Hacker News Reader

Comment by dheera | Hacker News Reader

I've thought of this but people also would need to use something durable, e.g. carbon-fiber-reinforced nylon.

You definitely don't want your battery pack shattering if you drop it.

The 3D printer world today seems to have an unfortunate love affair with PLA which is for both temperature and durability reasons a very bad choice for battery pack enclosures, and a bad choice for just about everything else. I'm really not sure how to get the word out there that PLA is outdated and should be deprecated for just about every 3D print use case. (Personally I use PETG For just about everything, but for a battery pack I'd probably want something carbon-fiber reinforced.)

View on HN

I think the larger concern should be fire safety. Normal 3d printer filaments don't have the fire retardants required by UL/etc for power electronics. I've experienced power supply and battery failures that were contained by their enclosures enough times in my life to take this very seriously. Had any of those cases been wrapped in common PLA/PTEG/ABS from a 3d printer, i'm sure it would have been much worse.

So, if your printing anything that has a likelyhood of catching fire you should probably be doing it with something like: https://www.3dxtech.com/flame-retardant-filaments/firewire-f... or just wrapping that part of the design in sheet metal.

How does this flame retardant ABS work in the event of a lithium battery fire? Are there flame retardant PETG? Are they any better than just printing with normal filament and spraying flame retardant on it?

I've been hesitant to use lithium batteries in home robotics projects, but I can't find charging and power mux boards for NiMH, so that has been a huge blocker in making robots that charge and operate at the same time.

Also I wonder if it is possible to put mini pressurized CO2 balls inside the battery enclosure such that if a fire happens they explode and release CO2, putting out the fire?

I suspect to have the right answer you will need to read the relevant standards/cert and understand how it interacts with your design.

But the general theory as I understand it generally is not that the plastic may not burn, only that they are self extinguishing. V-0 like that filament means it will extinguish itself within a maximum of 10 seconds after the ignition source is removed.

Regular filament acts more like an accelerant, put a flame near it and it burns quite vigorously, and for a long time. Meaning anything near it that can burn will likely catch fire too.

> Also I wonder if it is possible to put mini pressurized CO2 balls inside the battery enclosure such that if a fire happens they explode and release CO2, putting out the fire?

No, that may momentarily put out the flames but as long as the thermal runaway condition is still present it will reignite. The only way to stop a battery fire, which is actually a metal fire, is to permanently suffocate it - usually by dumping heaps of salt, cement or salt on it which melts and so removes oxygen from the fault area - and then, at least for electric cars, to dump the car in a container full of water and keep it there until dismantling, so that the water acts as a thermal buffer to prevent reignition.

You also have to realize things like this have to take scale into consideration. No one is going to do mass manufacturing of battery and electronics enclosures with 3D printing, this is a prototyping and hobby level manufacturing system. So the odds of injury and damage are a lot less with this as you automatically self-select down to a tiny fraction of the market share, and only a tiny fraction yet of those will ever experience an explosive outcome from a battery.

Also I'm sure that most of this market violates someones IP so again, this ensures that this type of thing (custom making battery enclosures) will never reach a critical mass where worrying about the world burning down due to non-fireretardent plastics is something to be concerned about.

But if a person is very concerned about this issue then certainly there are options out there.

I've become a large user of PET-G for some of the reasons you outline. I've never had a problem with strength the way that I use it in my designs, PET-G is plenty durable for a small device which spends 99% of its life inside a larger device. Also the heat resistance makes it tolerate the summer heat of cars which is also important.

Also PET-G has the added benefit of not giving me cancer like ABS. It definitely has a learning curve and I wouldn't fault anyone for going other routes rather than taking the time to get a feel for it.

Yep, same reasons.

My printer prints PETG pretty cleanly with no fuss. I imagine if someone used the same exact printer (Prusa i3 MK3), the same exact filament (eSun), my same exact settings (https://github.com/dheera/3d/blob/master/settings/slic3r-con...), and my same exact textured bed sheet, same exact procedure (Windex, then Magigoo, then print), they should have zero problems as well. It's not even tricky. It prints just like PLA with this "recipe".

The only problem seems to be that there is a lot of conflicting information about what works for PETG but people just need to think in terms of configuration sets like the above. If you use the Prusa PETG settings with eSun filament it fails miserably -- the prints come off the bed mid-print.

> I'm really not sure how to get the word out there that PLA is outdated and should be deprecated for just about every 3D print use case.

It works fine for 90% of use cases and is easy to use. Aside from that, you can buy virgin PLA filament which is pretty darn nontoxic. Most other plastics have a variety of negative effects on humans, from physical interaction with the finished product but also from the printing process.

There is a slightly higher difficulty with petg, my first print was with pla and had no issues, with petg I have had slight heat shrinkage and adhesion problems

My former employer wraps battery packs in flexible fiberglass sheet before potting them in epoxy; it makes them shatter-resistant.

Reply on news.ycombinator.com