeek]

1 hour ago, LaserEdge said:

My only concern with the LG M50T 21700 batteries is battery sag. The discharge rating on them is 7.3A. Substantially lower than 10A. It might translate to increased chance for cutoffs.

True, but any way you look at it, its less sag compared to the original 1080wh packs

[OneLeg]

For my weight, I find that if am running less than 800A theoretical aggegate discharge capacity I have stability or performance issue.  I convert all my unicycles to 20A batteries (MCM5, V8) at the cost of loss of Whr and range for stabiltiy and having lots of breaking power.

7.3A is not an issue as long as you have 120 or more cells in the pack.

[Nisse977]

Hi!

I have bought a 2nd hand 6 year old Gotway Tesla V1. It still works fine, feels powerful enough for my riding (in town to work) But it needs a modern battery...
I have made a few batteries for diffrent EVs like electric longboards, electric moped and some handicap vehicles.

My first electric longboard had a battery that I made out of Panasonic 18650GA cells, the same model that Tesla V1 (1020Wh) use. In my longboard I made a 12S3P battery and I was pulling a maximum of 40A from the battery. Thats like 12-13A per cell. These 18650GA cells does not work great with that current level, they are pretty weak (but I did not know that when I built that battery) I could see the voltage drop badly during hard acceleration, and spiking during hard braking.

Anyway..

This Tesla V1 are also using the 18650GA cells, but in a 20S4P configuration. When riding I can see in the EUC World App that I´m pulling around 40-45A from the battery at max. And this put a similar load on the cells like in my first longboard, witch results in bad voltage sag and spikes during accelerationg/braking.
Spec says its a 84v wheel and 1020Wh, but its really a 72v battery (nominal voltage) with a 84v fully charged voltage.
The specified capacity states 1020Wh, but is it really 1020Wh?
From test back in 2016, the 18650GA cells put out about 3200mAh at low current test 2-5A. https://lygte-info.dk/review/batteries2012/Sanyo NCR18650GA 3500mAh (Red) UK.html
This means that there is accually only about 921Wh in the pack, not 1020Wh. (3,2Ah x 4) x 72v = 921,6Wh.
But the problem is that the capacity of these cells is not very useful in this application because of heavy voltage sag.. If I ride the wheel down to 50% SoC and hit a uphill road and continue to ride at normal speeds, the voltage will sag down to 15% and the wheel will start beeping and tiltback so I have to lower my speed, therefore I need to always keep the wheel at minimum 60% SoC, but not over 90% SoC, then I get problems with braking caused by voltage spiking over 4,2v per cell... 
This problem would not happen if the battery was made out of high current cells. Or if i ride really slow, but who does that?

So now I am planning to building a new battery with modern high current cells. And from the looks of this thread, there is even a posibility to go for 21700 cells if I rearange some stuff inside the wheel, and cut away one side of the batterycompartments, like in the Mantraguy photos.
I am then thinking about 21700 cells like, Molicel, P42a, or P45B or P50B (expensive) other options are the new Ampace JP40 (or EVE 40PL) but Samsung 50S is a cheaper cell to buy, but not as high current as the others I mentioned, but maybe that would be okey, I dont know.
If I would go for 18650 cells, I guess Molicel P28B or P30B would be ok choices, but they are expensive.
The least expensive cells that I can buy is P42a or 50S.

Have you guys had any luck building better batteries for Tesla V1, or have just left that wheel and go on to newer models?

 

/Nisse977

[bpong]

i have to ask you this Nisse977;

do you object to upgrading your tesla to a recent euc model rather than building and experimenting with your tesla ?  there are a few eucs now available that are marginally heavier than your tesla...

[alcatraz]

6 hours ago, Nisse977 said:

When riding I can see in the EUC World App that I´m pulling around 40-45A from the battery

These are motor amps I've been told. These aren't battery amps. If I remember correctly the batteries experience a bit less.

6 hours ago, Nisse977 said:

This means that there is accually only about 921Wh in the pack, not 1020Wh. (3,2Ah x 4) x 72v = 921,6Wh.

18650GA might be tested ~3400mAh in low current draw but they're listed as 3500mAh. So 80 cells times 3.5Ah times 3.7v nominal voltage equals = 1036Wh. Yeah, that 1100Wh figure is weird.

6 hours ago, Nisse977 said:

Have you guys had any luck building better batteries for Tesla V1, or have just left that wheel and go on to newer models?

Sure the wheel would benefit from higher current cells but the 18650GA have decent current. They're not the worst lets say. Not even close. Those infamous battery fire wheels had LG 10A 21700 cells that tested 7A max in a lab environment. Furthermore they replaced three 18650 cells with two 21700 cells. So if the 18650GA would start to overheat at 10-15A x 3 (45A) the 21700 would overheat when passing 7A x 2 (14A!!!). 

Today you can find really nice 21700 cells like Samsung 50S that aren't very expensive and can do 20-25A (lab environment) while still containing 5000mAh. If you're going to go best of all worlds it's hard to find a better cell.

I'm personally very happy with the Panasonic 18650 in my wheels (I have several, all Panasonic 18650). They're sagging a bit because they're aging. IR is increasing with age. 

So I'd say it isn't the brand or model but that they're old. You just need something new that isn't low current and it'll feel better.

I have a Tesla V1 too and it's done 15000km and still going strong. My MSX is at 16000km. MCM5 at 6000km and Mten3 at 3000km. They all have original packs still. I just added balancing wires to some wheels to ensure all the voltages are perfect.  

[Nisse977]

22 hours ago, bpong said:

i have to ask you this Nisse977;

do you object to upgrading your tesla to a recent euc model rather than building and experimenting with your tesla ?  there are a few eucs now available that are marginally heavier than your tesla...

Hi!

No I am not planning to buy a newer wheel in the near future. But I might do that after a few years. I got my Tesla V1 for 400$, thats a ok price I think I guess Kingsong 16X would be a good similar wheel to buy. Do you have other tips or suggestions?

21 hours ago, alcatraz said:

These are motor amps I've been told. These aren't battery amps. If I remember correctly the batteries experience a bit less.

18650GA might be tested ~3400mAh in low current draw but they're listed as 3500mAh. So 80 cells times 3.5Ah times 3.7v nominal voltage equals = 1036Wh. Yeah, that 1100Wh figure is weird.

Sure the wheel would benefit from higher current cells but the 18650GA have decent current. They're not the worst lets say. Not even close. Those infamous battery fire wheels had LG 10A 21700 cells that tested 7A max in a lab environment. Furthermore they replaced three 18650 cells with two 21700 cells. So if the 18650GA would start to overheat at 10-15A x 3 (45A) the 21700 would overheat when passing 7A x 2 (14A!!!). 

Today you can find really nice 21700 cells like Samsung 50S that aren't very expensive and can do 20-25A (lab environment) while still containing 5000mAh. If you're going to go best of all worlds it's hard to find a better cell.

I'm personally very happy with the Panasonic 18650 in my wheels (I have several, all Panasonic 18650). They're sagging a bit because they're aging. IR is increasing with age. 

So I'd say it isn't the brand or model but that they're old. You just need something new that isn't low current and it'll feel better.

I have a Tesla V1 too and it's done 15000km and still going strong. My MSX is at 16000km. MCM5 at 6000km and Mten3 at 3000km. They all have original packs still. I just added balancing wires to some wheels to ensure all the voltages are perfect.  

Thanks for responding!

That was accual battery amps (the 45A), the motoramps are up to about 90-100A.

Yes the 18650GA was listed as 3,5mAh when they came out, and later on they where relisted as 3,35mAh.
In the test from 2016 when the 18650GA where new, the tester got about 3200mAh out of the cells at low current (2-5A) and the nominal voltage was stated 3,6v in the test, as on most li-ion cells, so 3,6v x 3,2 x 80 = 921Wh. But if we go by the spec 3,5Ah, we get 1008Wh witch is closer to Gotways spec of 1020Wh
I have tested some 18650GA cells myself, and at 1A load (that is really low) I got about 3200-3300mAh out of them. I have tested 36 of them, and that was the avarage capacity

From my experience with diffrent cells, the 18650GA is not a good cell for this application, it sags to much for my taste during load in this 20S4P configuration. But you are right, they are not new, so new ones would probably be better.
In another batterybuild I did, I used even worse cells, but I got the cells totally for free, and the battery is a 660cell 20S33P (yes 33P!) so it works nicely because of all cells in parallel. It´s a battery for a electric moped. Pulling only max about 1A per cell at full load

I think about using Molicel P42A, P45B or Samsung 50S in this batterybuild for the Tesla V1. I prefer to get as low of a voltage-sag as possible in front of getting the maximum range. But the cell-price is of course a factor to think about. I do not want to pay double the price for a cell that is just a little better then another one like the P50B cells...

The price I pay for 80-cells / The real total capacity and real total continuous ampdraw of 20S4P (Mooch test):
316€ Molicel P42a / 1152Wh, 120A
328€ Samsung 50S / 1440Wh, 80A
382€ Ampace JP40 / 1152Wh, 180A
428€ Molicel P45B / 1296Wh, 140A
1116€ Molicel P50B / 1440Wh, 140A (way to expensive cell for me)

To get a 382€ 80-cell pack with Ampace JP40, I need to buy 130pc of them for about 620€....


Nice to hear about your wheels and that they still use the original battery, good stuff 

 

Edited September 11 by Nisse977

[alcatraz]

If the 50S are too expensive check maybe for the Samsung 40T. They were among the first higher current cells offered a few years ago.

In China I've seen 50S for about 2usd per cell but perhaps they were used. 

Edited September 11 by alcatraz

[Nisse977]

22 hours ago, alcatraz said:

If the 50S are too expensive check maybe for the Samsung 40T. They were among the first higher current cells offered a few years ago.

In China I've seen 50S for about 2usd per cell but perhaps they were used. 

No the Samsung 50S is not expensive for me as I listed in my last post. But I do not know if I rather want the Molicel P42a instead, or P45B.
But there is no need to decide on that now, because I have a couple of batterybuilds to finish before taking on this one

 

[alcatraz]

Ok cool.

Here's a tip for you. Upgrade or protect the motor wire connectors on the V1 tesla. They used undersized banana connectors there, like 4mm on the v1. They can overheat as well as the motor wires. One motor phase overheated on mine to the point that the cable insulation turned from green to black and deformed. (And I'm not a power rider, I don't use high current batteries. I stay at max 35km/h. I don't gun it uphill or emergency brake.) 

Keep the three motor wires apart as much as possible. They seem to overheat closer to the control board first and not near the motor.

I got fiber glass heat shield hoses and put one on each wire instead of the pretty stupid original single piece for all three bunched up. 

The  male banana connectors can be widened by bending the segments apart. It can reduce resistance at the connectors. 

This info might be useful to you if you plan to build like a "race tesla". Just saying that even with the stock batteries somewhat aged and not riding like a madman (I'm on the 850Wh NCR18650PF) it can overheat. (I'm 65kg)

[Nisse977]

Ah thanks, really good tip there! 

I am 85kg and usually ride at 45km/h with my Tesla. 
The previus owner was a 100-110kg guy. I think he usually rode it at low speed, but he did ride it att 55km/h once. It was with fully charged battery so everything went good, no accidents. Then after that he read on forums that 55km/h is not a safe speed with the Tesla V1, so after that he kept it to max 45km/h.

[Sholphin02]

Very interesting development  @Nisse977 

if you begin your mod, please post it. I'll be most interested.

Might do the same.  

[RagingGrandpa]

On 9/10/2024 at 5:12 AM, Nisse977 said:

I have a Gotway Tesla V1.
It still works fine
feels powerful enough for my riding

But it needs a modern battery

"Needs" ?
Why?

Your wheel works!
Ride! 

(A theoretical concern about voltage sag for an old machine that has satisfactory operation already, is not a good reason to design a custom pack and BMS...)

Edited September 12 by RagingGrandpa

[Nisse977]

On 9/12/2024 at 9:28 PM, RagingGrandpa said:

"Needs" ?
Why?

Your wheel works!
Ride! 

(A theoretical concern about voltage sag for an old machine that has satisfactory operation already, is not a good reason to design a custom pack and BMS...)

With current battery that "works" I do not get very good range if I dont ride really slow. The voltage sag and spikes are big. 
Like yesterday I was out on a 17km ride and the battery fluctuated between 93% and 17% during that short ride.

Look here: https://euc.world/tour/637491240964064

So with a new battery I hope to get less voltage fluctuations and better range 

[alcatraz]

6 minutes ago, Nisse977 said:

With current battery that "works" I do not get very good range if I dont ride really slow. The voltage sag and spikes are big. 
Like yesterday I was out on a 17km ride and the battery fluctuated between 93% and 17% during that short ride.

Look here: https://euc.world/tour/637491240964064

So with a new battery I hope to get less voltage fluctuations and better range 

I'm curious. Do your packs charge fine all the way to 84.0v?

If you have a cell group or two that are dead it could explain the fluctuations.

[Nisse977]

54 minutes ago, alcatraz said:

I'm curious. Do your packs charge fine all the way to 84.0v?

If you have a cell group or two that are dead it could explain the fluctuations.

Yes it accually charges just fine to 84v so I do not believe I have unbalanced cells, but who knows.
 

[RagingGrandpa]

12 hours ago, Nisse977 said:

With current battery I do not get very good range
The voltage sag and spikes are big. 
Look here: https://euc.world/tour/637491240964064

Your recording began at 81V, not full. You rode 10 miles fast and used 50% SOC. So I expect your maximum range for a full battery is 20 miles cruising at 25mph. That's great for 1100wh! 

12 hours ago, Nisse977 said:

with a new battery I hope to get better range

Can't you find a used 1800wh EUC? 
Even the 2700wh EX.N is dirt cheap now... 

[alcatraz]

It's odd because my Tesla v1 that has 15000+km on its original 850wh packs (NCR18650PF) doesn't come anywhere near that min safety margin. Hmm...

[Nisse977]

10 hours ago, RagingGrandpa said:

Your recording began at 81V, not full. You rode 10 miles fast and used 50% SOC. So I expect your maximum range for a full battery is 20 miles cruising at 25mph. That's great for 1100wh! 

Can't you find a used 1800wh EUC? 
Even the 2700wh EX.N is dirt cheap now... 

Yes I did not charge it to 100% at that particular ride. But still big fluctuations between 93% and 17% during the ride.
I normaly don´t charge to 100% because I live on a hight, so almost everywhere I want to go, its starts with downhill. I charge to 90% most of the time. But I have tried 100% to  

Edited October 1 by Nisse977

[alcatraz]

How sure are you that your cells are balanced? The balancing currents provided by the stock bms' aren't designed to take out larger imbalances so it's important to charge to 100% frequently. Especially with older packs.

It will impact your safety margin because the voltage sag will be greater with unbalanced cells. Let's say you have most groups at 3.7v and two groups at 3.5v. When you accelerate the higher groups keep up but the lower two dip down below 3.0v.