Another EV bike :)

An area for all you bikers who now run your two wheeler on battery power, scooter owners to.
ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Another EV bike :)

Postby ChrisBarron » Fri Jul 12, 2013 8:45 pm

Hello, just sharing my project details, for comment as much as for amusement.
The bike is a derilict FZR1000 frame and wheels, with most of the plastics and fuel tank. I've sold thew engine/exhaust/electrics which all paid for the bike itself :)

The batteries were going to be lipo. I'ld collected a lot of 10Ah lipo packs consisting of 4 x 2500mAh packs in a parallel which werte designed as a travel battery gizmo, but they just couldn't be trusted to provide more than 1C discharge without getting too hot and the pouches swelling.

I've sold the lipo's and over about 8 months I've been buying odd ebay sales for used laptop battery packs. I've spent much of the past week consolidating what I have and reckon I have 700 useable 18650 cells, of between 1800 - 2600mAh capacity.

The pack is where the work is going to be. I've categorised the batteries - 700 have a voltage of 2.9V or above - 140 have a voltage of 2.5 - 2.9V (many of these seem to recover following a slow charge - and lots more below 2.5V (some recover some just get hot ;) !

Next job is to charge and discharge every cell to measure it's capacity. Although I have a cell analyser which can accurately analyse an individual cell, I obviously need something capable of bulk opperations ! For this, I've built a discharger which can discharge 8 cells through a fixed resistor at a time. The gauge value I'll probably end up using is simply time taken to reach 2.9V from a fully charged state. The dishcarger is pic based, has a simple 16 x 2 lcd display and costs about £15 to make, which is just as well because I'm going to build another one or two when more 18650 cell holders arrive.

To charge them up I've bought 22 single cell chargers which can pass 1 Amp during bulk charging. The charger's are cheap and seem to be reliable and quite intelligent, even slowly charging a cell from 2.3V with 90mA until the voltage reaches about 2.97V, when it goes into bulk charge mode with a CC/CV method.

I'm also buddying the lithium pack with a small lead acid pack. I've read a few articles about this mixed chemistry method, and there seems to be a few benefits. The lead batteries I have are 5Ah 12V good quality batteries from UPS supplies, capable of 40Amp each. I have 15 of them, so 25Ah at 36V seems right.

Unfortunately, my motor is from a golf cart and 24V-36V is about it's limit before it gets too hot. The controller is an old GE10 SCR controller. I'll build the battery pack to allow it to be reconfigured at a later date for higher voltages.

I'ld like to hear from anyone else who has built a 'massively parallel' laptop/18650 cell battery pack. BMS issues seem to be less important with such a large parallel pack but I'm still thinking I could build my own BMS based on a design I came up witha couple of years ago, should it be required.

Thats it for now !
Chris

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :) Some photos

Postby ChrisBarron » Mon Jul 15, 2013 11:00 pm

This is my battery collection http://www.flickr.com/photos/28496053@N04/9294245479/

I also have 100 or so of these http://www.flickr.com/photos/28496053@N ... otostream/

My 8-cell discharger http://www.flickr.com/photos/28496053@N ... otostream/

Discharger underside showing relays (£5.95 off ebay for relay board !) http://www.flickr.com/photos/28496053@N ... otostream/

The charger power supply, 5V at 80A http://www.flickr.com/photos/28496053@N ... otostream/

Just waiting for a few more cell holders to arrive to finish this one off (red led = charging, blue = not charging) http://www.flickr.com/photos/28496053@N ... otostream/


There will be ten battery slots each side to give the 20-cell charger http://www.flickr.com/photos/28496053@N ... otostream/

Chris

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :)

Postby ChrisBarron » Sun Aug 11, 2013 7:20 am

A short update covering the past couple of weeks, which have involved me fitting and removing cells in my charger bank and bulk discharger charger and then writing on the side of each one the duration of discharge through the 3R9 resistor.

I've probably reached about 600 cells now, I think I manage at least two groups of cells per day (a group is between 34 -40).

I started off by choosing only the cells which had a minimum voltage of 2.9V. I finished off that batch yesterday and decided to try some lower voltage cells, those with a voltage of between 2 - 2.9V.

Or so I thought. Because I actually started picking cells from the 'presumed dead batch' and started to process those, and too my surprise they nearly always seem to recover to a performance which is similar to that of those which were stored at 3.7V.

For example, I took two cells, one at 1.159V and the other at 1.219V and left them connected to individual cell chargers for about 6 hours, after which I installed them in the discharger. They both operated on the discharger for at least 9000 seconds, which is a very good value compared to cells which were at 3.7V which never managed more than 9000 seconds.

So it looks like a lot of the cells which I decided to scrap just on the grounds of their standing voltage may actually be very good cells which will recover after a spell on the individual cell chargers, which seem to do a great job of slowly raising the cell voltage before going into high current bulk charge mode.

I have a few cells which I repeatedly, and charged and discharged them in different slots in the respective units. I did this to check the plausability of the readings I got from a single cycle and in every case the value for each cell was within +/- 150 seconds of the average for that particular cell so I'm comfortable with the thought that the process I've adopted is reliable.

I've sought out the tabbing wire I want to use, 8mm x 0.2mm which is probably overkill but thats fine by me, and the spacers which will allow me to pack the cells into interlocking blocks are freely available.


It looks like I have more battery work to do than I thought now, but I'm not disappointed by that !

User avatar
timpootle
Posts: 1362
Joined: Wed Mar 14, 2007 12:13 pm
Location: Chorlton-cum-Hardy, UK

Re: Another EV bike :)

Postby timpootle » Sun Aug 11, 2013 11:04 am

Good work Chris! Keep up the posts - you're not just talking to yourself ;-)
Tim Crumpton

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :)

Postby ChrisBarron » Mon Aug 12, 2013 3:08 pm

Thanks Tim, I just took a couple more photos.

The charging/discharging 'station' at work http://www.flickr.com/photos/28496053@N04/9493073001/

And the cells which have been tested so far are all in this next picture, each has a number on the side which corresponds to the number of seconds it sustained a discharge through the 3R9 resistor for, before it hit 2.9V

http://www.flickr.com/photos/28496053@N ... otostream/

Today I dug out a few more cells which had been set aside for the recycling bin at the local dump because they sat at below 0.5V, but I'm surprised and pleased to say some of them have recovered perfectly, although I suspect their cycle life may have been permanently affected by such a long period of time spent at such a low SOC.

Chris

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :)

Postby ChrisBarron » Tue Aug 27, 2013 8:46 am

I've passed a milestone in the last week by finsishing off the characterisation of all 18650 cells (apart from 8 more I found today !)

I did a bit of random retesting of some cells and the capacity values i get on the first run are very consistent, usually a second and third cycle sees an increase in capacity.

As a rough guage, using a cell analyser which can exert a constant load, a value of 6500 seconds on my discharger equates to a capacity of approx 1600mAh at 0.9A constant load. I have a lot of cells which are above a value of 6000 seconds, the majority are above 6500 in fact.

I've moved onto testing the 100 or so prismatic cells which came in some battery packs. They can be made to fit the 18650 cell holders if yoiu use a clean dead 18650 to fill the remaining space !
http://www.flickr.com/photos/28496053@N04/9603666537/

Also ... http://www.flickr.com/photos/28496053@N ... otostream/

All I can say about these so far is that I haven't found a dead cell or a cell which didn't reach the charger's charge completion point at first attempt. The capcities are rarely below 6000 seconds and most are between 6500 - 7500 seconds. That has surprised me, and encouraged me at the same time. These cells have a built in thermal trip switch (bimetallic ?), which opens of the cells are shorted. The trip doesn't open even with a constant 10A load.

All the cells so far can be seen here (the coolbox has about 600 - 650 cells in it), minus the 50 or 60 prismatics which are still being tested http://www.flickr.com/photos/28496053@N ... otostream/


Once the testing is complete I can get on with sorting and making 10 packs of 3.7V each from the best cells. I'll probably need to make a tab welder soon !

mattcarr
Posts: 389
Joined: Sat May 12, 2007 2:27 pm
Location: Hampshire

Re: Another EV bike :)

Postby mattcarr » Tue Aug 27, 2013 9:40 pm

Wow Chris, that looks like lots of work being done there.

I am thinking of putting together a few smaller cells to make a 12 volt battery to run my camping fridge. Want to make a simple BMS to cut it off from over charge and discharge, and I would like to assist the battery with my small mobile solar panel.

Matt

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :)

Postby ChrisBarron » Tue Aug 27, 2013 10:45 pm

Thanks Matt. It's rarely difficult work, but it requires attention to remove each cell, write on it and then load up the charger and discharger with more cells. Each cell has 2 tabs which need careful 'encouragement' to let go !

I've made a couple of BMS type devices, what skill base do you have ? If you can program a microcontroller life is easy, but if not the classic analogue designs are still valuable.

There are quite a few custom battery management IC's of a reliable nature. Just adding a few minimal components gets you going.

Then there are the ready made Chinese pcb's, which have charge current and overdischarge protection. I think I saw a 4 cell/3A unit for a couple of pounds recently. I have a couple lying around here somewhere.

If you're arranging your cells in'massively parallel' format, say 10 cells or more in parallel, then BMS is not as important than if you have just 2 cells in parallel or even strings of single cells, due to the averaging out of cell characteristics as you add larger numbers in parallel.

It's probably still good to add some form of temperature awareness/monitoring, I had a few cells which didn't finish charging even after 10 hours which got too hot to hold. Having said that the most common failure seems to be an increase in resistance or even open circuit, which is fine in parallel packs because the fauly cell won't necessarily short it's pack neighbours out.

Chris

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :)

Postby ChrisBarron » Wed Aug 28, 2013 7:57 am

Matt, these are the sort of things I was talking about..
http://www.ebay.co.uk/itm/PCB-Charger-f ... 906wt_2650


This vendor sells them in packs of 5 (They call them 3-cell units, ask about their 4-cell ones)
http://www.ebay.co.uk/itm/5pcs-Charging ... 2509wt_932


If you are looking for something with a high current then here's a 30A version for less than £20 inc shipping !
http://www.ebay.co.uk/itm/14-8V-14-4V-4 ... 967wt_1051



And 10A ones, for LiFePo4....
http://www.ebay.co.uk/itm/Battery-Prote ... 975wt_1170

ChrisBarron
Posts: 244
Joined: Tue Dec 11, 2007 11:26 pm

Re: Another EV bike :)

Postby ChrisBarron » Wed Aug 28, 2013 7:48 pm

I expect to have completed testing all the cells I have tomorrow and plan to have a beer when I do !

But today I've done a little more testing of one of the prismatic type cells and am pretty pleased

For the cell in question, when charged and discharged on my homemade rig the discharge time was 6859 seconds

On the battery analyser, 1A charge and 1A discharge produced 1697mAh in and 1656mAh out
1A charge and 1.5A discharge - 1663mAh in and 1629mAh out
1A charge and 2A discharge - 1651mAh in and 1506mAh out

The same cell is back on my homemade cycler so I'll get another 'seconds' reading after these tests

I think I have about 120 of these cells capable of at least 1500mAh at 1A, which seems like a useful lump of capacity for about 4.5kg additional weight (each cell weighs 39g).

In the next few days I need to group the cells in terms of their capacity and count how many of each I have in order to plan the best way to build up the pack, as well as estimate the useful capacity.


Return to “Electric Motor Bikes”

Who is online

Users browsing this forum: No registered users and 14 guests