Something we should ALL be aware of

Never leave your charging batteries unattended…

3 Likes

The only charger I trust is the B6AC. I sure wish they’d come out with one that could handle more than 6 cells…

Totally agree, my B6AC never failed on me for the last 5 years. Lots of copies on the market though, new buyers should make sure they purchase the original one.

I only charge in the garden!

It’s not just the charger that you have to worry about.

The issues of high powered lipoly around saltwater are many and serious, and all batteries are vulnerable to workmanship issues from the factory.

I’ve been looking at some ways to make Lipoly safer. Everyone tells me that 18650 is safer (or other cylinder cells,) but I don’t think it has to be that way. The metal casing is a big plus, but they can still go critical and spray; look at the hoverboard fires. Lipoly has a power and weight advantage and is easily available as far as i know (can someone comment on that?) That’s a big advantage.

I believe we can address the safety problem holistically with four steps of safety:

  1. Prevent faults as much as possible through good design (naturally)
  2. Detect a problem when it (inevitably) happens
  3. Manage the problem as much as possible; alerting the user, possibly disabling charge, ducting gas and volatiles
  4. Educate owners not to charge without supervision, and to store charged packs away from fire hazards.

So in Prevent you could

  • good BMS design with balancing. many hot fires are from over charged cells in a stack.
  • over current protection, short circuit detect/protection, soft-start/load detect, under voltage detect
  • an aluminium casing as puncture is a huge issue
  • waterproofing obviously, coating of all circuits where possible
  • power enable over comms/protocol a good idea but not for everyone

In Detect you could

  • water leak sensor; salt water shorts is a high fire risk, and can be latent from corrosion
  • perhaps a drop/shock sensor
  • logging of data so you can see faults in battery, and in downstream
  • a full range of cell voltage monitoring that is logged to see later on

In Manage you could

  • audible alarm
  • disable charging, possibly disable discharge if a case for it
  • management of a fault, which may including overpressure rupture point (like a breather valve) allowing a predictable path for hot gas/flame
  • and an aluminium diffiuser in that path to cool gases escaping, which could stop them burning when they hit oxygen/air beyond the diffuser

In Educate you could

  • still thinking this part through. I wonder if an app that enables charging after you’ve ticked some boxes to say your not charging in your living room is a good idea? like the one DJI use to make sure you know the drone safety rules.

I’m scratching together a system around a Texas Instruments BQ76930 balancing front end, one per pack which is a pseudo distributed BMS. That’s the heart of it, the rest is a host micro, some comms, logging. I have some micro stuff that has logging on SD card. Bluetooth maybe but I’m looking at some new parts for Wifi+Google Cloud+phone gateway. Im going to pull it into a system for my own needs but if there’s interest I might release open source for other to pass around and improve.

Any and all comments or ideas I’d love to know.

1 Like

I agree with you the way to look at safety , but still i will prefer 18650 use for safety reason, mainly because with 18650 you know what you buy , lipo you don’t ! At the end it is all base on the company brand trust : gens ace … And there is not safety data available like 18650

For the Samsung

https://eu.nkon.nl/sk/k/30q.pdf

I all ready blew up a lipo ( 30% charged 16a 4s) and a couple 18650 ( Fully charged) , i prefer 18650 de gas than lipo fire

Both can be use in a efoil in a safe way , i think it is up to you what you feel more confortable lying on .

Similar to @Alexandre I am using 20700’s. I’ve ridden my esk8, with no battery cover at all, in an unexpected rainstorm and had no troubles at all - the board died, but it was the anti-spark switch that failed. The batteries and motor survived without issue.

That said I agree with everything you say.

I wouldn’t use LiPo mostly because I don’t know much about them, but also because of the vast amount of power that you are packing into the box. Those big LiPo packs have no protection whatsoever against neighbor cel failure so when it goes up it’s going to go up big. And while some DIYers are well versed in battery management some not so much.

I’m nervous about BMS’s onboard - I can see where a large amount of moisture or un-detected standing water could cause one of them to fail. At the very least I’d conformally coat the boards and use Dow-4 on all connections, but I think I would prefer an offboard BMS for charging (like my B6AC :slight_smile: ).

Lastly In my case I have a bit of extra space in my battery box so I will be physically separating each parallel bank by a small amount. This probably won’t make a difference, but the theory is that if one bank goes up maybe the neighbors will be spared. Also in my case the cells aren’t packed too close together so there is some air-cooling advantage.

Agreed on 18650 data sheets.

But if u read them the testing is excellent, and they never catch fire under a battery of tests… yet they still occasionally catch fire in real world. So a grain of salt?

And all 18650 packs have abutting cells and thermal runaway risk of neighbors.

But in reality I don’t have time for 18650 welds.i want my big hobbyking special lipoly with armour taked on. There are many other engineering challenges I’d rather focus on.

All electronics in efoils or auto apps needs to be waterproof. Use the ip67 connectors, and pot with resin or a good box :slight_smile: hard work but worth effort.