Oil immersion?

Ivo Staelens

Member
Joined
Aug 21, 2017
Messages
109
Good evening everyone,


Long time since I posted, but on the background I keep following everything. I'm also still building and collecting cells.

In the long run, I want to go off grid with my entire house and fire safety is a big concern for me.

I plan on putting the 14s powerwall outside but I also plan on immersing it in vegetable oil. I would be testing it with a small cell I built but I reason that a setup completely immersed in oil would avoid thermal runaway if one 18650 would go thermal.

Is my reasoning correct? Anyone ever tried this?
 

Sean

Member
Joined
Oct 8, 2016
Messages
946
Here's a clue to help you decide if your concept is flawed, or not.


image_jzwyxu.jpg
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,329
Even if the oil is absolutely non-conductive, the problem is "if" a thermal even occurs. If a cell goes all sparkler, it will be FAR higher temps than the flash point of the oil. In the end, you'll have a whole container of oil on fire and no way to put it out. And it doesn't matter what kind of oil you use, all oil is flammable at some temperature, and that temp is lower than what lithium will burn at.

Now, if you want to help make the inverter/charger a little more efficient under loads, then these could potentially be put in oil.

But, overall, I would recommend against oil submersion.
 

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
If you do plan to use oil, it has to be mineral oil. Organic oils will eventually go rancid. Apparently the smell of rancid oil is horrendous.
 

Amish_Fighter_Pilot

New member
Joined
Jul 15, 2019
Messages
14
It would be better to arrange the battery in such a way that cell fires remain isolated and that cells are immediately disconnected from the pack. I'm still pretty new to the world of battery building, but these things burn very hot if they ignite. Maybe some sort of ceramic separators between cell groups? Perhaps even better would be to cast sodium silicate into all the voids between cells(but definitely not blocking the positive end of the cell!). Really, any castable refractory would be good as long as it isn't a conductor or semiconductor(so no silicon carbide!)
 

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,239
I wouldn't do oil immersion for similar reasons as Korishan's post above.
Also any kind on fault fixing, measuring, maintenance, etc becomes a total nightmare.

There's several threads about fire proof battery boxes, etc:
https://secondlifestorage.com/t-Fire-explosion-proof-cabinets-for-safety
https://secondlifestorage.com/t-DIY-firesafe-casing-Test
https://secondlifestorage.com/t-Safely-housing-the-battery-bank
Separate building, eg a garden shed &/or 19 inch equipment racks &/or steel cabinets are a good start too.
 

completelycharged

Active member
Joined
Mar 7, 2018
Messages
1,008
Thermal runaway for an 18650 cell can start between 120C and 230C (1) so the cell has to be hot enough to burn your fingers and some spacing with an air flow would eliminate cell to cell thermal runaway (1), however typical powerwall builds are relatively tightly packed along with some battery packs with zero cell to cell spacing typically where nickel strip spot welding is done, hence the examples of a full thermal cascade of every cell in some packs.

Air spacing can be enough on it's own (to prevent a cascade) with adequate cell spacing then you only need to consider what the short duration flame from the valve gas release can set fire to nearby..

1. EnerSys 2016 test data for NASA
 

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
663
I agree on strongly against the idea of oil...
I am busy with some experiments for the case of a thermal runaway.
So far i have 3 setups for sand.
Even a setup is in the make with a sprinkler system, one filled with water and the other one i want to fill it with a special kind of fire extinguishing fluid,fairly new stuff, especially designed for li ion fire.

AAAAND i can not find the link anymore.......


its called f-500
https://www.pengsafety.nl/brandblus...ltr-speciaal-voor-lithium-ion-batterijen-769/

My link had 3 9 and 500 liter, was a store in The Netherlands, closer to you btw.
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,272
Agree with all the comments. And in particular, the 'practical aspects'... as I fool with my packs all the time. Not every day but every couple of months I'm adding a few cells to or swapping out a pack or building new ones or increasing the wiring size or 'something'. The message is that you want your battery bank as easy to access/maintain as you can within your design parameters!.

@competelycharged said
>some spacing with an air flow would eliminate cell to cell thermal runaway
I've used the standard 4x5 plastic 18650 cell holders. Just out of curiosity - does this qualify as adequate 'air flow' system to match your comment?
 

completelycharged

Active member
Joined
Mar 7, 2018
Messages
1,008
Search for "Thermal Runaway Testing of 18650 Lithium Ion Cells - NASA" it is a PDF with full testing.

I think the standard plastic spacers are too close together AND the plastic is flamable so when the cell heats up on the outside and starts to break down the plastic that adds to the combustable gasses.

If a single hot cell is the source ignition point with a suitable air gap it should be safe, however the air gap has to be all round and not just cell to cell.. If a single heater cell is allowed to have some time to warm up neighbouring cells, due to the lack of airflow then a cascade becomes more likely as the incremental heat required to set off the next cell is reduced.

The ideal pack may well end up larger, however it would be safer so depends on what the trade off is between build speed/parts/quality/ease to build/asthetics vs risk.
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,329
100kwh-hunter said:
https://www.nasa.gov/sites/default/files/atoms/files/therm_runaway_test_18650_li-ion_clobato.pdf

Reading it right now...very interesting, already got some answers.
Thanks completelycharged

Reading over it as well. Interesting in the graph with the summary of the 18 tests, it shows that during TR (thermal runaway) the temperatures experienced are very unpredictable.
The only safe way of containing any of these TR events would be that each cell would be vented into a steel tube, and the tube would need to be over the Pos end slightly as sometimes the whole innards of the cell gets ejected (depending on cell can design) :s
 

bogptrsn

New member
Joined
Oct 27, 2017
Messages
29
Wouldn't the oil prevent any sparks fromigniting the venting gas? Then the cells can get as hot as they like but no explosive fire.
 

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
663
Agree on the steel/alu tube, but in case of some sparks to prevent a cascade, to prevent ashort circuit,
There must be another solution added, like sand on top of your pack.
Or in stead that they are "standing", put them on their side, then would a steel tube be effective.

In my case for 100kw storage it would be impractical to cut or order 14.000 steel tubes and make cell holders out of them.

If you place sand on top of your cells and you get a TR the sparks will not short circuit the pack.
But the heat that is generated by the TR cell can not escape, and thus can light up the sleeves of the other cells.

To make 140 (in my case) buckets with a soft plastic bottom and fill that with sand, its for me also not practical.
It could work.
A TR must blow fire/sparks to let the plastic melt, so the sand will go down on the cells.
But as we could see, some TR only had some smoke, for sand in plastic bags on top of your cells you need some fire.
So it won't be hot enough, so no need for a melt.

And the certainamount of sand needed to cover a pack.
Also when the sand ispouring down, must be hold to cover the pack, so i does not flow away.
Sand has a other problem, weight.

To be practical, for a safe fire preventing solution i think one or two sprinkler per pack, with that fl-500 stuff.
The standard sprinkler would go off at 68C for 1,50E each.

In all the above cases, you need to give each pack its own room or if something might happen, lose your whole pw, instead of one pack.

Oke we make sure that we only keep the good cells, we fuse, and do our calculations.
Even with all that safety connected, it could be possible for a TR, just....it....might....be...could...be...happening.
Even if there isone cell that could go into TRand there is a verybig chancethat it will stay with that one cell.
What if that one in a million shot is going to happen???
On yt you have some vids of 100 cells, 500 max, on security cams.(bigger??? show me, please)
But now 14.000 cells???

Soon i will have the projects done and i will post some pictures and results, time.....also waiting for the rubber boat from china.

Best


If a cell goes into TR, it can ignite the oil, you really don't want that.
Most oils will dissolve some plastic over time, and get rangid.

Best
 

chuckp

Active member
Joined
Jul 29, 2018
Messages
343
100kwh-hunter said:
I agree on strongly against the idea of oil...
I am busy with some experiments for the case of a thermal runaway.
So far i have 3 setups for sand.
Even a setup is in the make with a sprinkler system, one filled with water and the other one i want to fill it with a special kind of fire extinguishing fluid,fairly new stuff, especially designed for li ion fire.

AAAAND i can not find the link anymore.......


its called f-500
https://www.pengsafety.nl/brandblus...ltr-speciaal-voor-lithium-ion-batterijen-769/

My link had 3 9 and 500 liter, was a store in The Netherlands, closer to you btw.



I've also been looking at F500 for a fire suppression system, my initial thoughts were for a glass bulb type sprinkler system connected to a pressurised vessel containing a 3-6% solution of F500.
This vessel would be connected to a nitrogen cylinder for the system propellant.
The sprinkler line would have a solenoid valve that fails open, this would be controlled via a fire detection system and held closed, once a sudden heat rise and or smoke was detected it would de-energise the solenoid and allow the F500 mix to the sprinkler heads. The sprinkler heads are now primed ready to actuate if the temperature continues to rise.

As this would in a sense be a two knock system it should prevent any accidental discharge from the system. i.e. accidental damage to a sprinkler head, false alarm or power failure at the solenoid etc, it would need a fire detection and a heat source present before the F500 could be released.

I have almost all of the materials needed to implement the system already, I only need to find a UK supplier for the F500 encapsulation agent in the smaller quantities i require.

Also getting the battery house built so I have somewhere to install it might help LoL.
 

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
663
@ chuckp:
I have almost all of the materials needed to implement the system already, I only need to find a UK supplier for the F500 encapsulation agent in the smaller quantities i require.

I like your idea of the valves.
Its one sprinkler that is going off, the pack does not need 100 litres of that stuff, 2 max.

But oke, if i am going to buy, i can send you some? if you would like of course?

With best regards
 

chuckp

Active member
Joined
Jul 29, 2018
Messages
343
100kwh-hunter said:
@ chuckp:
I have almost all of the materials needed to implement the system already, I only need to find a UK supplier for the F500 encapsulation agent in the smaller quantities i require.

I like your idea of the valves.
Its one sprinkler that is going off, the pack does not need 100 litres of that stuff, 2 max.

But oke, if i am going to buy, i can send you some? if you would like of course?

With best regards

Thanks thats great.

I already have everything needed to put in a nitrogen sprinkler system.
Adding the F500 would only be an additional solution storage vessel, adding a solenoid valve that Ive probably got somewhere and a fire detection panel, smoke and heat detectors, I can grab those from work.

Actually now I think about we have an old stainless steel water softener kicking around that I could remove the brine tank from.
 

Bubba

Active member
Joined
May 9, 2018
Messages
272
chuckp said:
Thanks thats great.

I already have everything needed to put in a nitrogen sprinkler system.
Adding the F500 would only be an additional solution storage vessel, adding a solenoid valve that Ive probably got somewhere and a fire detection panel, smoke and heat detectors, I can grab those from work.

Actually now I think about we have an old stainless steel water softener kicking around that I could remove the brine tank from.


Here is an snip from an article about Lithium battery fires:

The FAA special conditions document explained, In general, lithium batteries are significantly more susceptible than their [nickel-cadmium] or lead-acid counterparts to internal failures that can result in self-sustaining increases in temperature and pressure (thermal runaway). This is especially true for overcharging, which causes heating and destabilization of the components of the lithium-battery cell, which can lead to the formation, by plating, of highly unstable metallic lithium. The metallic lithium can ignite, resulting in a self-sustaining fire or explosion. The severity of thermal runaway due to overcharging increases with increased battery capacity due to the higher amount of electrolyte in large batteries.

https://www.ainonline.com/aviation-...thium-fires-generate-myths-and-misinformation

I noticed the Self-Sustaning bit. This means that Nitrogen will not work. Nitrogen's purpose is to displace Oxygen. Self sustaining reactions will still burn in a Nitrogen rich environment.

Sounds like cooling is the best thing to do and that F500 Encapsulator Agent sounds interesting though.
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,329
Due to the compounds in the electrolyte, Oxygen is present, this is why it's self sustaining. Agree with Bubba on the cooling is the only option. Using hadron or other suppressants have been topics of discussion before and it all leads back to this: Keep the cells cool at all times. This is why keeping an eye on the temperatures is so important with lithium-cobalt (and it's close cousins)*. If temps raise too high, kill the charge/discharge to those cells. It's better to loose power in the house/equipment than it is to have a fire you can't control.



* - there's also lithium manganese and other variants that are susceptible to these fires. Lithium Iron (LiFePO4) is one of the few that have a hard time igniting and is far safer than the other chemistries. This does not meant they "won't" catch fire, but that it is much harder to do so.
 
Top