Tesla modules in boat

3283 Views 80 Replies 10 Participants Last post by remy_martian, 2 mo ago

Ohno

I am converting a 25 hp gas outboard motor to an 18kW electric PMAC motor for use in a hydrofoil assisted power catamaran. Converting an outboard to electric is something I've done a couple of times but this time I'd like to make my battery safer. My last electric boat is a little lacking in the safety department - Electric Foiling Catamaran from Recycled Parts.

This time I'd like to use 6 Tesla Gen2 XS modules. 3 modules will go in a box on each side of the boat. This is a 72v system so I'm thinking that the 3 modules in each box will be wired in series then the two boxes combined parallel.

What I'm hoping you very knowledgable people here can help me with is battery box design. My thought is to make a watertight (within the limits of the IP67 rated vents) steel box out of .050" - .080" thick sheet that fits fairly close to the size of the 3 modules. That steel box would have a battery box vent Vent valve for battery box - EVcreate
and a layer of furnace insulation on the outside. The steel boxes would be mounted inside of composite boxes that are used as seats and will also hold BMS, contactors, fuses, emergency shutoffs, etc.

This is a carbon/honeycomb boat and I'm fighting to keep the weight down so that I can drag around those heavy batteries without too much performance loss. It kills me to think of 100 lbs of steel boxes onboard but I'm not seeing a way to mitigate the fire risk without something like that. It sounds like aluminum's melting temperature is not up to the job.

Any thoughts? Am I overthinking this? I'm not pulling a lot of current, maybe 200 amps continuous. I think that's like .5C. Charging C rate would be substantially lower.

Save

Reactions: 1

1 - 20 of 81 Posts

Ohno

Discussion Starter · #3 · 4 mo ago

reiderM said:
For what it's worth, if the modules catch on fire, you're not going to stop the fire from spreading regardless. When a factory EV catches on fire, it's often just allowed to burn out on its own since lithium batteries supply their own oxygen which makes extinguishing them exceptionally difficult. The battery box vent just prevents the battery box from becoming a pressure cooker and possibly exploding in the event of a fire (not ideal!). So whether the box is steel, aluminum, plastic, or otherwise, you're in for a very bad time if the modules catch.

An ounce of prevention is worth a pound of cure. There are only two ways which will cause the batteries to light: physical damage to the cells (eg. a puncture), and overcharging above 4.2v. Lithium batteries can also thermal runaway above about 70 C, but I have a hard time seeing a scenario in which your modules would get anywhere near that threshold given the discharge and charge rate you mentioned.

With all that being said, you certainly don't have to stick to steel. Many folks have used aluminum without issue.

What are you using for BMS/charger/fusing/contactors etc.?

Thanks very much for your thoughts. I would happily give up the idea of building a steel box if it won't provide enough, or any, added safety. That said, I have to believe that a battery fire could be contained in a steel box if it was heavy enough but that it's not worth the expense and weight for EV manufacturers to do that.

Probably the safest thing would be to use LiFePO4 batteries. They are heavy and stupidly expensive though.

I have questions about what BMS, charger, contactors, fuses and switches I need at the batteries. So far my plan is to use an Orion BMS, a fuse on the negative cable of each battery box, and maybe a power cut off on each box? The contactor is inside the outboard motor cowling along with the controller and motor. Those are the basics at this point. In addition there are various gauges, motor/controller coolant system, DC-DC converter, etc.

Here again any ideas you have about this stuff would be much appreciated. For instance: My two battery boxes are 4' apart. Each has a 72v battery. The boxes are connected by a conduit. My plan is to connect those two batteries in series. Can I use one Orion BMS for this set up? (Note: I edited my original post I switched "parallel" and "series")

See less See more

Save

Reactions: 1

Ohno

Discussion Starter · #6 · 4 mo ago

reiderM said:
Yeah, I mean if the metal is thick enough at a certain point of course it will be a non issue. A foot of steel will definitely stop the spread of a battery fire inside (and you can probably ditch the blowoff valve at that point too)! I like the idea @kennybobby mentioned with the nested boxes and a vent, if you really want to be safe. But rest assured that many people have li-ion systems with aluminum siding in the mm-range that have run for ages without issue. Tesla modules also have cell level fusing which is nice in case of a cell failure etc. Everyone's risk tolerance is of course different, so ultimately it's up to you. Li-ion batteries when managed properly are quite safe (see the fire rate of EVs versus ICE vehicles).

So you'll have a 144v system? Or a 72v system? 6s would be 144v, 3s2p would be 72v. Keep in mind that a BMS like the Orion cannot manage separate paralleled strings. I'll plug the SimpBMS here as I am the retailer for it in North America (so I may be a little biased!) as it can manage paralleled or series modules without issue and is a good bit cheaper and simpler than the Orion for Tesla modules. If you are interested in that just shoot me an email at [email protected] and I'll send you all the docs on that etc.

The BMS should control at least one contactor separately as well to disconnect the battery from any load in the event of a battery failure - this is a key aspect of safe battery operation. The BMS should also control charging to prevent overcharging.

I am leaning toward making a thin wall (.090"?) aluminum box with a welded aluminum frame made of whatever shape extrusions are needed to hold the batteries.

The set up is 3s2p. Since I have two battery boxes I suppose I need a contactor in each one connected to the BMS?

I am definitely interested in the SimpBMS. I see a link to the documentation on eBay. I suppose that's the same documentation you're talking about?

Save

Reactions: 1

Ohno

Discussion Starter · #7 · 4 mo ago

kennybobby said:
From what i remember, the 2 boeing 787 screamliner APU starter battery fires were contained in aluminum boxes; their "fix" was to build a vault to hold this 36V 8s lithium pack with a vent tube to exit the airplane outer skin in the event of another fire. There are other lithium packs for other functions in the electronics bay, but they are not in any vault and they don't have the same 4-level bms, and probably aren't held on a full-voltage trickle charge at all times during a flight.

i think you could build an aluminum battery holder and be fine with a good bms and attention to not over-charging. And a good fire and smoke alarm system to let you know if you need to get out quick.

What sort of duty cycle would you have on the electric motor operation or its purpose? Once the wind is up a catamaran would haul ass against any electric motor prop drive wouldn't it?

I expect to run the motor at around 200amps depending on what the 'sweet spot' ends up being for speed vs power consumption. This is a powerboat so no sails are involved!

Save

Ohno

Discussion Starter · #9 · 4 mo ago

electro wrks said:
In one of Tesla's original patents, it states that their battery fires can burn at 850 C (1562 F). As aluminum melts at ~660 C (~1220 F), it might be a poor choice as a material if you don't want your battery box to potentially burn through and melt in a fire. Stainless steel melts at ~ 2X the temperature of aluminum. It might make a better choice, particularly in a marine environment.

That 850C is a really interesting number.
I wonder if there is an example of a battery box containing a battery fire?

Save

Ohno

Discussion Starter · #11 · 4 mo ago

reiderM said:
Yep, all documentation is there. Plus I can do a better price ($20 less) off eBay since eBay's fees are quite high for sellers.

Not necessarily a contactor in each one, but you should definitely have a fuse between paralleled strings. IMO it's better to leave modules permanently paralleled when there are cell-level fuses (as is the case with Tesla modules) but the opinions on this vary. I can give you more details on the pros/cons of that if you're interested

i'd be happy to leave off the contactors. The less unnecessary electrical stuff on a boat the better I say.

So, a fuse between paralleled strings. Also a fuse to protect the battery to controller cables? I'm also wondering if I need some kind of manual shut off at the batteries?

The SimpBMS looks like the way to go. I'll contact you about that as I get all of my ducks in a row.

Do you have any thought about a reputable source for Tesla modules? I'm looking at these - TESTED Gen 2 Tesla Model X S 85 90 Battery Module 24V 250Ah 5.3 kWh w/ Warranty | eBay

That looks like a 13' Whaler in your photo. Classic!

Save

Ohno

Discussion Starter · #18 · 3 mo ago

oudevolvo said:
Still need to do a writeup on my root cause analysis on my fire.
Regarding battery boxes: the fact that I used 2mm stainless steel battery boxes with vent valves are probably why there is a car to rebuild.
It contained the fire of three Tesla modules until there was nothing to burn inside anymore.
And the front tires still held air.
Thought on the root cause I’ll share on my rebuild blog Electric Volvo Amazon wagon from 1967 and converted to 100% electric but still want to test some hypothesis.

Thanks for responding. It looks like you did a fantastic job in building your car. I am sure like many people I would love to read about the cause of your fire.
Have you shared anywhere photos of your battery box?

Save

Ohno

Discussion Starter · #19 · 3 mo ago

kennybobby said:
In a catamaran with packs in each hull, should the connecting wires be de-energized when the vehicle is Off, or will they be Hot all the time?

If using used modules from a wrecked car, then just assume that you will have a fire--are you wanting to save the vehicle or just save your life? Is it just a run-about type of boat or a live-aboard?

An EV has safety aspects to protect first responders in the event of a crash that a Boat doesn't necessarily need, just depends upon your level of risk tolerance

I don't yet know enough about this stuff to say if there is any value in de-energizing the wires between battery boxes. It seems like the power will be on whenever there are people onboard.

This is a runabout. Really if the boat is consumed in flames there is no good place to go. That's why my OP - should I build a battery box that contains a fire? I guess "how safe is safe enough?" is a question that only I can answer in this case. I mean, I don't wear a parachute in small planes even though they occasionally fall out of the sky.

Save

Ohno

Discussion Starter · #22 · 3 mo ago

remy_martian said:
In the case of a boat, I'd cut a square hole in the bottom of the boat then seal a stainless steel battery box with a plastic bottom to that hole - in the event of a fire, the flaming batteries get delivered to Davey Jones locker.

Ha ha! I've seriously considered that. As usual the devil is in the details. For one thing there's a hydrofoil directly beneath the batteries. The batteries are in two boxes on deck. Floating them off with a helium balloon might be easier. 300lbs each though? Maybe not.

Save

Ohno

Discussion Starter · #23 · 3 mo ago

oudevolvo said:
Thanks, yes, in this blogpost https://www.oudevolvo.nl/en/blog/2018/09/27/stainless-steel-battery-boxes-for-tesla-modules/

And when I stopped blogging updates can be found on Instagram @oudevolvo eg around

http://instagr.am/p/CNErkAKH7hG/
And on Facebook OldVolvo

Your build is very impressive. What beautiful work.

Can you think of any way that you would change your battery box design to better contain a fire?

Save

Ohno

Discussion Starter · #26 · 3 mo ago

remy_martian said:
The wands for height control are brilliant...I'm now wondering if there's a mechanical wave height averager that could be introduced using adjustable-valving shock absorbers (or even corn starch dampers, lol).

Smarter people than me have figured out how to control ride height, roll and pitch electronically. I'm thinking all the sensors that control a quadracopter would do the job. Somehow tie those things to servos and hydraulics.

My new boat skirts the whole control issue by keeping the hulls slightly in the water.

Save

Ohno

Discussion Starter · #27 · 3 mo ago

kennybobby said:
In a catamaran with packs in each hull, should the connecting wires be de-energized when the vehicle is Off, or will they be Hot all the time?

If using used modules from a wrecked car, then just assume that you will have a fire--are you wanting to save the vehicle or just save your life? Is it just a run-about type of boat or a live-aboard?

An EV has safety aspects to protect first responders in the event of a crash that a Boat doesn't necessarily need, just depends upon your level of risk tolerance.

Cool boat build

I don't know. Do car guys de-energize cables that run between battery packs?

Save

Ohno

Discussion Starter · #33 · 3 mo ago

oudevolvo said:
Thanks!
In the first place my fire mitigating action will be not to use Tesla batteries anymore.
Nowadays there are safer batteries that will give me the right volume, weight and kWh at 96s.
There are two things that could have been better on the box I made:
1) Smarter vent valve positions (so the exits would be where even less damage would occur).
2) A hose connection (not sure how to seal/close that when not in use) onto the box that the firemen could have used to put water inside the box to even better cool at the exact right place

What an interesting idea to provide a fire hose connection. It might be asking a lot that a firefighter would have the presence of mind and courage to connect to it but I can picture a fill pipe that exits at the car side with a valve that can withstand maybe 800c? It could not be a liquid tight valve probably. I wonder if such a thing exists?

Do you think in your case the vents directed heat at combustable parts near the battery box? Or could the heat radiating from the box have ignited nearby combustibles?

Thanks so much for taking the time to respond. I am in awe of what you produced. Such a heartbreaker to see it damaged.

Save

Ohno

Discussion Starter · #34 · 3 mo ago

remy_martian said:
Safety requirement, not merely best practice. It used to be "best practice", which is nowadays unacceptable for occupant and first responder safety in roadgoing and competition vehicles. There are too many conversions being done these days to where someone will wreck and electrocute a responder....then we are done as hobbyists as we either get regulated to death or outright banned. There's no nicespeak on this - if there's more than 60V on ANY wire, it needs to be able to reliably get shut off in the closed box that put the voltage out.

"Wetted area" increase...yuck. But you're the boating genius here.

Meanwhile, yes, you could use a flight controller's pitch, yaw, and roll sensor to move the surfaces. Instead of your wands, use the four motors and props from a quadcopter, each placed in each of four corners of your boat that are actually flying in air (a couple of feet above water so they don't dunk) to servo the position in the air, then run those down with a control rod/wire to a trim tab on the foil control surfaces. The trim tab acts as a force multiplier and is usually how autopilot servos move control surfaces on aircraft. This should only pay attention to the gyro on the flight controller and will not care about wave heights and should not require a single line of code to be written.

That would be a lot of fun to play with - on the next boat maybe. This one needs to be a more practical daily transporter from our non ferry serviced island. I'm even hoping it can push a small steel hay barge around a couple of times a year, but that might be asking too much.

Save

Reactions: 1

Ohno

Discussion Starter · #35 · 3 mo ago

oudevolvo said:
Thanks!
In the first place my fire mitigating action will be not to use Tesla batteries anymore.
Nowadays there are safer batteries that will give me the right volume, weight and kWh at 96s.
There are two things that could have been better on the box I made:
1) Smarter vent valve positions (so the exits would be where even less damage would occur).
2) A hose connection (not sure how to seal/close that when not in use) onto the box that the firemen could have used to put water inside the box to even better cool at the exact right place

On second thought it might be that introducing water into a sealed box where temperatures may be 800c might cause the water to flash to steam and explode. I will leave that for others to test.

Save

Ohno

Discussion Starter · #40 · 3 mo ago

oudevolvo said:
Thanks for your kind words. I think the flames coming from the vents ignited surrounding stuff. Higher up there was mostly radiation and there the damage is not that bad although it has burned there and also underneath the dash.
See also http://instagr.am/p/CiVWyI8tkMz/ Regarding the hose connection, in my view it can add value in two stages:
1) At the very beginning of a fire where you might be able to cool it down to stop thermal runaway
2) At the end of a fire to ensure it does not light up again
Then just a metal (quick connect) cap can do the trick.
At the hottest point during the fire I'd not suggest to use/open it.
In my box, three Tesla modules burned until there was nothing to burn anymore.
When the heat was more or less gone, the firefighters tried to open it, but I suggested to cut off the MSD instead so that was the way to extra cool down instead of the hose connection. See photo 2 here http://instagr.am/p/CWxV47Yojzt/
But in your case (if you still want Tesla modules on a boat) I'd say try to ensure each module is contained so in worst case it can burn without igniting others.
Think the issue in a Tesla (and others) and the main reason for submersing cars is that you never know if the surrounding batteries got just too hot or not.
In my case I was quite sure the front would not re-ignite since there was simply nothing left to burn.

So it seems that your battery box worked well to contain the fire. Possibly if vents are directed away from combustibles and with the addition, perhaps, of some shielding and/or insulation a battery fire might not necessarily be a disaster.
I'm sorry that I can't simply build a battery box that could be jettisoned into the water. It seems like the most elegant option since we are surrounded by water. Maybe on the next boat!

Save

Ohno

Discussion Starter · #42 · 3 mo ago

oudevolvo said:
True, but in a car it is quite difficult to keep the flames away from anything that can catch fire at these temperatures. It's just too tight.
And I have a lot of secondary damage. Powder from an powder extinguisher someone who wanted to help used got all the way in the cabin (so I need to replace all wiring).
So ensure you have a proper fire extinguisher to put surroundings out.
But since you are on a boat, can't you just cool and suppress a thermal runaway?

I like your use of the word "just"! But if someone is there at the time to detect a problem then, with good pump, maybe a battery fire could be controlled. Your hose access to the battery would be a good thing then for sure.

You are so right about the fire extinguisher. I had not thought of that as important for a battery fire because, what's the point? But it's all the other flammable stuff that needs to be protected.

Save

Ohno

Discussion Starter · #46 · 3 mo ago

floydr said:
Shouldn't a thermistor or a bunch of them warn the operator in time to flood the battery compartment with continuously flowing water thereby cooling The offending cells/ modules before you would have a problem with high heat >100ºC of course you would lose the modules in the battery compartment affected by high heat. This could even be automated if any thermistor goes above such and such temperature the battery box is disconnected from motor and water starts to flow in and flow out though ports which would be opened when pump is activated.
later floyd

I like this. Someone, other than me, needs to sacrifice some batteries to test it.

Save

Ohno

Discussion Starter · #47 · 3 mo ago

remy_martian said:
You can always put the battery box in its own dinghy and tow it.

Use fiber rope for the tow and it self disconnects in a fire 😂

Perfect.

Save

Ohno

Discussion Starter · #48 · 3 mo ago (Edited)

remy_martian said:
In the case of a boat, I'd cut a square hole in the bottom of the boat then seal a stainless steel battery box with a plastic bottom to that hole - in the event of a fire, the flaming batteries get delivered to Davey Jones locker.

You know I think your idea might work. It means moving the batteries back 18" and out 12" and so messes with my weight distribution but I can compensate for that. By adding one bulkhead I can create a sealed compartment in each hull. The batteries could sit in a open top box in that compartment and the bottom of that box would be the bottom of the hull that has been cut out. The details of making it watertight and structurally sound are not at all daunting to me. Neither is coming up with some kind of mechanism that lets the box drop.

What I don't know is how the various electrical/electronic components will handle this. Especially the BMS.

I picture it working this way:

Over temperature alarm sounds.
Someone, or some kind of automation, 'pulls the pin' and the offending box drops, say to 18" below the hull, and floods with water.
Now comes the tricky stuff: This boat's batteries are in two packs, one pack in each hull. They are in a 72v 3s2p configuration. Ideally I would be able to disconnect (Anderson connectors?) the submerged battery and plug the battery cables back in to re-energize the controller/motor so we can limp home using the dry pack.
Is the BMS going to hate this and refuse to allow the remaining dry battery pack to function?
I'm thinking only the batteries will take a swim, Or maybe a contactor and/or fuses will get wet as well. The BMS wires would get wet too. Everything else would be dry.

Any thoughts?

See less See more

Save

1 - 20 of 81 Posts

Top Contributors this Month