21700 cells comparison

[fred8569]

Hello everyone,

 

For those who don't know, 1rad is going to release his own upgraded model of the Nikola 100v,

 

I think he made his own batteries using INR-21700-50E cells from Samsung (which are, I guess, different from what there are in the original and new Nikola models - I don't know yet the reference of the cells)

 

Anyone to discuss the pros and cons of those different cells to help me make my choice ☺️ ?

Edited November 27, 2019 by fred8569

[Chriull]

40 minutes ago, fred8569 said:

Anyone to discuss the pros and cons of those different cells to help me make my choice ☺️ ?

 

The main difference between 21700 and 18650 cells is their size.

As the number of cells in series has to be the same to reach the 100V, with the 21700 cells the number of parallel cells will be less.

So the drawn current per cell for the 21700 will be higher for the 21700 as for the 18650.

Higher drawn current per cell means higher voltage sag (if the internal resistance is comparable) and more "lost" capacity. If you look at discharge graphs from for example dampfakkus.de you see this capacity loss.

I was quite excited from the data of the INR21700-3(4)0T - very low internal resistance and low capacity loss at high currents! Just their nominal capacity is not too high....

The INR-21700-50E seems to have comparable internal resistance as the normally used 18650 cells, but higher capacity...

So depending on the specific battery config for these cells the INR-21700-50E could maybe compete with the 18650? But quite impossible to say without knowing the exact configs and especially having no discharge graphs. Even with knowing this it will still be not easy as the burdens of an EUC are very dynamic and depend on enviroment and personal driving style...

From the datasheet i found for this cell and if the 21700 and 18650 config have about the same capacity i think i'd prefer the 18650 version... But that's an very unqualified guess with much gut feeling involved...

Maybe best to discuss this directly with chris from 1rad - he's very reasonable and experienced in this topic!

[Mike Sacristan]

It is already available. 1radwerkstatt know a thing or two about batteries so I am sure their choice is well thought out. They have been doing custom packs for a few years now.

https://www.1radwerkstatt.de/epages/80603321.sf/de_DE/?ObjectPath=/Shops/80603321/Products/GotWay_Nikola

One guy received his 3 days ago. My friend is second in line and is awaiting his.
The weight is around 28.5kg and it's a 24s6p config just like the 1845Wh 18650 version.

Can also be ordered with Sanyo 4250mah 15Amp vs 10 amp (2260wh).

2664Wh battery pack from 1RadWerkstatt with 21700 cells and SMART BMS

(From the battery, all data can be read out via Bluetooth: voltages (also from the individual cells), current (charging current or consumption while driving), number of cycles, remaining capacity, remaining range (calculated from actual consumption and actual distance traveled via GPS) - comprehensive Datalogging)
- App available for iPhone and Android

- Fail-safe and with all protection circuits

- Charging possibility for 8A / limited by the charging plug connection

[Chriull]

1 hour ago, Mike Sacristan said:

...SMART BMS

(From the battery, all data can be read out via Bluetooth: voltages (also from the individual cells), current (charging current or consumption while driving), number of cycles, remaining capacity, remaining range (calculated from actual consumption and actual distance traveled via GPS) - comprehensive Datalogging)
- App available for iPhone and Android

- Fail-safe and with all protection circuits

- Charging possibility for 8A / limited by the charging plug connection

That sounds really great!

.... maybe a smart bms nikola+ for next spring...

 

Edited November 27, 2019 by Chriull

[fred8569]

5 hours ago, Chriull said:

 

The main difference between 21700 and 18650 cells is their size.

As the number of cells in series has to be the same to reach the 100V, with the 21700 cells the number of parallel cells will be less.

So the drawn current per cell for the 21700 will be higher for the 21700 as for the 18650.

Higher drawn current per cell means higher voltage sag (if the internal resistance is comparable) and more "lost" capacity. If you look at discharge graphs from for example dampfakkus.de you see this capacity loss.

I was quite excited from the data of the INR21700-3(4)0T - very low internal resistance and low capacity loss at high currents! Just their nominal capacity is not too high....

The INR-21700-50E seems to have comparable internal resistance as the normally used 18650 cells, but higher capacity...

So depending on the specific battery config for these cells the INR-21700-50E could maybe compete with the 18650? But quite impossible to say without knowing the exact configs and especially having no discharge graphs. Even with knowing this it will still be not easy as the burdens of an EUC are very dynamic and depend on enviroment and personal driving style...

From the datasheet i found for this cell and if the 21700 and 18650 config have about the same capacity i think i'd prefer the 18650 version... But that's an very unqualified guess with much gut feeling involved...

Maybe best to discuss this directly with chris from 1rad - he's very reasonable and experienced in this topic!

Another comparison I was interested in was between the original 21700 cells from gotway in their new Nikola 1800Wh and those used by 1rad (if there's enough elements from gotway)

[Chriull]

38 minutes ago, fred8569 said:

Another comparison I was interested in was between the original 21700 cells from gotway in their new Nikola 1800Wh and those used by 1rad (if there's enough elements from gotway)

I haven't heard anything specific about the 21700 cells used by GW till now

Afaik chris i can't believe he'd take anything worse - alone his new smart BMS sounds like a huge leap forward!

[FreeRide]

Does anyone one here besides Afeez Kay have the Gotway Nikola 21700? 

[fred8569]

I've finally found the reference of the OEM

 

The batteries are groups of NCR 21700A.

 

So we are taking about Panasonic 18650 GA vs Panasonic NCR 21700A vs Samsung INR-21700-50E

[Chriull]

8 hours ago, fred8569 said:

I've finally found the reference of the OEM

 

The batteries are groups of NCR 21700A.

Not much to be found with a quick google search. Seems to be quite new/not to much available at the market (for private?)

Just some sites specify/advertise 15A.

8 hours ago, fred8569 said:

 

So we are taking about Panasonic 18650 GA vs Panasonic NCR 21700A vs Samsung INR-21700-50E

The 50E als has 15A specified as max non continous current.

Seems the wheels with these cells are/will hit the market before these cell tests become ready...

[Mike Sacristan]

There was a discussion regarding the 50e on Facebook.https://m.facebook.com/story.php?story_fbid=1640872466053955&id=527048414103038
 

Eddie Moy: Those batteries suck. They are just high capacity, but performance sucks.

Damilola Alade: Eddie Moy which will you recommend.. money no object. Thanks 

Eddie Moy Damilola Alade: the ones that can do 40amps continuous. But the capacity is lower. Haven’t seen both high capacity and high drain. I like to see those used in the higher 126V versions as you won’t need that many parallel packs.

1Rad Werkstatt: Eddie Moy this is wrong
high drain is not needed in EUC application!
This is slow discharge device and motor power Limits primary amps (3000W/90V=33A MAX / 6Cells = 5,5A MAX drain single cell - and this is MAX and short, not continous) amps not all😄

 

[Chriull]

16 minutes ago, Mike Sacristan said:

There was a discussion regarding the 50e on Facebook.https://m.facebook.com/story.php?story_fbid=1640872466053955&id=527048414103038

There are nice details for the smart BMS, too:

"current drain from Mainboard readout":

... so these nominal cell current questions could be backed up with real measurements

but most importantly:

"balance works active in charge, not only at charge end (you can charge every time to 80%/90% and all is balanced)"

and

"voltage from each cell"

so one sees once the pack gets old/cells aged too much and it could get more risky!

16 minutes ago, Mike Sacristan said:

Eddie Moy: Those batteries suck. They are just high capacity, but performance sucks.

Damilola Alade: Eddie Moy which will you recommend.. money no object. Thanks 

Eddie Moy Damilola Alade: the ones that can do 40amps continuous. But the capacity is lower. Haven’t seen both high capacity and high drain. I like to see those used in the higher 126V versions as you won’t need that many parallel packs.

1Rad Werkstatt: Eddie Moy this is wrong
high drain is not needed in EUC application!
This is slow discharge device and motor power Limits primary amps (3000W/90V=33A MAX / 6Cells = 5,5A MAX drain single cell - and this is MAX and short, not continous) amps not all😄

 

+1.

Although already the "MSuper V3+ 1600Wh" had an mechanical max output power of about 4.6 kWh (https://airwheel.ru/2017/10/27/test-monokoles-na-dinostende/ - could eventually be not the most accurate measurement, but the range should be about right).

With all the efficiencies regarded (motor, motor driver, ...) still more than this is needed to be supplied from the batteries!

I'd assume the Nikola should be at least in this range, most probable above this.

But as the cells are all rated for 10A continous current there are 6 x 10A = 60A at battery side available. Making 90V*60A = 5.4kW battery output power available!

With the 15A that would be 90V * 6 * 15A = 8.1kW!

For peak burdens! Imho if this power is "used" for a bit longer times some mosfets/wires/connectors will burn anyhow - or in the best case the overheat alarm will trigger.

... and these 10/15A are "just" the nominal recommended currents from the manufacturer. For real peak "emergency" burdens they easily can deliver 2-3 times this values.

If one compares this - in normal european households with power outlets limited to 16A (fuses) the max power to get for one device (induction plates, stove, boiler...) with three phases is 3 * 230V * 16A = 11 kW.

So, yes - i'd have to agree fully with chris from 1rad - 6p 10/(15)A are (more than) enough for EUC power supply!

Some differences could be in the usable capacity between the different cells. As for example for the NCR18650GA with nominal 3300mAh with 2A from 4.2V downto 3.3V one gets 2529mAh, with 5A 1800 mAh and with 10A ~1200mAh (http://www.dampfakkus.de/akkutest.php?id=609, The official NCR18650GA datasheet gives better results - has nice discharge graphs for 2,4,6,8,10A included!)

[alcatraz]

In China there are two cheap high capacity 21700 cells. The Lg m50 and tesla refurbished. Both rated 5000mAh.

Tesla I suspect are Sanyo/Panasonic but I don't know.

[Mike Sacristan]

4 hours ago, Chriull said:

+1.

Although already the "MSuper V3+ 1600Wh" had an mechanical max output power of about 4.6 kWh (https://airwheel.ru/2017/10/27/test-monokoles-na-dinostende/ - could eventually be not the most accurate measurement, but the range should be about right).

With all the efficiencies regarded (motor, motor driver, ...) still more than this is needed to be supplied from the batteries!

I'd assume the Nikola should be at least in this range, most probable above this.

But as the cells are all rated for 10A continous current there are 6 x 10A = 60A at battery side available. Making 90V*60A = 5.4kW battery output power available!

With the 15A that would be 90V * 6 * 15A = 8.1kW!

For peak burdens! Imho if this power is "used" for a bit longer times some mosfets/wires/connectors will burn anyhow - or in the best case the overheat alarm will trigger.

... and these 10/15A are "just" the nominal recommended currents from the manufacturer. For real peak "emergency" burdens they easily can deliver 2-3 times this values.

If one compares this - in normal european households with power outlets limited to 16A (fuses) the max power to get for one device (induction plates, stove, boiler...) with three phases is 3 * 230V * 16A = 11 kW.

So, yes - i'd have to agree fully with chris from 1rad - 6p 10/(15)A are (more than) enough for EUC power supply!

Some differences could be in the usable capacity between the different cells. As for example for the NCR18650GA with nominal 3300mAh with 2A from 4.2V downto 3.3V one gets 2529mAh, with 5A 1800 mAh and with 10A ~1200mAh (http://www.dampfakkus.de/akkutest.php?id=609, The official NCR18650GA datasheet gives better results - has nice discharge graphs for 2,4,6,8,10A included!)

Thank you for this analysis. I have done similar research and have come to similar conclusions but it is comforting to read your numbers and thoughts.


 

[Kristof Willen]

21 hours ago, FreeRide said:

Does anyone one here besides Afeez Kay have the Gotway Nikola 21700? 

I ordered one and received it last Friday.  Bad weather here, so not much mileage yet...

[Chriull]

2 hours ago, Kristof Willen said:

I ordered one and received it last Friday.  Bad weather here, so not much mileage yet...

Just out of curiosity - the GW 21700 or the 1rad? Or any other? Which config?

Looking forward to reviews!

[alcatraz]

Best is to wait for test results. Look for reviews.

Cells can be specced for 15A but during testing they might run too hot or voltage drop too far. Inversely some cells rated at 7A could be fine at 10A.

There are fine differences between cells that will affect the feel of a wheel.

[Chriull]

9 hours ago, alcatraz said:

Best is to wait for test results. Look for reviews.

+1. For the abovementioned LG MG50 21700 i found in some forum discharge graphs that showed that a firmware should be changed to stop riding from the by now 3.3V to ~3V or almost half very very much of the great capacity is lost at a bit higher discharge rates.

But again - no LG cells are used, no "tustworthy" discharge graphs are given and no real max/average battery burdens are known...

So some only real test result would be by now a rider as @Marty Backe doing his mountain ride and stating a (much) different or about the same capacity usage!

Edited November 29, 2019 by Chriull

[Kristof Willen]

9 hours ago, Chriull said:

Just out of curiosity - the GW 21700 or the 1rad? Or any other? Which config?

Looking forward to reviews!

It's the default 100V Nikola+ from Gotway, but with the 21700 cells, so delivering a 1800Wh. 

You can read my review journal here :