Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Ebike battery help
#11
(01-20-2019, 12:12 PM)Iker138 Wrote: Hi everyone!! I am making a ebike battery for a 1000w motor. I am making with 18650 recycle of laptop batteries. At full throttle it consumes 26A.
[...]


My electric skateboard is running a 10s12p battery from harvested laptop cells.  It runs 750W of motor, which translates to ~ 20A peak.  I have had this running for months now, and generally thrash the living daylights out of it, and taken some logs from it running.  Here are some thoughts:

1/ if you can, use all cells with same make/model.  this will spread the load evenly throughout discharge curve.  IE panasonics give out more power up high (>3.9V) while LG's give out more down lower (3.7-3.9V) ... or is it the other way around ... anyways, hoping you get my point.

2/ ALWAYS pay EXTRA attention to battery construction when used in mobile situations.  EG I stack mine 'parametrically' (each row offset half a cell width, so it looks more like a honeycomb than a spreadsheet).  So this means I risk the 'rows' (banks) of cells rubbing against each other and shorting out the -ve shell if there is damage on adjoining sleeves/wraps.  Heat [dissapation] also can become an issue when I go for extended runs with Wide Open Throttle.  but this is because my harvested laptop cells suck badly -- they are NOT rated for high discharge.

3/ I have found that with careful throttle management I can avoid too much sag.  even better with a 'rest' (all stop) between fast stretches, which allows the pack to 'bounce back' up, and get the most energy (range!) out of the battery.

I have only JUST got my hands on some cells rated for high discharge ... so I understand if laptop cells are all you got ... use 'em! ;)
--    ----    ----    ----    ----    ----    ----    ----    ----    ----    --
all systems nominal... rock on.
-----------------------------------------------------------------------
Reply
#12
I have sold Boston Power Swing batteries to several Ebike users ...and they are very happy with them. You might at least give these a look. I also have some prebuilt packs listed in the marketplace here. Yes this is a little bit of a sales pitch ...but really it is meant to be a suggestion that the experience I have had is that Ebike folks have been very happy with the Boston Power Swing solution. Buy them wherever you like ....but at least look at them as an alternative. If you go that way ...do not use the Boston Power Sonata ...it has a lower discharge rating.
Reply
#13
(01-22-2019, 04:11 AM)spinningmagnets Wrote: It sounds like you want to draw 26A from a 6P pack. That peak draw will be roughly 4.3A per cell. There are 2000-mAh cells that are made for high discharge, and there are also some 2000-mAh cells that are made for low cost. Laptop cells are the low cost version, and they will get hot at 4.3A

I understand your desire to save money by using cells that are available at very low prices, but I recommend that you focus on cordless tool cells, which are made for high current. If there is no room in your budget for cordless tool cells, you would need to use a much larger pack of laptop cells.

At 13S, a 6P pack is 78 cells, and that might be a large pack to try and fit it into your bicycle frame. This is why you should use cordless tool batteries, even if it costs more. You can get very high power form a 5P pack, and the cells will not get hot.

I only make this batteru for ONLY test the ebike conversion kit. Only 1 cycle. Now i write this post to know the opinion about mi new pack that i want to make. 2000mAh 13s15p. 1,73A per cell. 
Thanks for your recomendations!!

(01-22-2019, 04:30 AM)DCkiwi Wrote:
(01-20-2019, 12:12 PM)Iker138 Wrote: Hi everyone!! I am making a ebike battery for a 1000w motor. I am making with 18650 recycle of laptop batteries. At full throttle it consumes 26A.
[...]


My electric skateboard is running a 10s12p battery from harvested laptop cells.  It runs 750W of motor, which translates to ~ 20A peak.  I have had this running for months now, and generally thrash the living daylights out of it, and taken some logs from it running.  Here are some thoughts:

1/ if you can, use all cells with same make/model.  this will spread the load evenly throughout discharge curve.  IE panasonics give out more power up high (>3.9V) while LG's give out more down lower (3.7-3.9V) ... or is it the other way around ... anyways, hoping you get my point.

2/ ALWAYS pay EXTRA attention to battery construction when used in mobile situations.  EG I stack mine 'parametrically' (each row offset half a cell width, so it looks more like a honeycomb than a spreadsheet).  So this means I risk the 'rows' (banks) of cells rubbing against each other and shorting out the -ve shell if there is damage on adjoining sleeves/wraps.  Heat [dissapation] also can become an issue when I go for extended runs with Wide Open Throttle.  but this is because my harvested laptop cells suck badly -- they are NOT rated for high discharge.

3/ I have found that with careful throttle management I can avoid too much sag.  even better with a 'rest' (all stop) between fast stretches, which allows the pack to 'bounce back' up, and get the most energy (range!) out of the battery.

I have only JUST got my hands on some cells rated for high discharge ... so I understand if laptop cells are all you got ... use 'em! Wink
Okey!! Thank you a lot of for your recomendations. I agree with you that with a careful management of the trottle we avoid too much sag. Im making this bike for my and maybe to a presentation of a projects competicion. I dont wanna go at full throttle. Only sometimes. My biggest question is if it is safe. I know that the battery will not degrade fast because also i have a 250w mode in the controler so this was very beneficial to the battery. Apart of that, my rides times they are very shorts, so only discharge like %25 of the capacity. And to charge im thinking to adjust the parameter of the bms to cut-off at 4,1v. Have you any video about your skate?. Thank you and sorry for my english xD
Reply
#14
(01-22-2019, 04:30 AM)DCkiwi Wrote: 1/ if you can, use all cells with same make/model.  this will spread the load evenly throughout discharge curve.  IE panasonics give out more power up high (>3.9V) while LG's give out more down lower (3.7-3.9V) ... or is it the other way around ... anyways, hoping you get my point.

To be explicit, my context is USED LAPTOP cells, rated for LOW DISCHARGE.

Additionally there is another reason I strive for 'homogeneous' cells in my esk8 battery.. With all the cells being stressed so near to their [dis]charge limits it is essential that they all work in harmony.  This specifically includes individual cell capacity.  Hence a completely different beast of a battery than for instance, an 80P powerwall which might have cells with capacities that vary by 100+ %.  It doesn't matter that a 2000mAH cell is charging at twice the current as its 1000mAH capacity tested parallel neighbour, because its in an 80P a 40A battery charge is only an average cell charge rate of 500mA anyways!  

To summarize -- matching the capacities (reducing deviation) and discharge curves (brand/version) of cells in a battery becomes more important as the currents (charge or discharge) come close to cells max spec.  Tuning a battery like this can help reduce maintenance and improve lifespan/reliability.


(01-23-2019, 05:40 PM)Iker138 Wrote: Okey!! Thank you a lot of for your recomendations. I agree with you that with a careful management of the trottle we avoid too much sag. Im making this bike for my and maybe to a presentation of a projects competicion. I dont wanna go at full throttle. Only sometimes. My biggest question is if it is safe. I know that the battery will not degrade fast because also i have a 250w mode in the controler so this was very beneficial to the battery. Apart of that, my rides times they are very shorts, so only discharge like %25 of the capacity. And to charge im thinking to adjust the parameter of the bms to cut-off at 4,1v. Have you any video about your skate?. Thank you and sorry for my english xD

Yes I highly recommend you enter that competition.  I recently participated in a Maker Faire (Wellington 2018) and it REALLY boosted along a lot of my projects, experience, and confidence too.  We'd love to see your pictures, keep us all posted.  : )

Me, I am very heavy handed on the throttle, and am more in the mode of making my stuff perform like I wanna drive, rather than the other way round.  If I discharge 25% of my battery on one ride, it WAS a long ride.  lol.  about the only time I would dare to charge my battery over 4.1V/cell would be to 'trick' the charger into giving full CC by raising the CV limit to 4.12V and then it would just truck through to 4.10V without slowing down at the end.  But then I'd manually stop the charge and pack/ride it from there.  No video on my esk8, am a bit camera shy, but I will start another thread here soon ok, not wanting to hijack this one.  Your english is better than my [pick any language].  Thanks for sharing.
--    ----    ----    ----    ----    ----    ----    ----    ----    ----    --
all systems nominal... rock on.
-----------------------------------------------------------------------
Reply
#15
(01-23-2019, 07:57 PM)DCkiwi Wrote:
(01-22-2019, 04:30 AM)DCkiwi Wrote: 1/ if you can, use all cells with same make/model.  this will spread the load evenly throughout discharge curve.  IE panasonics give out more power up high (>3.9V) while LG's give out more down lower (3.7-3.9V) ... or is it the other way around ... anyways, hoping you get my point.

To be explicit, my context is USED LAPTOP cells, rated for LOW DISCHARGE.

Additionally there is another reason I strive for 'homogeneous' cells in my esk8 battery.. With all the cells being stressed so near to their [dis]charge limits it is essential that they all work in harmony.  This specifically includes individual cell capacity.  Hence a completely different beast of a battery than for instance, an 80P powerwall which might have cells with capacities that vary by 100+ %.  It doesn't matter that a 2000mAH cell is charging at twice the current as its 1000mAH capacity tested parallel neighbour, because its in an 80P a 40A battery charge is only an average cell charge rate of 500mA anyways!  

To summarize -- matching the capacities (reducing deviation) and discharge curves (brand/version) of cells in a battery becomes more important as the currents (charge or discharge) come close to cells max spec.  Tuning a battery like this can help reduce maintenance and improve lifespan/reliability.


(01-23-2019, 05:40 PM)Iker138 Wrote: Okey!! Thank you a lot of for your recomendations. I agree with you that with a careful management of the trottle we avoid too much sag. Im making this bike for my and maybe to a presentation of a projects competicion. I dont wanna go at full throttle. Only sometimes. My biggest question is if it is safe. I know that the battery will not degrade fast because also i have a 250w mode in the controler so this was very beneficial to the battery. Apart of that, my rides times they are very shorts, so only discharge like %25 of the capacity. And to charge im thinking to adjust the parameter of the bms to cut-off at 4,1v. Have you any video about your skate?. Thank you and sorry for my english xD

Yes I highly recommend you enter that competition.  I recently participated in a Maker Faire (Wellington 2018) and it REALLY boosted along a lot of my projects, experience, and confidence too.  We'd love to see your pictures, keep us all posted.  : )

Me, I am very heavy handed on the throttle, and am more in the mode of making my stuff perform like I wanna drive, rather than the other way round.  If I discharge 25% of my battery on one ride, it WAS a long ride.  lol.  about the only time I would dare to charge my battery over 4.1V/cell would be to 'trick' the charger into giving full CC by raising the CV limit to 4.12V and then it would just truck through to 4.10V without slowing down at the end.  But then I'd manually stop the charge and pack/ride it from there.  No video on my esk8, am a bit camera shy, but I will start another thread here soon ok, not wanting to hijack this one.  Your english is better than my [pick any language].  Thanks for sharing.
Yes, i suposse that i post some images but this project gets a lot of time to buil it. I need more cells and received all components from alliexpress. Im new in this forum. Can i put you like "friends"? . I dont no if i can xd.
Reply


Who read this thread?
59 User(s) read this thread:
Funkadelicious (08-13-2019, 01:39 PM), Legolfeur95 (01-23-2019, 08:17 PM), Richard (01-23-2019, 06:26 PM), c70r (01-22-2019, 02:05 AM), Powersaver (03-12-2019, 12:49 PM), sparkgap (01-22-2019, 08:43 PM), LEDSchlucker (01-24-2019, 03:25 PM), OffGridInTheCity (06-21-2019, 11:38 PM), Arnestobar (01-24-2019, 02:05 AM), newone972 (09-20-2019, 06:19 AM), ajw22 (06-21-2019, 09:39 PM), Speicher (01-28-2019, 05:43 PM), Generic (06-22-2019, 01:48 PM), Iker138 (01-25-2019, 08:00 AM), ma3006he (01-28-2019, 11:13 AM), mskip (02-02-2019, 12:05 PM), maozmicky (09-01-2019, 10:57 PM), Hotsy-totsy (08-23-2019, 10:01 PM), Bobthebuilder (06-28-2019, 06:58 AM), wetandfire (06-09-2019, 12:15 PM), billvon (06-21-2019, 09:37 PM), daniel84uk (06-06-2019, 06:33 PM), Scripcarul (05-05-2019, 06:09 PM), energy_man (06-21-2019, 09:28 PM), gueorgi (03-08-2019, 03:05 PM), Leafnut (03-08-2019, 06:18 PM), Lucan (07-09-2019, 09:32 PM), i-craft (01-22-2019, 04:56 AM), Bastler_Ro (01-23-2019, 06:49 PM), Wolf (01-21-2019, 03:21 PM), Ronald Hyde (01-28-2019, 10:10 PM), DCkiwi (06-24-2019, 03:07 AM), DarkRaven (01-23-2019, 11:19 PM), Geek (06-24-2019, 01:50 AM), BlueSwordM (01-22-2019, 03:42 AM), Stefaan De Ridder (01-24-2019, 06:43 PM), watts-on (01-24-2019, 09:07 AM), rebelrider.mike (01-30-2019, 05:04 AM), PAF (01-23-2019, 08:33 PM), Korishan (07-03-2019, 11:01 AM), mike (01-24-2019, 02:21 AM), Sean (06-21-2019, 09:58 PM), daromer (01-20-2019, 08:42 PM), Filip (01-23-2019, 08:34 PM), djuro (02-25-2019, 04:08 PM), spinningmagnets (01-22-2019, 04:11 AM), Leipi35 (01-22-2019, 11:56 PM), jaeger-audiodesign (06-22-2019, 08:10 AM), Beholder (06-21-2019, 09:46 PM), emuland-metroman (01-24-2019, 03:41 PM), chuckp (01-24-2019, 01:13 AM), Riplash (01-20-2019, 02:51 PM), dibs (09-08-2019, 08:10 AM), cadric (01-21-2019, 10:07 AM), Dala (01-21-2019, 11:31 AM), rodagaster (01-22-2019, 12:33 AM), Headrc (06-22-2019, 01:27 AM), Chablis_m (01-23-2019, 10:27 PM), hbpowerwall (06-24-2019, 04:07 AM)

Forum Jump:


Users browsing this thread: 1 Guest(s)