Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Ebike help
#1
Hi Guys,

So I've been harvesting cells for a few months now. I've got about 120 good cells (most have 2000+ aH) and my goal is to make a ebike. I got some question that hopefully you guys can help me out with. I'm trying to make a battery that can power my ebike so it can take me to work and back, and thats roughly 30 miles. I'm a big guy. I weigh 230 pounds (104.3 kg) and im 6 feet 3 inches (190.5 cm) tall. 

How many wH should i aim for?

I'm goign to buy a conversion kit that has a 48V1000W brushless gearless hub motor. I believe i can make a 4.94 kWh battery (103aH * 48v) with 52 cell (103aH/2000mAh **roughly). Is my thinking wrong on this? is this enough?

Guidence appriciated.

Thanks.

EDIT:
Turns out i did my math wrong... I am making a 13s4p battery. So its 4 * 2000mAh = 8Ah. Then it would be 48.1v * 8Ah = 388.8wh. Is this right?
Reply
#2
Are you getting a bafang motor set up? I have that motor and run a 13.2ah battery and get roughly 70km out of a charge. I'm 80kg and 167cm, I am also a Fit person and still ride it as much like a normal bike as possible. If I don't put much effort into pedalling and use it more like a motor bike I can get around the 20 to 30km from a charge. This is me riding at an average of 30km per hour.
Reply
#3
Hey mate,

I've made 2 ebikes and learned a lot.  Both have a 48V 1,000W hub motor on a 26in rear wheel.

1) 1st attempt used a 13S8P battery made from recycled laptop cells (av 1.95Ahr so ~ 750Whrs)  1,000W hub motors pull about 20+ Amps so using laptop cells in an 8P configuration was too few (2.5A discharge per cell)

Would go about 20km with a few mild hills (I'm 85kg) and battery would get hot, voltage would sag badly and BMS would cut out if pushed too hard.

2) 2nd attempt much better - 14S10P using Samsung INR18650-20R (and 25R) from recycled Ego 56V packs - each cells happy at 20A discharge so no more voltage sag.  Capacity is 20.58Ahrs or 1.07 kWhrs - easily gets me 26km to work including 3 very steep hills (Strava says 520m elevation gain).  

Volts drop from ~58.8 to ~50.9V - so roughly down to 3.64V per cell so somewhat under 50% capacity (very roughly).

26km trip takes around 42min averaging 38km/hr - takes 5hrs on a 2A lead acid charger to get the volts back up.

A lot of fun to ride.  Done ~ 1,000km on it so far.

Recommendations:

a) use high current cells to avoid voltage sag you get with laptop cells
b) bus bars between cells in series to be min 20A capable (~ 4.6mm2 DC cable)
c) replace plug in connectors from controller to hub motor with soldered joints (a couple had melted when I looked)
d) upgrade thin wires from controller to hub motor with at least 4.6mm2 cable ~ 20A capable
e) make sure your bike has disk brakes ;D
BoomBoomShroom and Bdaearth like this post
Reply
#4
If you haven't sourced your motor yet put some thought into a mid drive so you can still have 10 gears on the rear wheel.

http://dillengerelectricbikes.com.au/ele...afang.html

This is the kit I put on my bike but this was well before I found anything about doing a complete build yourself. I have the 11.6ah battery not a 13.2 as said previously, it was the only option when I got it, but you are building your own so you don't need to worry about that.
Reply
#5
Hello Emerca, first of all sorry for my english, im spanish. I wat to make a battery for my 48v 1000w ebike kit. My ideas was: 13s10p battery with old laptop cells (all 2000mA). The controler can maximun off 26A. So 26/10= 2,6A per cell? Apart of that i wwant to put 5A fuses each cell? That was safety?, to discharge each cell to 2,6A? Thank you for readingSmile
Reply
#6
(11-04-2018, 01:50 PM)IkerTronic Wrote: Hello Emerca, first of all sorry for my english, im spanish. I wat to make a battery for my 48v 1000w ebike kit. My ideas was: 13s10p battery with old laptop cells (all 2000mA). The controler can maximun off 26A. So 26/10= 2,6A per cell? Apart of that i wwant  to put 5A fuses each cell? That was safety?, to discharge each cell to 2,6A? Thank you for readingSmile

Hi IkerTronic - ha your english is fine mate - far better than my spanish!

Sorry to say I can't recommend laptop cells for an ebike - unless you go 13S20P (or go a lower wattage motor).  

IMHO laptop cells are usually rated for max discharge at ~500mA - I have tested a few thousand at 1,000mA (on Opus testers) and they don't get too hot - that's why I say 20P min for laptop cells.  If it's your 1st pack and want to have some fun and don't mind damaging cells - your 13S10p will work for a few hundred km.

Another suggestion is to go 16S10P - the BMS is still cheap and the increased volts will lessen the amps.

The main problem is that an ebike battery using recycled cells must have a BMS and at say 20A on laptop cells they will heat up quite a bit, voltage will sag (amps increase) and will cause a low voltage disconnect - just when you don't want one.

Maybe do what I did and build a laptop pack, have some fun and see what happens while you get some high current cells from power drill packs??


And yes I fuse both +ve and -ve with .26mm copper wire to protect against shorts etc.

Good Luck!
Reply
#7
(11-04-2018, 03:52 AM)emerca Wrote: Hey mate,

I've made 2 ebikes and learned a lot.  Both have a 48V 1,000W hub motor on a 26in rear wheel.

1) 1st attempt used a 13S8P battery made from recycled laptop cells (av 1.95Ahr so ~ 750Whrs)  1,000W hub motors pull about 20+ Amps so using laptop cells in an 8P configuration was too few (2.5A discharge per cell)

Would go about 20km with a few mild hills (I'm 85kg) and battery would get hot, voltage would sag badly and BMS would cut out if pushed too hard.

2) 2nd attempt much better - 14S10P using Samsung INR18650-20R (and 25R) from recycled Ego 56V packs - each cells happy at 20A discharge so no more voltage sag.  Capacity is 20.58Ahrs or 1.07 kWhrs - easily gets me 26km to work including 3 very steep hills (Strava says 520m elevation gain).  

Volts drop from ~58.8 to ~50.9V - so roughly down to 3.64V per cell so somewhat under 50% capacity (very roughly).

26km trip takes around 42min averaging 38km/hr - takes 5hrs on a 2A lead acid charger to get the volts back up.

A lot of fun to ride.  Done ~ 1,000km on it so far.

Recommendations:

a) use high current cells to avoid voltage sag you get with laptop cells
b) bus bars between cells in series to be min 20A capable (~ 4.6mm2 DC cable)
c) replace plug in connectors from controller to hub motor with soldered joints (a couple had melted when I looked)
d) upgrade thin wires from controller to hub motor with at least 4.6mm2 cable ~ 20A capable
e) make sure your bike has disk brakes ;D

I did some more research and was going to attempt a 13s10p setup. But from the info you guys seem to be giving me it seems like i need to avoid the voltage sag, so I'll attempt a 14s10p. I just need more cells. I sadly did a count recently and only have about 90 cells. Could have sworn i had close to 120 lol.

I plan to have a bus bar and fuses on my pack. What kid of bms do i need? Is it a 14s 52v 20A?
Reply
#8
Hey all. first post..

Hey BoomBoomShroom...

Was reading in this thread and thought that I'd throw in my 2 cents.
(Links and photos to follow)

E-bike - 18months riding approx 2000km

Background: I have a 1000w rear hub (brushless direct drive) from a cheapo chinese kit.
Ended up upgrading the controller and added a lcd display. I ride all year (in winter I stop riding once it goes below -20c - it's just too fracking cold)

Winter Riding Tip: studded front tires, disk brakes and personal body armour... pain is a harsh teacher.

My bike is a Raleigh Airlite MTB with front suspension only.
Regular brakes - definately will upgrade to Disk brakes - even better hydralic Disk, as I have not been able to slow down fast enough many times
and have had 2 accidents now, broken ribs and all. sigh. I like to go fast, keeping up with the traffic.
E-Bike Specs: (next post will have all the links and Pics)
1000w rear hub motor
controller
Thumb throttle
LCD screen (upgrade)
PAS
cutout style break levers and 1 HWBS
Heavy duty Rear Rack
Heavy duty Pannier bags (for battery on one side and cargo on the other)
Wire Basket for my backpack, more cargo
Heavy duty center mount dual leg kickstand

I've added 12v LED motorcycle brake light strip to the rear and to the front bendy motorcycle turn signals, to controll the lights and horn - a motorcycle handlebar mounted switch with on/off (lights) horn button, and left right turn signals. This all runs off a 3s2p 18650 pack (laptop batteries) with a 12v relay for the turn signals and a 12v to 5v stepdown for the rear LED and the front headlight. In retrospect I'll upgrade the secondary - lighting battery to a 4s3p as the turn signal relay (12v) tends to stop working fairly quickly as the 3s battery filly charged drops below 12v fairly quickly, and then the lights work (5v) but the turn signals fail (low voltage)

Batteries:
I watched many of the early bike battery videos on youtube and started with building a 48v 13s12p hotglued together with solid wires everywhere and no fusing (Ya I know)
Then I started buying BMS from china and even a 14s bluetooth bms (xoaxing or something like that...)
So now I have 5 -48v batteries (all have cheapo BMS) and one 52v battery (with the BT BMS) for my e-bike and I'm finding that the cell groups are drifting further and further apart, even when using the bms's to charge and discharge (run the bike)

Odly enough, the first battery I built - with the least knowledge and hotglued together and ducktaped up - is the longest lasting and strongest (it is 12p while the rest are 8p or 10p)

In my very limited experience and knowledge - Laptop scavanged batteries are ok, IF you use enough of them in parallel (ie: 12p or 14p)

My daily commute is about 25km, keeping up with traffice (35-45km/k) mostly flat but with one large double overpass, and almost always a headwind..
I pedal, but just enough to keep the PAS powering the bike, so it might as well just be throttle, ya I'm lazy that way... it's built for transport, not excercise.

My 8p13s 48v pack gets quite warm (no actual temp measure - but quite warm to the touch) after first leg of my commute 11km
My 10p13s 48v pack (same distance and speed - approx) gets warm, but not as much
My 10p14s 52v pack (same distance and speed - approx) isnt getting warm at all.
My 12p13s origional Ducktape franken battery - cool as a cucumber.

In future all my batteries for the 48v1000w hub motor will be 52v 14s14p with BMS's and additional pigtail leads for external balance charging and individual cell group charging to off set the Drift that the cheapo chinese BMS's cannot account for.

I know that my battery testing was lacking, I would charge, test with Opus once, group them by mAh, then build a battery once I had enough. - This yeilded batteries that would only last for a few dozen charges before drifting out of balance so far that the cheapo BMS's cannot cope with it.

Future Battery testing: Charge, let sit for 1 week, check V, discard the self discharging batteries, Test on the Opus for capacity and record on the cell mAh and V, Test and mark IR (internal resistance) with both Opus and proper IR tester, then group by mAh.

Future Packs will be built with the Pack builder spreadsheet that one of the members here posted (way cool) soldered together with fuse wire and 2mm buss bars (household wire stripped), BMS, extra leads for external charging (BMS or 1cell at a time) Silicone load and charge wires with anderson 90's then yoga matt foam and heatshrink.

Future plans: E-Cargo Trike, off road stupidly fast fully suspended trail bike, super efficient light weight PAS commuter, some form of pedalelec 4 wheeler, Powewall (thanks HBPowerwall, I have followed your vids from the beginning) PV system to go off grid. Farm

I have learned everything from the Interwebs and all the very kool people who take the time and effort to post their experiences and Videos, man the work that goes into those is astounding... My Thanks to the Community, and all they give so freely.
Reply
#9
Hey, I'll tell you what I do. I'm building an eBike myself.

When I design a battery, I start with the power requirement. In the case of an eBike motor, generally they'll give you Watts and Volts. More tricky is how long do you want the battery to last per charge? So for example, let's say a 48V 1000W motor, and we want to have it last for an hour at full power. We can then figure out the characteristics that the battery needs to have:

48V
1000W
20.8A  =1000W/48V
1 Hour
20.8 Ah  =20.8A*1h
998.4Wh  =20.8Ah*48V

Next we need to know the capability of each cell I'll be using. Easy if you buy a bunch of the same cell, you can look up the data on the forum database here. More difficult if you have a bunch of different cells. You'll have to find an average or think conservatively. For example, how many Amps can each cell give, and how many Ah does each have. Just remember putting cells in Series, you'll add the Volts, but the Amps and Ah don't add up. Each cell will experience the total current in the Series, and the capacity is an average of what each cell has. And putting them in parallel, you'll add the Amps each cell can do, and add up the capacity of each cell. But the Voltage will be the same.

Anyway, let's say we have cells at 3.7V (4.2-2.8V) 2Ah, and can deliver 1A safely. We can use that to determine first, how many in series:

48V/3.7V=12.97 cells.

Of course, you can't have 12.97 cells, so round up to 13 cells. Motors, and motor controllers, generally operate over a range of Voltages, so you might compare the range of the motor/controller to that of the battery you might make. For example:

12s = 50.4V-33.6V
13s = 54.6V-36.4V
14S = 58.8V-39.2V

Let's say 13s ends up matching best. Next, we need to figure out how many in parallel we need. The two factors that will determine this is Wh, and Amps. Watt-hours are interesting, because they add up whether you put cells in series or parallel. So from above we know we need 998.4Wh total. We can figure out the Wh of each cell:

3.7V(nominal)*2Ah= 7.4Wh
998.4Wh/7.4Wh= 134.9 cells total.

Again, we can't have 0.9 of a cell so round up to 135 cells. So we can figure how many to put in parallel because we know how many we'll have in series already:

135/13s= 10.4p (Round up) 11p

That's just for Wh though. Still need to know how many are needed in parallel to satisfy the Amps:

20.8A(total)/2A(per cell)= 10.4p (Round up) 11p.

That's weird, that they match in this example. For me they usually don't and I have to choose the higher number. But in this case, 13s11p. Notice that the total number of cells, because we had to round up, is now 143 cells.

You can also check your work by comparing the battery you've just designed to the original specs you needed:

48V and 54.6V-36.4V  Yep!
20.8A and 22A (11p*2A/cell)  Yep!
998.4Wh and 1,058.2Wh (143cells*7.4Wh/cell)  Yep!

I hope that makes sense. I put all this in a spreadsheet so every time I design a new battery, it does most of the work for me. Smile
-Mike G
Reply


Who read this thread?
64 User(s) read this thread:
Funkadelicious (08-13-2019, 11:01 AM), Iker138 (03-25-2019, 02:20 PM), Sunrise (01-15-2019, 09:19 AM), kabnikoff (01-30-2019, 03:48 PM), DTLSchiller (01-13-2019, 05:27 AM), andy chalmers (02-03-2019, 08:19 PM), emaidkd (12-29-2018, 12:52 PM), -lafayette- (12-22-2018, 11:15 AM), LEDSchlucker (02-07-2019, 07:32 PM), blind (12-17-2018, 01:26 AM), hnetto (11-29-2018, 02:02 PM), newone972 (02-13-2019, 02:09 PM), Bedo (11-11-2018, 05:33 AM), Bdaearth (11-09-2018, 10:52 AM), IkerTronic (12-14-2018, 05:13 PM), mskip (02-10-2019, 11:58 AM), psuggmog (02-12-2019, 03:03 AM), maozmicky (09-01-2019, 10:37 PM), Thehelper (07-24-2019, 04:34 PM), Hotsy-totsy (08-23-2019, 10:00 PM), malo01 (07-14-2019, 02:46 AM), stok (07-03-2019, 06:40 AM), Bobthebuilder (06-27-2019, 02:33 PM), Zeko (06-10-2019, 02:36 PM), OldOakTree (05-13-2019, 12:42 AM), JohanHem (04-23-2019, 10:00 PM), loyd (03-12-2019, 10:26 PM), energy_man (06-21-2019, 09:17 PM), GeneralDJ (03-08-2019, 09:25 AM), Greenbatt (03-08-2019, 07:22 PM), guurue (02-10-2019, 08:18 PM), stevelectric (11-06-2018, 10:24 PM), Bayyn (Yesterday, 05:18 PM), emerca (11-06-2018, 09:45 AM), Geek (02-08-2019, 01:09 AM), ivnz (01-30-2019, 08:46 PM), Stefaan De Ridder (11-05-2018, 03:17 PM), jesusangel (11-03-2018, 09:31 PM), watts-on (02-07-2019, 10:21 AM), rebelrider.mike (02-07-2019, 05:54 AM), wim (11-05-2018, 07:47 AM), jdeadman (11-05-2018, 01:50 PM), PAF (10-31-2018, 02:56 PM), Korishan (02-07-2019, 11:27 PM), mike (11-04-2018, 06:27 PM), Sean (11-04-2018, 07:57 AM), daromer (11-04-2018, 02:40 PM), slimf (11-13-2018, 02:49 AM), BoomBoomShroom (11-19-2018, 03:32 PM), Generic (02-07-2019, 04:17 PM), collinkoolen999 (11-14-2018, 09:07 AM), Bastler_Ro (01-30-2019, 06:20 PM), Wolf (01-30-2019, 05:52 PM), Ronald Hyde (11-12-2018, 11:07 PM), Juan Manuel Sagasti (12-19-2018, 06:28 PM), Beholder (11-04-2018, 10:51 AM), chuckp (01-12-2019, 06:53 PM), djotter (02-27-2019, 12:31 AM), Bubba (11-04-2018, 01:08 PM), cadric (01-19-2019, 06:20 PM), Chablis_m (02-07-2019, 09:57 AM), rtgunner (11-04-2018, 04:23 PM), brwainer (11-08-2018, 05:07 AM), hbpowerwall (11-03-2018, 09:10 PM)

Forum Jump:


Users browsing this thread: 1 Guest(s)