PSA: Spark bikes could be using fake 18650 cells in their battery packs. Edit: confirmed LG M36 cells in 17AH pack

HKPolice

New Member
Got my Spark City 17AH today, popped open the end of the battery pack and immediately noticed that the cells are in light blue wrappers. This makes no sense because LG only makes two 18650 cells with this color, the MH1 which is rated @ 3200mah or the M36 rated @ 3600mah.

The battery pack is in a 13s 5p confirguration, so 3.2AH*5 = 16AH or 3.6AH*5 = 18AH. Neither matches the 17AH pack rating.

What makes things even more confusing is that sparkbikes themselves told me they're using LG F1L 3350mah cells which have light purple wrappers: https://lygte-info.dk/review/batteries2012/LG 18650 F1L 3350mAh (Purple) UK.html

This leads me to suspect that the cells are not LG at all, or the pack is using lower capacity MH1 3200mah cells which means it should be labelled as a 16AH pack.

If you have the 17AH pack, please unscrew the bottom cover and take a peak @ the cells. It's just 4 phillips screws.
 

Attachments

Handlebars

Active Member
The glue they used is excellent though. 😀
Careful with metal tools once it's open, people. I used only plastic tools and fingernails in case I slipped up.
 

techeazy

New Member
If no prove to state this is legit, I prefer to stay away from Sparks. I still think what you pay is what you get.
 

emtbdude

Member
HKP, if I were in your situation, I would return the bike immediately and even demand return shipping expenses.

No one should ever tolerate counterfeit product or even misrepresentation of this nature. If they got a pedal spec wrong, or substituted a headset, that's one thing. But using counterfeit cells is 100% intolerable.

Contact the company asap, return it, and get a real bike.
 

Handlebars

Active Member
I didn't want to go further because I've never taken a bike battery apart before, but I'm interested to see what the cells say on them since there are LG cells with that color wrapper. If it's a case of 16Ah sold as 17Ah, they owe customers a bit of change, I figure. If they are just cheapo generic batteries they owe customers a good bit of change. But if they are LG and add up to 18 Ah I'll shut right up. But that isn't very likely. :p
 

HKPolice

New Member
I'm setting up a rig that will measure battery capacity while discharging using this watt meter: https://www.amazon.ca/Digital-Wattmeter-Checker-Balance-Voltage/dp/B07GJ33PT1

Should get some results this weekend.

Digging further into LG cells, the M36 isn't officially rated @ 3600mah, but more than half the vendors on the net are making that claim. LG's data sheet states the nominal capacity as 12.5Wh with an average voltage of 3.63v which makes the true capacity rating around 3443mah or 3450mah when rounded. I think the reason why vendors are claiming 3600mah is because the LG M26 cell was rated @ 2600mah so M36 logically should be 3600mah.

I'm hoping that these are real M36 cells which would result in a theoretical pack capacity of 17.25AH, rounded down to 17AH. I don't think there are any generic chinese cells capable of more than ~3000mah capacity. So If I get close to 17AH in a discharge test, then M36 confirmed.
 
Last edited:

Handlebars

Active Member
I'm setting up a rig that will measure battery capacity while discharging using this watt meter: https://www.amazon.ca/Digital-Wattmeter-Checker-Balance-Voltage/dp/B07GJ33PT1

Should get some results this weekend.

Digging further into LG cells, the M36 isn't officially rated @ 3600mah, but more than half the vendors on the net are making that claim. LG's data sheet states the nominal capacity as 12.5Wh with an average voltage of 3.63v which makes the true capacity rating around 3443mah or 3450mah when rounded. I think the reason why vendors are claiming 3600mah is because the LG M26 cell was rated @ 2600mah so M36 logically should be 3600mah.

I'm hoping that these are real M36 cells which would result in a theoretical pack capacity of 17.25AH, rounded down to 17AH. I don't think there are any generic chinese cells capable of more than ~3000mah capacity. So If I get close to 17AH in a discharge test, then M36 confirmed.
Thank you!
 

Handlebars

Active Member
Digging further into LG cells, the M36 isn't officially rated @ 3600mah, but more than half the vendors on the net are making that claim. LG's data sheet states the nominal capacity as 12.5Wh with an average voltage of 3.63v which makes the true capacity rating around 3443mah or 3450mah when rounded. I think the reason why vendors are claiming 3600mah is because the LG M26 cell was rated @ 2600mah so M36 logically should be 3600mah.

I'm hoping that these are real M36 cells which would result in a theoretical pack capacity of 17.25AH, rounded down to 17AH. I don't think there are any generic chinese cells capable of more than ~3000mah capacity. So If I get close to 17AH in a discharge test, then M36 confirmed.
The M36 looks like a very good battery to me. Probably looking to see slightly over 3300mAh each cell at normal discharge rates?

And they are rated so good after 1000 cycles...double the cycles many others are rated after.
4.3.4Cycle LifeCells shall be charged and discharged per 4.2.3 for1000 cycles. A cycle is defined as one charge and one discharge. 1001st discharge energy shall be measured per 4.2.1 and 4.2.2.≥80%
 
Last edited:

HKPolice

New Member
Discharge test done @ 1-1.2A load per cell.

I wasn't able to get a picture before the pack cut out @ around 38.4v output (2.95v per cell) since I had no idea what the BMS shut down voltage was. The final capacity was around 15.5AH which works out to about 3100mah per cell @ 2.95v EODV (end of discharge voltage). This matches up pretty well with LG M36 specs which gives 3206mah @ 2.8v EODV and most likely eliminates the possibility of LG MH1 cells which only provide 3080mah @ 2.8v EODV.

So M36 cells confirmed, spark bikes was wrong to claim F1L cells. I'm still not satisfied with the 17AH rating though since the high cut off voltage of 2.95v per cell means max theoretical capacity is only 15.5AH and in real world riding you'll be lucky to get 15AH since using anywhere near full power will discharge the cells at much lower efficiencies. The bright side of this is that the battery pack should last at least 500 cycles before there is any noticeable degradation.

Another negative side about having such a high cut off voltage is that if you try to use full power when the battery is around 40v, it would be easy for the voltage to sag below 38.4v which triggers the pack to shut down even though the cells could still have 15-20% capacity. I think 2.8v cell or 36.4v pack is a more reasonable cut off voltage and shouldn't cause much degradation over 38.4v.
 

Attachments

Last edited:

Handlebars

Active Member
I've put 1200km on it and run full power a lot and haven't had a shut down ... so maybe the small controller is keeping that in check?
I've only run slightly below 20% a couple of times though, so maybe it doesn't even apply to my useage.

Thank you for doing the test, HKPolice!

I think you are correct to say that rating it as 17Ah isn't good practice - even if technically it could apply...if you can't get it out of the pack.
 
Last edited:

HKPolice

New Member
I've put 1200km on it and run full power a lot and haven't had a shut down ... so maybe the small controller is keeping that in check?
I've only run slightly below 20% a couple of times though, so maybe it doesn't even apply to my useage.

Thank you for doing the test, HKPolice!

I think you are correct to say that rating it as 17Ah isn't good practice - even if technically it could apply...if you can't get it out of the pack.
It was fun doing the test, I haven't seen anyone else do a discharge test on these Reention dorado packs so I figured why not.

I asked sparkbikes & they said that the controller is 26A max. The motor has no limit and will use whatever amperage you throw at it until it overheats.

During testing, I pressed the power button on the pack occasionally to check if the meter was accurate and it showed 0 bars at around 42v. Not sure how the bike's controller interprets battery capacity, it could be reporting 20% @ 44v or higher. This leads me to question the controller's minimum working voltage as well, maybe it's higher than 38.4v which would leave even more battery capacity unused.
 
Last edited:

harryS

Well-Known Member
48V ebike controllers usually are set to go off around 40-41 volts. This value is often printed on the controller label.

Some can be tweaked +/- a volt or two around this number via display setup.