Icon-YouTube(250x93px).png
Icon-LinkedIn(250x93px).png

© 2017 Ian Watts

Icon Projects(48x49px).png

Lithium Battery Packs

...using 18650 cells...

I never like to 're-invent the wheel'. Indeed, what's the point ? Unless I truly believe I can make a better one, or more realistically, one that's better suited to, or further optimised for, my intended use.

When it comes to battery packs... well... whilst I don't want to and don't have the technical expertise to re-invent a battery (I'll leave that to others) I do find myself wanting to :​

  • remain flexible in terms of physical layout / size

  • remain flexible in terms of voltage / capacity

  • select a battery technology that I (kind of) understand and transfer that understanding (which is always evolving) across multiple projects

  • be able to re-use / re-task batteries should a project become obsolete

  • be able to 'upgrade' / modify a battery solution to suit ever-changing requirements. Quite often I find myself re-assessing my initial plans and changing a design when a plan doesn't quite meet expectations for some reason !

  • reduce costs of bespoke battery pack manufacture and (often, but not always) build time is faster than placing an order and waiting for delivery. 

Whilst there are undoubted benefits to the 'DIY' approach (above) there are some 'downsides' :

  • time to completion of first pack. It took quite a while to investigate the plethora of options :

    • cell formats : pouch, prismatic, cylindrical,

    • cell size (voltage / capacity),

    • cell chemistries (SLA, Lithium manganese oxide, Lithium nickel manganese cobalt, Lithium cobalt oxide, Lithium iron phosphate, Lithium Titanic Oxide...)

    • cell interconnections (solder, spot-weld)

    • cell fusing (to fuse, not to fuse)

    • pack BMS (include one or not)

    • which battery charger ? (purchased cell packs generally come with specified charger options, not so with the DIY approach)

    • what final packaging (tape, heat-shrink, custom-made (plastic or aluminium), 3D printed)

To be fair, subsequent pack manufacture is much quicker once the above decisions have been made.

  • cost to build / the equipment to manufacture a quality and safe pack :

    • soldering iron (low wattage (wire connections) and high wattage (80+watts for cell caps and o/p terminations))

    • spot welder (should one decide to spot-weld)

    • glue gun (to hold the individual cells and misc parts in place during manufacture)

    • heat-shrink (for connections and maybe final pack wrapping)

    • outsourcing / 3D printer (for any final 'packaging')

As with 'Time', above, the cost to build a single pack can be (and often is) 'expensive' BUT if multiple packs are to be built, then the initial equipment outlay can be spread across the first and all subsequent packs.

In order to move on, I split battery manufacture into various sub-headings... detailed below :

Battery type / chemistry / composition

Following my research into battery chemistries and formats, you can read my reasons for selecting 18650 format, LiFePo4 as my chemistry of choice <Here>

Solder or Spot-weld ?

Solder or Spot-weld the individual cells ? You can read my thoughts and what I opted to do <Here>

To fuse or not to Fuse ? - That was my question

Following my research into fusing or not fusing individual cells, you can read my reasons for selecting to fuse <Here>

Battery Management System (BMS)

Following my research into Battery Management Systems, you can read why I use a BMS on EVERY pack I build <Here>

Isolation, Pack Protection & Charging

My output isolation (On/Off), protection and charging methods are detailed <Here>

Final Packaging

Sketchup design and 3D printing results can be seen <Here>

Improvements (future work)

Future improvements would include :

  • a coulomb counter (battery meter / fuel gauge), preferably to link directly with the onboard / original fit fuel gauge.

  • Heating. Probably limited to heat-during-charge only and if required. Packs could then be left in cold conditions and, if a charger is connected, could be kept warm (maybe a PTC heating element, heatsink, fan and a simple logic circuit).

  • Cooling. EV packs would undoubtedly benefit from being cooled during hard use / high temperature conditions. The idea would be to employ the same (heating) fan but without the heating OR, possibly, the peltier effect to achieve heating OR cooling as required...

... Just thoughts...