Sanyo NCR 18650GA

this one is fairly old (2014), but currently the only thing we got:

http://img4.hostingpics.net/pics/668723esboardevaloct14.jpg

someone should start a thread about this and collect data properly. with enough and properly verified data, one could maybe have a hit on quite a few interesting topics: single vs. double motor, 63 vs 50mm, foc vs bldc, vesc vs. other escs, impact of KV values, impact of wheel choices and and and!

I do agree about creating a new thread about this.

I know this opinion is vastly unpopular, but I think the amp requirement for e-boards are ‘‘overengineered’’. And for everyday use you don’t need these 30A continous.


From the chart alone, it tells that it takes around 100w of energy to go 20kph.

That would equal to ~4.5A for 24v (6s), ~2.7A for 36v (10s) and ~2.4A for 44v (12s) system.

As about the consumtion climbing hills, I think someone mentioned it could take around 20-30A. But, as the place I’m living basically has no hills, I would count something like 15A for 6s/22.2v (333w), 9A for 10s and 8A for 12s. That should suffice to climb some streets with a slope and also for acceleration.


So I think it comes to the fact, that the end user / battery pack builder should really consider what his power requirements and use would be like.

well, first of all, dont follow that quadratic air drag line - you can see that most boards level out at 200watt+. you have the belt that costs energy, the deformation of wheels - obviously none of these builds were really getting below 200Watt.

I have done a quite constant speed 23kph average run with a vesc foc driven single motor and it took me 9Wh/km. using the 23kph average, Im getting a 207Watt average power consumption as well (thats 1.5A per cell) - riding conservative and without changing my velocity all that much.

a 27kph average run gave me 13Wh/km, thats 300watt already (thats 2A/cell)!

but now comes the important point: thats on average!

I know for certain that I can easily hit 60A battery current with just a single motor build - for my 10S battery thats more than 2000Watt of power, that has to be kept up for a few seconds at least (for my battery thats 15A per cell - LG HG2 rated 20A continuous). I recently pushed my vanguard really hard - basically full throttle ↔ full braking and my battery, thats already over-engineered for 80A continuous current, was slightly more warm than my hand (through enclosure and felt sheet)!!

nothing is more deadly for a lipo and liion battery than getting warm, so here goes your overengineering. peak loads can simply be much much higher than these averages!

as a test, why dont you limit your vesc battery current to 20A and see how much fun youll have with that! :smiley:

1 Like

Ok, You got a good point there, I agree. The temperature of battery should not be very high.

Although, I think it once again comes to the point of what kind of board and for what use it is going to be. If you want a racer board, then, yeah, those 60A, will perhaps weakly sustain your energy needs.

But if you are on a ‘‘budget’’ and don’t plan to go speed racing with over 25kph (~15mph) and just cruise at 15-20kph (~10mph), then you can live off with lower draw battery setup.

I would like to quote the ride time / distance thread, where people usually have 40+ minutes of ride time, if they were going at 800w all the time and their batteries are in 100-200wh range, then they would not be able to do so.

a thread like that would be great, but it needs a knowledgeable moderator to put the data into a context that allows comparisons and to set the standards to enter the database to begin with.

ride time thread is a good example for something that isnt very useful in my eyes. :neutral_face:

I actually found it is already made!

I think there are a couple of other posts related to this question.

Should have searched earlier but Im glad I did it know :slight_smile:

Standards would be great. It seems people tend not to spend very much info on at what speed/wheels/gear ratio they are using.

not much info in that linked thread, just a few random peak and average numbers without any context. :confused:

to make this properly, youd have to create a reasonable template that people have to fill out before entering an organized database.

Ok, scrolled till the end and have to conclude the same


Should we make more ‘‘live’’ that old thread or make a new one?

I will create a database and some UI to get nice results. Like a poll.

That would be great!

Already messaged whitepony concerning the template / questions.

Ok, I hope it does ‘‘assemble’’ soon enough for the first ones to be able to post some of the data they have gathered!


Perhaps, it can go as:

Technical: Board type (mtb or longboard) Gear ratio Wheel size Weight of the board + rider

Electrical: Motor Battery voltage Battery capacity Battery’s cont amp draw / c rate

Statistics:

Peak draw Avg draw

Max speed Avg speed

Km/mi ridden Energy consumed Wh per km/mile

Ride conditions: Flats / Hills: if hills: time ridden on them, suspectable/ possible incline % Road / terrain condition


Calculations: (automatically done)

Max Wattage used (amps x volts) Average Wattage used (amps x volts) Wh (Ah x volts) Wh per km / mile ( Wh / distance) (
)


Attention should be placed on that table in the bottom left corner.

(taken from: http://www.nature.com/articles/srep12967/figures/5)

25ÂșC = 77ÂșF, 55ÂșC = 131ÂșF

For that specific battery (which is LiCo chemistry), when comparing 25Âș C and 55Âș C operating temp, only 2.45% capacity decrease difference over 100 cycles.

At 200 cycles the difference becomes more wider - 7.19% (3.29% vs 10.48%) capacity loss for running at 25Âș or 55Âș C.


So, let’s look at the real loss when subtracted from total capacity of the battery.

As for 1500mah battery, at 100 cycles that would be: 1474,2 mah total (25Âș) and 1438,95 mah total (55Âș) at 200cyles: 1450,65 mah (25Âș) and 1342,8 (55Âș)

Now a 5000mah pack would be 3.33 times larger 10ah = 6.66 larger.

If the same percentage applies, then at for 5000mah and 100cycles = 4914 mah (25Âș) and 4765,5 mah (55Âș) at 200cyles = 4835mah/164,5mah loss (25Âș), 4476mah/524 mah loss (55Âș)

For 10ah

100 cyles =
at 25Âș C → 9828 mah / 172 mah loss at 55Âș C → 9531 mah / 469 mah loss

so - 2.72 times more

200cycles = 25Âș C - 9671 mah / 329 mah loss (25c),
55Âș C - 8952 mah / 1048 mah loss (55c)

3.19 times more


Enough calculations for one day. I think the test was done by keeping the battery at that temperature all the time! I do wonder how big difference there is for LiMn batteries (the high discharge ones) compared to the one in test (LiCo).

So, in general, little ‘‘bursts’’ of heat shouldn’t damage cell that much, if the heat does not reach 70-80 degrees celsius, I think. Especially considering that LiCo bears high drain current worse than LiMn.

So yes, I think it is a tricky matter to precisely estimate / calculate capacity loss over time, since if we take two different batteries, their real capacity may vary depending on load / amp draw. That’s of course why the majority stay on the safe side, not to degreade / heat their batteries excessively.

If I find something more related to this, for example, why exactly tesla has battery cooling, then I might post more info.


Im still confident that the real amp draw should be found.

Once we have that, we can more accurately estimate what batteries we might need, how hot / warm they may run and what capacity loss we may witness. We have 10A / 15 A / 20A / 25A draw configurations available. Plus there is capacity and price


We will need a battery calculator at some point in time.

Different max discharge rates X batteries in parallel X running temp X Capacity.


If we add battery cooling, then we would be able to ‘‘step into’’ using batteries which get warmer or does not have as high discharge rate.

Not saying that running batteries hot all the time is a wise idea, it may pose a risk that some batteries get weaker (balance problems) but I would aim, that in the future we better understand how all of the system operates and can design it a way we need without choosing a ‘‘safe’’ estimate.

This should become wildly important for someone who is going to build larger or more than one pack.

One comment about something that is important for some users: wind resistance. I live in the SF Bay Area, where in summer we almost always have 15-20kts winds in the afternoon by the bay. So even though it is flat, riding upwind is very taxing on the ESC. Even at low speeds (in which case the RPM is lower and therefore the motor is less efficient even
). I built my system initially for 6S but I’m already considering going to 10S so I can draw fewer amps. When I ride against this type of wind I have to push so much more to keep the same speed.

One question: I am waiting for a cable (damn unusual VESC connectors) to connect a bluetooth module to the VESC. How are you guys logging the data from the VESC? I’d love to collect these data riding full speed against the wind / no wind / with the wind. Doesn’t have to be real-time. I have an openlog board.

Leo

1 Like

@Okami you didnt read everything of that article - the authors Qm is an artificially calculated capacity of the battery - the problem is that youll have to get to your electrons somehow and the increased inner resistance of the battery will lead to additional heating and a reduced discharge capacity. the authors of the paper discuss that before the conclusions.

its dangerous to just cut pictures completely out of context! its more reasonable to look for papers, that do the things that we do: just use a battery and count the Wh you got out of the battery. for example this: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.606.4672&rep=rep1&type=pdf

which kind of draws a different picture: at 500 cycles, the drop in capacity as function of temp has been: room temp: 22.5% drop 45°C: 26.5% drop 50°C: 43.2% drop 55°C: 70.5% drop

after pushing my board I surely have been at 45°C battery temp and its over-engineered already with 80A continuous current!

also interesting the batteryuniversity read in that respect, especially the tables, most impressively the storage table! http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

Ok, I do agree that the paper I found was not that reliable concerning our situation, especially with the little 1500mah battery which was not 18650 and also the few cycles (250) compared, that some users may push the battery way further. Will have to sit down and read all of that paper + yours completely. It does point out some very important factors etc related to batteries.

And yes, I somehow tottally forgot about the wind! In that matter there is a need for higher amp setup.

Is this battery suitable? Is the chemistry safe? http://eu.nkon.nl/rechargeable/18650-size/icr18650he4-2500mah.html

It is a good one and widely used together with Sony Konions, Samsung 25Rs and LG HE2s. @whitepony: have you seen the new Konion VTC6? 3120mAh at 30A according to nkon website. But 8.45€ per cell is quite a lot :cry:

looks too good to be true! Ill wait for some discharge results on the vaping forums :stuck_out_tongue:

Well apparently your beloved HG2 perform almost identical http://www.mountainprophet.de/forum/viewtopic.php?p=5955&sid=fa87c944b949823865a689d8c87a02f7#p5955

so no reason to switch - especially with the price difference

1 Like

hELLO , I new to this thread I just bought 84 this type of battery But I hesitate about the originallity I buy this battery at local store


1.When I buy from titan battery.com, the battery have “C” mark, but when I buy at Local store The Mark is “B”,

2.I Set with NIMH operation at the charger, discharge set to 2amp When The Charger begin, it show discharge at 1.4 amp, the voltage drop from 4.2 to 3.88 Volt , this is normal? this is good resistance? I have no graphic, I try with note,

when the voltage at 3.4 Volt, the “Mah” that I got is about 1100 Mah
 when the voltage at 3.0 Volt, the “Mah” that I got is about 2300 Mah
 when the voltage at 2.8 Volt, the “Mah” that I got is about 3205 Mah
 everything is normal?

I did not try The “C” Mark Yet

Discharge with SKYRC balance charger, 2 amp discharge, until 2.8 Volt, I got about 3205 mah

thank YOU, I upload the Picture

It apear to be ok, for a better measurement take a resistor of about 10 ohms and measure the loaded and unloaded

With your numbers the internal resistance 228 mOhms, but the charger is taking into account the wires resistance