Two electric unicycles + 140kWh of dead cells + Solar panels (Bayliner hybrid)

[HOTOREL]

**********************************************
December 2023 Update is HERE
**********************************************

Current project state:

Yacht: Bayliner Bodega 1982, 45ft LOA

Battery:
- 48V 4kWh 13S2P Old Leaf Battery modules
- 12V 400A Lead Acid (In plan to get rid off from it)
- 100.8V 9kWh 24S18P 21700 li-Ion (in the process of creation) about 9kW

Solar: 3 panels for 550W (1650W total)
Generator: 3kW + 7.5kW (in the process of repair)
Port motor: 2.5kW nominal 100.8V DC
Starboard engine: 350 Chevy V8


***********************************************
Original post below
***********************************************

Hello everyone, I want to ask for your opinion about my project. I'm thinking to invest in an old boat yacht and rebuild it for a fully electric propulsion system. I don't think that someone tried to do something like that before and I have a lot of doubts if this is possible. I hope I can find answers to my questions here.

The yacht + engines
So here is the thing, a potential boat which I want to buy is the "Bluewater coastal cruiser 51". They are super cheap compared to all other yachts because of the hull design and because most of them have really old gasoline engines. They usually use "Chevy 454" or "Mercruiser FWC 502" engines for more fresh years. The first thing I want to change is to swap them for Tesla Model 3 drive units. Those drive units have less continuous power but I am not planning to fly 20 knots, stable 5-10 knots will be more than enough for me. The only information which I was able to find on the Internet about those Drive units is that continuous power is around 35 kW (47 Hp). So we will have around 70kW for both of them.

Battery construction
Now let's talk about the battery, because it is a really big yacht I am planning to use 28 EG4-LifePower4 batteries. Which will provide around 143kWh of juice. I know that they are server batteries. They will be installed inside of the boat so if they get wet it means there is already nothing to save) And here is the first question. I want to know what you think about the way how I want to connect them together. Because I have never seen anything like that before and I think I am missing some important point somewhere.
I want to connect them 7s4p so this way I will have a nominal 358.4V 400A battery which should be ideal for the two Tesla Drive units. I hope... Because the original Tesla battery has a pick output of 1000A for a couple of seconds but because I will not accelerate from 0-100 for 4 seconds I hope I will not have that current. (What do you think about it?)

Why 7s4p?
Here is the unusual part which I was talking about, sorry for a lot of words. Because I am planning to install this system on the yacht I don't need fast charging system. Because I can slow charge my battery at any marina or using my generator (about it later), or solar panels (also a bit later). I decided that I will use an individual 2000W Victron system for each 4p element of my battery. I know it is kinda a lot of Victrons BUT they are the only one (or at least I didn't find any others) who support up to 7 Victron devices connected together and communicating with each other. So in this case I will have only one battery on the boat which could be used for the main engines as for the home appliances because for each Victron unit, it would be an individual 4p 51.2V battery. I am still working on the scheme but I can attach later what I have so far.
So here is the question: how dangerous is it and why nobody did that before? I am pretty sure both of these questions are connected somehow)

Solar panels and generator
The last part here is the sources of power that I am going to use to recharge that thing. The minimum I want to install is 8 big residential panels for 400W each. I am glad that this yacht has ideal construction for that. All this power will be managed by Victron solar controller as well as the 11kWh generator and the 30A shore power.

Battery location
If you have read till here and still think this is real it is really nice because here is the next thing to think of. There is a lot of space on the floor of this yacht (I already did all 3D scans of the real one) but in this case, I would need to use cables with different lengths for different 4p elements of my system to the nearest Victron unit. Which I guess is not good. As another option, I can try to install all of them in a kind of server ruck but their total weight will be around 1260 kg which can affect the center of gravity which is also not the best thing. What do you think about that?

Tesla battery?
I think I need to explain why I don't want to go the "classic" way and just use a Tesla battery. Everything is simple, in this case, I will need to transform 48V to 400V using invertors which will lose some percentage of my energy, and generate additional heat, and I still will need one more battery with a more powerful Victron system for household use...

Range
Let's try to imagine that this thing is real and I didn't die from the short circuit while I am building this thing. As I understand the Tesla Drive units will consume around 70kWh at the full load. So in this case, if I will use them at 40-50% I should have around 4-5h of sailing time without solar or generator help? Or am I missing something?
I really want to say thank you to everyone who wants to participate in the discussion for this project. I was thinking about that for the last year or so and have a lot of details that I will be happy to share with you. For now, I just want to see if there will be anyone who I can discuss my ideas with because chatGPT already don't know the answers to my questions)

 

[Rocketman]

I have been toying with doing the same thing but with a smaller boat.

Here are my thoughts…

I don’t think your charging scheme will work (practically). A 48v charger (Multiplus) can only charge 48v batteries not a 358v battery. Now with Anderson plug connectors you could break down that big battery back to a 48v to charge. Then you also need to VERIFY that you can put 7 of the batteries in series. I think that will blow the Fets in the BMS. Also, I think you will need much more solar.

Here are what my ideas have created. (My dream is about a 40’ catamaran).

Motor and controllers - Esco electric Marine motors, either a 96v or 144v one.

Battery Get many of the Seplos battery boxes 16s 280ah cells. Put 2 or 3 in series - with disconnects between to comply with ABYC standards. (No bms). There will probably be two batteries in parallel.

For the BMS - a contactor based one - Batrium or Orion (I use Batrium on my RV - but for this I would reevaluate). Both of these can handle very many cells in series - so they can handle the high voltages.

For Solar - Morningstar has one mppt that I believe can handle 96v battery. Some solar goes here.

For generator/shore recharging. Electric Car Parts has several battery chargers that can handle 96 or 144v DC batteries) from 120v AC power. They also have one that can handle 312v and 540v - wow!

That is the propulsion battery (times 2 for the boat).

You also need a large house battery 48v.
Victron Quattro 10k, mppt’s (more solar goes here) and then four or six of the 16s 280ah battery boxes in parallel.

The house battery can recharge the propulsion batteries by converting to 120v then using the big battery charger for each side.

Generator/Shore that can charge the house bank, which can also charge/pass through to the propulsion batteries through the 120v.

A big cost is each 48v battery case - by the time it is done it will be around $3.000. Twelve or more of the batteries for the boat starts to get to some serious dough! Plus motors!

Good Luck

 

[niktak11]

The BMS in the EG4 batteries almost certainly can't be connected 7 in series

 

[HOTOREL]

The BMS in the EG4 batteries almost certainly can't be connected 7 in series

I don’t want to connect BMS in series. Only 4 of them are in parallel. So each 4p will be pretty independent and will be monitored by Vectron

 

[niktak11]

I don’t want to connect BMS in series. Only 4 of them are in parallel. So each 4p will be pretty independent and will be monitored by Vectron

Maybe I misunderstood this statement

"I want to connect them 7s4p"

 

[HOTOREL]

Maybe I misunderstood this statement

"I want to connect them 7s4p"

Yeah, I think I will attach the scheme tomorrow ? Sorry that I wasn't able to describe my ideas properly

 

[yabert]

The only information which I was able to find on the Internet about those Drive units is that continuous power is around 35 kW (47 Hp). So we will have around 70kW for both of them.

This is the continuous power for Tesla induction motor (old model S). The Model 3 motor have a way better cooling capacity and an higher continuous output power.
Enough to cruise at highway speed with a huge Promaster van: https://ingenext.ca/pages/ev-kit-conversion
Also really good result for a racer near you who swap model S for a model 3: https://www.youtube.com/channel/UCMryo0vWzx1Q8vrLAP7APLg

 

[yabert]

Tesla battery?
I think I need to explain why I don't want to go the "classic" way and just use a Tesla battery. Everything is simple, in this case, I will need to transform 48V to 400V using invertors which will lose some percentage of my energy, and generate additional heat, and I still will need one more battery with a more powerful Victron system for household use...

I don't understand. 2 Tesla 3 battery will give you 150 kWh and you simply need an high voltage inverter (400Vdc to 120/240AC) if you need to have AC inboard.
To add, with complete tesla 3 battery the charger and the 12v DC/DC are include.
Depending of your load, a simple 12V inverter could power the boat AC loads.

 

[Shimmy]

I'm not familiar with the state-of-the-art, but there were a lot of hybrid diesel-electric sailboat conversions being done ~10 years ago. To pull it off though you are going to need a native high voltage battery; the 48V server rack units have a maximum system voltage of ~60VDC. Also... you realize that would be 2,800 pounds of battery; are you going to be able to save the weight elsewhere and maintain the boat's center of gravity?

I'd also guess this is an application where NMC batteries would be better than LFP due to weight.

The other thing I know nothing about is how grounding would work with the high voltage DC system... that would scare me.

 

[elusivesd]

I'm not familiar with the state-of-the-art, but there were a lot of hybrid diesel-electric sailboat conversions being done ~10 years ago. To pull it off though you are going to need a native high voltage battery; the 48V server rack units have a maximum system voltage of ~60VDC. Also... you realize that would be 2,800 pounds of battery; are you going to be able to save the weight elsewhere and maintain the boat's center of gravity?

I'd also guess this is an application where NMC batteries would be better than LFP due to weight.

Many larger boats/sailboats intentionally *add* weight to the keel in the form of Iron or lead, specifically to improve stability. It would theoretically be possible to have a "lithium keel", which would consume no additional space onboard... Of course, this depends on the type of boat - ideal for a displacement boat (monohull), but not so great for a cat.

The other thing I know nothing about is how grounding would work with the high voltage DC system... that would scare me.

Grounding and corrosion aside, 300V+, on saltwater, on a moving boat? This is very different from electronics getting wet.

Look into a 48V drive, off the shelf conversion units do exist. Some will convert your 48V system to whatever your motor requires via inverter. Did I mention 48V? It is a popular, and relatively safe voltage for this purpose.

 

[yabert]

the original Tesla battery has a pick output of 1000A for a couple of seconds but because I will not accelerate from 0-100 for 4 seconds I hope I will not have that current. (What do you think about it?)

You will not have that current if it's to move at 5-10 knots.
Tesla peak output can be over 1000A because there is something to put power to (torque x RPM): the tires.
And peak power happen at relatively high speed.
Trying to put power at the prop in the water is different and if you can have the RPM in the formula RPM x Torque=Power, it would be tough to pass high torque in prop. Especially at low speed.

 

[wholybee]

Many people have converted old yachts to electric. It is at this point a well understood thing, with lots of products purpose built for it.
I am not at all familiar with Tesla motors. 35kW seems like a very low figure for one of them. However, 70kW is appropriate for a typical displacement yacht of 50 feet. And your expected speeds seem reasonable.

I would not even have Tesla power units on my list of possible engines. There is a long list of motors built for what you want to do.
I would not exceed 96V. Probably 48V. 48V batteries are easy to build and charge. I would have a selection of motor controllers, etc. You absolutely want LiFePO4 batteries. That will rule out Tesla packs.

I would start with a power budget. Figure how many HP (kW) you need to move the boat the speeds you want. Then how many kWh of batteries you will need for the run time you want, and what charging you will need to support that. Then start looking at products that meet those needs. But, Tesla products are not made for boats and thus would not likely work best.

 

[Hedges]

BMS have a limit to how many can go in series. Some 12V batteries can be connected for 48V, some can't.
For the Tesla drive, you need several hundred volts, so it would be several 48V server rack batteries in series, maybe 7s as you said.
No chance the server rack BMS can handle the -300V they will see if one disconnects.

I think a bypass diode for each would protect it. Revere biased with 48V or so across it in normal operation, forward biased with a diode drop of maybe 2V when BMS shuts down. Do you draw 100A? Then you need to dissipate 200W, and/or detect the shutdown and stop drawing current.

But even though I think BMS between the battery terminals can be protected this way, I'm not certain that any other electronics, communications, possible EMI filters to chassis won't fail if the battery circuits is biased to several hundred volts.

With just the cells and not the 48V BMS, you could connect one of the modular BMS such as REC, which can support up to 128s.

You probably want to put things in an airtight box to keep salt atmosphere out. Includes some bags of desiccant, and cooling if needed. Could be heatsink with fins inside & out, fan blowing over it. Or water cooled, hopefully tubes outside and metal carrying heat from fins inside.

 

[yabert]

You absolutely want LiFePO4 batteries. That will rule out Tesla packs.

Except if you put your hand on a Tesla 3 SR with an LFP pack ?
There is now thousand of those on the road and possibly a lot at scrap yard.

 

[willo]

Yeah, you're gonna want a 50A shore power connector on that thing.

For the voltages/loading I'd go with REC BMS and custom packs. 300ah cells will serve you better than a crap ton of 100ah cells.
Rec supports master/slave BMS deployments, You can build multiple 48v high capacity packs that'll work in unison this way.

If you must use those rack mount boxes, then I'll tell you to open every single one and hit all the connections inside with corrosion prevention. (T9, etc)

You should really get a quote from thunderstruck on this. In the marine environment, you'll find that staying at 48v is much, much safer.
A guy here just had his boat totaled after hitting something on the ICW - it put a strut through the hull.
If that boat had 48v in there, it'd still be safe. If it had 300+, it would have become very dangerous.

I am more than happy to talk about this more, I've been diving into the marine builds in the last year with my sailboat (which still has a diesel but I'm ok with that)

 

[HOTOREL]

Hello everyone and sorry for missing for a long time. I had a lot of time to think about everything you recommended to me and I already started to go towards my goals. First of all, I bought Bayliner Bodega 45ft LOA and this is the Yacht I’m going to proceed with. So there is a lot to cover and I will try to explain everything as much as I can but please let me know if I made any mistake anywhere. I would love to discuss everything with you.

First of all the Yacht. I tested different speeds on it and found out that around 4-5 knots is the most comfortable speed for travel around. Still need to install the autopilot but it is another topic) So I measured the RPM of the driveshaft at this speed and it was about 700-800 RPM which is not a lot. Because I lost my source of income not so long ago I had a significant cut on the conversion budget so I had to use al my engineering knowledge to make sure it would be even possible. So keep this in mind some of the decisions that I will describe below were only chosen because of the reduced cost for the whole project. But still, it should be powerful enough to move around. Because I had access to a lot of different 100V electric motors at my previous job I decided to give a try to 4kW direct drive electric motor from Begode T4 electric unicycle. There were also more powerful ones to choose from but they required more complicated transfer plates and some of them couldn’t fit directly into the shaft because of the small space near the engine there. You can check the 3D scan of my engine compartment here:

https://sketchfab.com/3d-models/model-1446240d4b0848679452535ea67725dc

Because I live full-time and travel on this yacht I can’t disassemble everything so I have to do all the conversion in small stages which is completely fine. For now, I want to start all trials on the Port side (because the port side engine is already dead) so if someone wants to repeat anything like this later on here are the technical drawings for the T4 motor to connect it bolt on to the drive shaft. So Yeah I already have the test motor and ND96360 Fardriver controller. I just ordered it from Aliexpress. In a couple of days, I will meet with my friend and will try to connect it all together on the stand for the first time. For tests, I will use just a 100V battery from another EUC (Electric unicycle). Will update you here about the tests. Plus next month I should receive all the parts to bolt it on the drive shaft and we can start the actual tests. The cool thing is that there is enough space to put the motor without removing the old engine so it will save a lot of money for me during the tests plus now it works as a ballast to keep the center of gravity in its place.

What is my current setup?

Now I have two wrong-connected 13S1P batteries made from Nissan Leaf (going to fix it soon) You shouldn’t connect the Nissan Leaf battery in the 13S configuration. Only 12S or 14S because otherwise you will end up connecting your load to the balance terminals which is not good for the battery. I have one EG4 3000W mppt solar charger inverter and three 550W solar panels. Plus 3kW portable generator because the original is dead and also behaves as a ballast now. So current battery capacity I have is about 4-5kW and it is only 48V so it was stage one to make sure I will have at least some electricity. There will be a couple of more stages in between but the final setup I want to come to at the end of the summer is shown below:

I’m still planning to use Nissan Leaf battery modules, here is why:

Price!

I mean you can buy 420 Nissan Leaf modules for only 5000$ which is about 130kWh of capacity there is nothing like this on the market for the same price. Yeah, they are old and heavy. So I would prefer to change the battery for the next old generation in 3-4 years for a fraction of the price than spending 10 times more now. Plus I have no idea what I’m going to do with this boat in a couple of years. Maybe I will buy a bigger one. Still thinking about that. So the second concern is that they are heavy. 420 modules should be around 500kg which is almost the same as two current gasoline engines which I have to throw away so it is even ideal for me.

The setup:

The hardest part was to engineer the elegant solution to convert these 48V batteries to 100V without burning anything and make it relatively simple to do. So here is what I’m planning to do (pretty soon on the smaller 20kWh scale) I’m planning to connect leaf modules in 12S4P configuration with a bit of a twist that there will be an additional terminal with bypass of BMS. I’m planning to use this terminal only to discharge (for a series connection of two huge battery banks connected in series) and it will be additionally protected with a relay Which should disconnect the load as soon as BMS shuts down the power on the main terminals. In this case, we will not over-discharge the battery and will not damage them. I have to do it this way because as you mentioned earlier I can’t connect BMS in series (or they will cost too much) in my scenario I can use really cheap 40A BMS because combining 12S2P of them will give me 480A 96V which is more than enough for anything.

Charging:

To charge it all I’m planning to use my MPPT inverters but they should be connected to one battery bank and have a sync cable to make sure they will now burn at the same second I will turn them on. So, because we already have two battery banks we will use two absolutely independent 120V systems on the board. It allows me to utilize two of my solar arrays for each system, and use one shore cable to charge both of them and one generator. The only one downside I won’t be able to combine all power in one combiner box and will have to make another one and think ahead and split loads properly. But again in the future, this system is easily scalable because I could change only one inverter which requires more power because it is completely independent systems.

100.8V:

As you see on the scheme there is a relay that connects two battery banks, this relay will be controlled by an inverter output signal. So the logic is simple if both invertors are in default state it means that the battery is in normal condition and we can use it for the main engine. As long as any of the inverters will shut down the load it will automatically disconnect battery banks from each other and will protect the battery from over-discharge. This is the first thing that should happen during discharging, the relay on each battery is like second-stage protection from over-discharge. Let’s say when one of the batteries will degrade more than others it will shut off the bad battery module but all other batteries still will perform.

So in this scenario, I will have only 12 BMS connected in parallel at any time which is safe and doesn't break any requirements.

Why don’t use separate batteries for the house loads and separate for the electric motor?

Because it is stupid. First of all, you will have two huge battery banks which you have to charge. So in my case, I can use this 130kWh for the whole week without recharging it at all or cruising the whole day without charging, and if I’m going to charge I’m going to charge only one battery, Again, it is easy to maintain and monitor the system when you have one big source of energy and you just monitor it. You are not going to lose a lot of energy while you converting your energy. Every step-up and step-down inverter has its own idle consumption and losses. So the less you have them the more reliable your system is.

So yeah, this is the current state of the project. I know that I didn’t cover a lot because I already think that this is a pretty big thing to read already. But I will try to not disappear again for a long time and will keep you updated on the project and will discuss it with you.

 

[AntronX]

I mean you can buy 420 Nissan Leaf modules for only 5000$ which is about 130kWh of capacity

Are you sure you are buying modules with good SOH ?

 

[HOTOREL]

Are you sure you are buying modules with good SOH ?

I’m going to buy with about 70% SOH which is more than enough for the next several years. Again even if they are going to die in a couple of years I can buy a new set for a fraction of the cost of the new one.

 

[AntronX]

I’m going to buy with about 70% SOH which is more than enough for the next several years. Again even if they are going to die in a couple of years I can buy a new set for a fraction of the cost of the new one.

Li battery capacity degradation "falls off a cliff" when nearing 60% SOH. You would be buying a battery already degraded past it's rated cycle life which is 80% SOH. Don't waste your money and go with new LFP cells.

 

[HOTOREL]

Li battery capacity degradation "falls off a cliff" when nearing 60% SOH. You would be buying a battery already degraded past it's rated cycle life which is 80% SOH. Don't waste your money and go with new LFP cells.

The cheapest deal for LFP that I found was a 10240Wh server rack from SWA for 2249 CAD per module. Let's compare it to the Nissan Leaf price:
I can buy right now 432 Nissan Leaf modules for $5784 (7842 CAD). I need 432 for my configuration and it will provide me with a 138kWh battery. So if I want to build the same size of the battery with LFP cells it will cost me 30388 CAD which is more than three times more expensive. I didn't put here prices for the mount rack system or wires (because I will need them in any case), or delivery (100% from China will cost more than from the US) but what we can add is a BMS for the Leaf battery. One BMS costs about 25CAD which adds 600CAD to my total of 8442CAD


Now imagine that they are going to last for 2-3 years (the bigger your battery then the slower you are going to discharge it so the longer it is going to stay alive) I will just change them for something more new and efficient or just update my pack for a less dead one for a fraction of the cost.

Again I agree that it is right to use LFP cells but it is out of budget for a reason