[xevion]

Today I was lucky enough to pick up this 1976 Sebring-Vanguard Citicar from someone who bought it in the 80s and stored it ever since.

With ~3k miles it's in good shape for it's age, the body doesn't have too much cracking which I'm told is common for these. My goal for this project is to restore the exterior/interior and modernize the electronics. I'm coming to this forum as a newbie to EV work, and am looking for some advice on getting going with this build. I have experience wrenching on ICE cars, and have a background in electrical engineering & CS, but I'm not sure where to start/how far to go with this project. Today I cleaned out a few things and took stock of what I had gotten myself into.

The batteries hold enough charge to spin the rear wheels with no load, but not much else (~5V each when I pulled them).

They also weighed an absolute ton. My understanding is that batteries like these are recyclable, and some auto parts stores will even give you an in-store credit for bringing them in. I know the electrical system in the car works (everything light up, the main battery charger works, etc.) but the 50 year old wiring worries me, so I imagine I'll be changing a lot of it out. In terms of modernizing the electronics I'm thinking there are three key components to upgrade here: batteries, motor controller, and motor.
The batteries are a necessity because the old ones are toast. My dream would be ~50 miles of range out of a pack with modern battery tech, but I'm not sure what I should be looking for. I also know new battery chemistry means new battery charger (the old one seems comically simple).
Next up is the motor controller. The current "motor controller" is a combination of big relays and a big nichrome resistor that basically amounts to a three speed switch. I have a feeling this will work as a temporary solution but seems like a natural next step upgrade once I sort the batteries out.
Lastly the motor. If the current one is in good working order, I don't really see myself messing with it. An induction motor would be nice, but I don't think it justifies the hassle of swapping the motor. That being said I don't have a sense for the pros/cons of going induction over DC.

That's all for now. I'll be working on this tomorrow and I'll keep an eye on this thread. Any help/advice is greatly appreciated!

 

[OR-Carl]

Haha, this is a really fun looking project, I am excited to see it come back to life.

My understanding is that batteries like these are recyclable, and some auto parts stores will even give you an in-store credit for bringing them in.

I just scrapped a few batteries, and they are worth 0.15$/lb at my local scrap yard at the moment. Auto parts places have a core charge of about 20 bucks, which is a better deal if they will take a bunch of them for store credit.

Your assessment on what needs to be done is spot on. If all the "electronics" in the drive system are still functional, then you can probably push their replacement down the road. As long as you get back to your original battery voltage, they should work fine. Or at least as well as such a crude system ever worked . You likely have a 96v system, with 8x 12v batteries. You will be a little limited in what sort of battery you can cram into that battery box, but lithiums will give you a big boost even if they have a nominal rating that is simply equal to the lead acids they replace.

There are a couple of routes you could go for batteries. Used EV batteries are the current DIY solution of choice, as they are cheap and already really well engineered for the application. This build might be tough, as the OEM batteries are usually in fairly large modules, and obviously you will only be using a partial pack. Leaf cells might be a good fit, you will have to do some measuring though to see what would fit in that space.

Another option, that I would not suggest for a full-sized car might actually work out for your case; which is to get prismatic LFP cells. This used to be the only way to get large lithium batteries, but they are not really ideal for mobile applications. There is also some concern that when you buy small quantities online, you might be getting QC rejects.

Lastly, if you have lots of free time, this might be a good candidate for building a custom 18650 pack. This route would likely let you cram the most battery into the space, but it is a LOT of work.

As for the motor: if it is still working, then there is really no reason to mess with it at this point. The gains from switching to AC would be very small for the amount of work and expense it would add.

And that leads to the final point: budget. I suggest you do some research about what batteries might cost, as that will be the biggest expense for your project. Dont forget about a BMS, and a programmable lithium battery charger. Anyway, do some reading, and post more questions as they come up.

 

[brian_]

You likely have a 96v system, with 8x 12v batteries.

I assume that the batteries shown under the seat and in the driveway are the whole set, and they are eight "6V" (three-cell) lead-acid... very typical for a golf cart. So, nominally 48 V rather than 96 V.

Think "golf cart" rather than "early EV".

You will be a little limited in what sort of battery you can cram into that battery box, but lithiums will give you a big boost even if they have a nominal rating that is simply equal to the lead acids they replace.

There are a couple of routes you could go for batteries. Used EV batteries are the current DIY solution of choice, as they are cheap and already really well engineered for the application. This build might be tough, as the OEM batteries are usually in fairly large modules, and obviously you will only be using a partial pack. Leaf cells might be a good fit, you will have to do some measuring though to see what would fit in that space.

Another option, that I would not suggest for a full-sized car might actually work out for your case; which is to get prismatic LFP cells. This used to be the only way to get large lithium batteries, but they are not really ideal for mobile applications. There is also some concern that when you buy small quantities online, you might be getting QC rejects.

I agree.

The tougher part about using production EV modules is that they are configured for much higher voltage, so arranging for only 48 V or so means either a very small pack or having to connect modules in parallel.

Since lithium-ion cells are lower in density (mass per unit volume) than lead-acid, any replacement that fits in the same space will be lighter. It will also have more energy capacity, if it fills the space even close to as effectively.

The original batteries are likely the GC2 (yes, "golf car") sized:

260 mm long x 180 mm wide x 274 mm tall (10.24 x 7.09 x 10.79 inches)​

30 kg​

The same volume (about 12 litres per battery times eight batteries for about 100 litres) of lithium-ion cells would have an energy capacity of about 35 kWh, but realistically they won't likely pack that nicely.

If the space is four GC2 battery lengths long, two GC2 battery widths wide, and one GC2 battery height tall, that's
1040 m (41") long x 360 mm (14") wide x 274 mm (11") tall

A stack of 24 original-style Nissan Leaf modules would fit in there, configured 6 sets of 4 modules, with each 4 modules wired in parallel. That's half of a complete Leaf pack (the part under the rear seat of the Leaf), which is 12 kWh to 20 kWh depending on the generation of Leaf modules used.

Up to six of the VDA 355-sized modules would sit side by side with lots of space, and just four of them in the 3S4P configuration such as the LG "X590" modules available from Zero EV would add up to the right voltage and hold about 10 kWh (and weigh only 51 kilograms - or 112 pounds - compared to the original 240 kg or 530 lbs).

There are also some prismatics in other chemistries (not just LFP), and prismatics do offer the best hope for optimal packing. At least 15 kWh of NMC cells would fit, in just one layer. If LFP cells are used, about 16 of them in series (multiplied by a suitable number in parallel) would run at about the right voltage; in other lithium-ion chemistries about 12 of them in series (again multiplied by a suitable number in parallel) would be suitable.

For example, a 16S4P (64 cells, paralleled in sets of four, 16 sets in series) configuration of a LiFePO4 High Power Cell (3.2V/60Ah) from GWL (which happened to come up in another forum discussion) would provide 12.3 kWh in 109 kg, and fit easily.

 

[OR-Carl]

Nice catch on the batteries, I looked right past the fact that there were only 3 cell caps on them!

 

[xevion]

There are a couple of routes you could go for batteries. Used EV batteries are the current DIY solution of choice, as they are cheap and already really well engineered for the application. This build might be tough, as the OEM batteries are usually in fairly large modules, and obviously you will only be using a partial pack. Leaf cells might be a good fit, you will have to do some measuring though to see what would fit in that space.

Another option, that I would not suggest for a full-sized car might actually work out for your case; which is to get prismatic LFP cells. This used to be the only way to get large lithium batteries, but they are not really ideal for mobile applications. There is also some concern that when you buy small quantities online, you might be getting QC rejects.

Lastly, if you have lots of free time, this might be a good candidate for building a custom 18650 pack. This route would likely let you cram the most battery into the space, but it is a LOT of work.

Thanks for this info, getting batteries ordered is a top priority so I'll start to look into all of these options. I'm not exactly sure what kind of capacity (in kWh) I should be shooting for, but I think this will likely be dictated by the physical space I have and how much I want to spend.

As for the motor: if it is still working, then there is really no reason to mess with it at this point. The gains from switching to AC would be very small for the amount of work and expense it would add.

This pretty much settles it for me, if the motor is good it's staying in.

Since lithium-ion cells are lower in density (mass per unit volume) than lead-acid, any replacement that fits in the same space will be lighter. It will also have more energy capacity, if it fills the space even close to as effectively.

The original batteries are likely the GC2 (yes, "golf car") sized:
260 mm long x 180 mm wide x 274 mm tall (10.24 x 7.09 x 10.79 inches)30 kgThe same volume (about 12 litres per battery times eight batteries for about 100 litres) of lithium-ion cells would have an energy capacity of about 35 kWh, but realistically they won't likely pack that nicely.

If the space is four GC2 battery lengths long, two GC2 battery widths wide, and one GC2 battery height tall, that's
1040 m (41") long x 360 mm (14") wide x 274 mm (11") tall

A stack of 24 original-style Nissan Leaf modules would fit in there, configured 6 sets of 4 modules, with each 4 modules wired in parallel. That's half of a complete Leaf pack (the part under the rear seat of the Leaf), which is 12 kWh to 20 kWh depending on the generation of Leaf modules used.

Up to six of the VDA 355-sized modules would sit side by side with lots of space, and just four of them in the 3S4P configuration such as the LG "X590" modules available from Zero EV would add up to the right voltage and hold about 10 kWh (and weigh only 51 kilograms - or 112 pounds - compared to the original 240 kg or 530 lbs).

There are also some prismatics in other chemistries (not just LFP), and prismatics do offer the best hope for optimal packing. At least 15 kWh of NMC cells would fit, in just one layer. If LFP cells are used, about 16 of them in series (multiplied by a suitable number in parallel) would run at about the right voltage; in other lithium-ion chemistries about 12 of them in series (again multiplied by a suitable number in parallel) would be suitable.

For example, a 16S4P (64 cells, paralleled in sets of four, 16 sets in series) configuration of a LiFePO4 High Power Cell (3.2V/60Ah) from GWL (which happened to come up in another forum discussion) would provide 12.3 kWh in 109 kg, and fit easily.

I really appreciate the detailed response. I'm looking forward to seeing how the reduced weight effects things, those golf cart batteries are so heavy. How much does delivery factor in to the cost of batteries? I can't imagine it's cheap to ship them. Does it make sense to try to look for local ev sellers? I'm located in New England for reference.

Nice catch on the batteries, I looked right past the fact that there were only 3 cell caps on them!

I made the same mistake.

 

[xevion]

Spent a good 8 hours getting things cleaned and disassembled today. Here are some of the highlights:

Pulling the original charger out. It really is just an off the shelf golf cart battery charger shoved under the dash. I may look to sell this since it works.

The batteries that were in there were really nasty, some baking soda confirmed they were covered in battery acid. Yuck.

Could this duct tape be.....factory? It's goes between the frame the body, and I can't see how someone would have gotten it in there unless the two were separate.

The vent tubes under the dash just fell apart in my hands. They should be easy to replace.

Found some vacant wasp homes under the dash. These were fun to take out, especially the one on the left which was built on top of a circuit board.

Here's a close up look at the mechanism that runs the 3 position pedal. The first click runs the 48 V pack as 2 24V packs in parallel, and then runs them through a resistor. The second position takes the resistor out of the equation. The third gives you the full 48V. I'm thinking it's going to be easier to just get a modern motor controller instead of shoehorning a modern pack to work with this system. That being said the car is wired to work like this right now, so maybe I should give it a try. Thoughts?

The previous owner did some work on overhauling the old hydraulic brake system, which is awesome. Less awesome is this homemade parking brake setup zip screwed to the bottom of the car, which doesn't seem to actually function.

The rear drums actually don't look that bad, but I can't seem to figure out why the e brake cable won't engage the brake shoes. More investigation is required. Also this stack of washers looks a little suspicious.

This is the air intake to the motor. Luckily there is a grate preventing this stuff from getting sucked in, but it being here doesn't inspire lots of confidence.

 

[xevion]

Here's the blower that keeps the motor cool. That second picture is what's left of the impeller. Good thing blower motors are easy to come by. Might also try to 3D print another impeller since the motor seems to work just fine.

 

[OR-Carl]

Does it make sense to try to look for local ev sellers?

Yeah, I would definitely look. You might start prowling the local craigslist for EV Batteries; thats how I found my pack. It was an outfit that specialized in replacing and refurbishing hybrid battery packs, so they no doubt had a bunch of contacts keeping an eye out for batteries to salvage. I see an ad all the time for a place that does Nissan leaf battery upgrades, and that might also be a good bet for tracking down a partial pack. I do not have experience with getting batteries shipped, but I do know it is possible.

I'm thinking it's going to be easier to just get a modern motor controller instead of shoehorning a modern pack to work with this system. That being said the car is wired to work like this right now, so maybe I should give it a try. Thoughts?

Yeah, I am starting to think that a new controller might not be a bad idea. 48 volts seems like it is just going to be a hassle to deal with. Parallel stings should really be avoided if possible. It will greatly simplify your BMS - which you should include with just about any lithium battery pack. And reconfiguring a pack would be a lot of work if you are going to plan on putting in a new controller later anyway. Anyway, I like the pictures, do you have any more of the motor (and what its approximate dimensions are?)

 

[xevion]

Yeah, I am starting to think that a new controller might not be a bad idea. 48 volts seems like it is just going to be a hassle to deal with. Parallel stings should really be avoided if possible. It will greatly simplify your BMS - which you should include with just about any lithium battery pack. And reconfiguring a pack would be a lot of work if you are going to plan on putting in a new controller later anyway.

As much as I want to slap two 24V packs in here and get this thing on the road, I want to do it right the first time even more. I'm thinking it might make sense to start with what the motor needs and work my way back. Size a good controller for the motor, then a good pack for the motor controller.

do you have any more of the motor (and what its approximate dimensions are?)

This is the best shot of the motor from today, I'll grab some more next time I'm under there. I had the motor info somewhere but I can't find it at the moment. I think this is a 3.5hp GE motor.

 

[MattsAwesomeStuff]

I wouldn't toss your golf cart charger.

I'd use it to charge your lithiums.

Just know that "48v" of lead acids can be as high as 60v. I would put a capacitor across the output of the charger and measure its voltage when turned on. You'll see the peak to which the charger actually charges. Size your lithiums to be slightly higher voltage than that, and they'll never overcharge. It's probably just a big dummy transformer with magnetic shunts to limit the max current (lead acid batteries too are an almost bottomless current sink when empty, the charger will have to gracefully limit the current so, no issues there).

A BMS... you could get away with a low-power e-bike or scooter BMS, if you need to use one at all. For how few cells you have, ~13 or so, I would be tempted to just buy 13 little 1/2" tall panel meters for $1 apiece, mount them to a sheet of plastic or wood, and give them a glance once in a while before driving to make sure they're roughly equal. Else, just top up the weak ones manually. This is what many e-bike builders do. With low cell counts, you have low likelihood of imbalance.

Speed control... lots of low voltage golf car controllers available on the cheap. Nothing fancy needed.

Damien's had a dead simple Prius DC controller brain on the backburner for a year or more, but if that ever gets fleshed out it would be more power than you'd ever use for something around $300-ish including the Prius inverter.

The thing I love about these cars is that they're so simple, and so terrible, that anything you want to do you can do, and it improves the car

 

[wjbitner]

I wouldn't toss your golf cart charger.

I'd use it to charge your lithiums.

Just know that "48v" of lead acids can be as high as 60v. I would put a capacitor across the output of the charger and measure its voltage when turned on. You'll see the peak to which the charger actually charges. Size your lithiums to be slightly higher voltage than that, and they'll never overcharge. It's probably just a big dummy transformer with magnetic shunts to limit the max current (lead acid batteries too are an almost bottomless current sink when empty, the charger will have to gracefully limit the current so, no issues there).

A BMS... you could get away with a low-power e-bike or scooter BMS, if you need to use one at all. For how few cells you have, ~13 or so, I would be tempted to just buy 13 little 1/2" tall panel meters for $1 apiece, mount them to a sheet of plastic or wood, and give them a glance once in a while before driving to make sure they're roughly equal. Else, just top up the weak ones manually. This is what many e-bike builders do. With low cell counts, you have low likelihood of imbalance.

Speed control... lots of low voltage golf car controllers available on the cheap. Nothing fancy needed.

Damien's had a dead simple Prius DC controller brain on the backburner for a year or more, but if that ever gets fleshed out it would be more power than you'd ever use for something around $300-ish including the Prius inverter.

The thing I love about these cars is that they're so simple, and so terrible, that anything you want to do you can do, and it improves the car

Hi Matt!

Not to quibble, but..

Size your lithiums to be slightly higher voltage than that, and they'll never overcharge.

If it's truly a big transformer, then the output voltage is dependent on the line voltage, which can vary significantly. The only guarantee is that the frequency will average to 60 hz. Even that
varies somewhat. Just wanting to be a little cautious, since some lithium's will burn quite energetically if over-charged..

Hi xevion:

Since your top speed is probably limited to ~40-45 mph, and your desired range is 50 miles (what's your desired budget?) I'd look for 2 Telsa packs which should give you ~10Kw, and then consider two more if needed for your range goals. Of course you have to figure out where they will fit.. Fun project!


Bill

 

[xevion]

A BMS... you could get away with a low-power e-bike or scooter BMS, if you need to use one at all. For how few cells you have, ~13 or so, I would be tempted to just buy 13 little 1/2" tall panel meters for $1 apiece, mount them to a sheet of plastic or wood, and give them a glance once in a while before driving to make sure they're roughly equal. Else, just top up the weak ones manually. This is what many e-bike builders do. With low cell counts, you have low likelihood of imbalance.

I do like the idea of keeping things simple, as long as it's not at the expense of safety.

Speed control... lots of low voltage golf car controllers available on the cheap. Nothing fancy needed.

Yea I'm hoping to be able to pick one of these up second hand, just need to figure out how big the motor is so I can size the controller appropriately.

some lithium's will burn quite energetically if over-charged

Yea this is something I definitely want to avoid. From what I remember the stock charger is just a huge transformer, I'll take a closer look tonight.

Since your top speed is probably limited to ~40-45 mph, and your desired range is 50 miles (what's your desired budget?) I'd look for 2 Telsa packs which should give you ~10Kw, and then consider two more if needed for your range goals. Of course you have to figure out where they will fit..

The speedo only goes up to 45, stock these things maxed out at ~35. My range will probably dictated to a certain extent by my budget. I'm hoping to keep the battery/bms/charger under $2k, but maybe that's just not realistic. A quick google puts tesla packs at around $800-$1000 so that might be an option. Some back of the napkin math puts the capacity of the original pack at ~10KW (and 500lbs). The original batteries were rated at a range of around 40 miles. What's needed to charge/maintain 24V tesla modules?

Original pack calculation:
200Ah @6V = 1200 Wh * 8 batteries = 9600 Wh

 

[brian_]

... I'd look for 2 Telsa packs which should give you ~10Kw, and then consider two more if needed for your range goals.

Of course this don't mean two Tesla packs (which would be heavier than the car), but two Tesla Model S or Model S modules.

These are 6S modules, so nominally 22.5 V per module (but over 24 volts each at full charge).

 

[xevion]

Of course this don't mean two Tesla packs (which would be heavier than the car), but two Tesla Model S or Model S modules.

These are 6S modules, so nominally 22.5 V per module (but over 24 volts each at full charge).

Haha yes, my mistake. 2 modules, not packs.

Sent from my Pixel 3 using Tapatalk

 

[OR-Carl]

What's needed to charge/maintain 24V tesla modules?

First off, have you measured that space under the seat? Most Tesla modules are big.

As for charging, and this goes for nearly all lithiums, It depends how risk averse you are. Overcharging lithium cells is bad news, and discharging them too low is also to be avoided. The safest way to do it is to have a BMS that monitors each cell in your battery, and reports that info over CAN a couple times a second. Then you have a CAN enabled charger that reads those messages and adjusts the charging current based on SOC. As soon as the first cell gets to your user-defined cut-off, the charger shuts itself down. At the same time, the BMS should be checking for cell imbalances and bleeding down the highest voltage cells to get all the voltages to match.

I personally would be very nervous to plug a dumb charger into a 10kwh bank of lithium cells and walk away. Under charging the bank would add a margin of safety, but you are then buying and hauling more battery than you are using. You also end up having to decide how many extra miles you are willing to wager in a game of "will-this-voltage-burn-down-my-car-and-maybe-garage/house." LFP cells, which do not experience thermal run-away, would be safer without a BMS. Still, over voltage can ruin the cells. I would think of a BMS as an insurance policy. A 24 cell bms from Thunderstruck is like 450 bucks, which would get you to about a 90volts-nominal system. It can tie into a CAN-enabled charger, or control a relay if you want it just to use it to disconnect a dumb charger. This may all sound like paranoia, but catching an e-bike battery on fire is not the same thing as lighting up a few hundred pounds of lithium cells inside a vehicle.

 

[brian_]

What's needed to charge/maintain 24V tesla modules?

They're 22.5 volts nominal, not 24 V.

In a Tesla (as in any other EV), the modules are managed by a BMS (battery management system) which monitors module temperature and the voltage of each group of parallel cells, tells the charger what rate to charge at, tells the charger when to stop, tells the motor controller how much power it can use, and balances cell groups by partially discharging the most-charged groups to bring them down closer to the voltage of the other cell groups.

Original pack calculation:
200Ah @6V = 1200 Wh * 8 batteries = 9600 Wh

Ideally, yes, but that 200 Ah capacity is for very slow discharge (over 20 hours, which is 10 amps for a "200 Ah" battery, so about 480 watts for the whole 8-battery 48-volt set). Useful capacity in a car will be a fraction of that.

 

[xevion]

First off, have you measured that space under the seat? Most Tesla modules are big.

Took some quick measurements, very conservatively the space I have to work with is 15"x43"x9" (l*w*h). A quick google puts the tesla packs at ~12"x27"x3" so I should be good for two packs should I choose to go that route. The photos below measure the height, length and width respectively at the narrowest spots.

I personally would be very nervous to plug a dumb charger into a 10kwh bank of lithium cells and walk away. Under charging the bank would add a margin of safety, but you are then buying and hauling more battery than you are using. You also end up having to decide how many extra miles you are willing to wager in a game of "will-this-voltage-burn-down-my-car-and-maybe-garage/house." LFP cells, which do not experience thermal run-away, would be safer without a BMS. Still, over voltage can ruin the cells. I would think of a BMS as an insurance policy. A 24 cell bms from Thunderstruck is like 450 bucks, which would get you to about a 90volts-nominal system. It can tie into a CAN-enabled charger, or control a relay if you want it just to use it to disconnect a dumb charger. This may all sound like paranoia, but catching an e-bike battery on fire is not the same thing as lighting up a few hundred pounds of lithium cells inside a vehicle.

Yea if I choose to go lithium I'm definitely not going to skimp when it comes to battery safety. A fully CAN-enabled setup may be beyond what's needed for this project, but we'll see.

Ideally, yes, but that 200 Ah capacity is for very slow discharge (over 20 hours, which is 10 amps for a "200 Ah" battery, so about 480 watts for the whole 8-battery 48-volt set). Useful capacity in a car will be a fraction of that.

Makes sense. This means my conservative estimate for the original battery capacity was even more conservative then I thought.

 

[Electric Land Cruiser]

Cool project! Nice find I have seen several of these but none as in good of shape as that one. Way cool that the original batteries still held a charge, that is really amazing.

A couple points about resurrecting vintage EVs I've learned that may help you:

-The battery bank is 48V nominal, but that means it probably has a max voltage of 60V and a minumum of 40V. You need to know what the minimum voltage cut-off is before choosing your lithium pack, or at least it's a good idea. For instance you might think "I want a 14S lithium pack because that gives me a nominal 48V!" but the lithium can be discharged much further than lead-acid, and most of the lithium's energy is in the bottom half vs the top half of the voltage curve. So the car will shut down at 40V but the 14S lithium pack can still be used down to 35V so there's energy left on the table. In this case you could go with a 15S or even 16S lithium pack and just make sure to keep the maximum voltage to 60V max and end up with more range.

-Like others I would be wary of a 45 year old charger, however if you test it rigorously enough there's no reason it cannot be used with a new lithium pack. Of course, no matter what you need a BMS for your lithium pack, but there's no reason you can't charge the pack with the original charger as long as the voltages are all compatible. You'll likely want a more modern charger at some point though, as you'll have way more capacity than stock to recharge and also probably want the option to use public charge points.

-Figure out what the motor is. Take some photos and run the model numbers. That will dictate whether you should try to keep the original controller (probably not) or get an aftermarket one. Will also dictate if you can increase voltage above 48V for more power or replace with a different motor.

-Luckily a 48VDC car is very similar to golf carts as brian_ mentioned so things like motor controllers, chargers, DC-DC converters, even Sepex motor controllers are all readily available and cheap!

Good luck resurrecting this little goofy car! Can't wait to see what it can do with modern tech.

 

[MattsAwesomeStuff]

If it's truly a big transformer, then the output voltage is dependent on the line voltage, which can vary significantly.

True, regardless, you would want a simple high voltage cutoff. No reason not to, it's simple, you already have contactors, and it will prevent 99.99% of rare problems.

Heck, I think even something dumb like a UPS charger would have that built in.

No need for fancy charging profiles or any of that BS. Feed it voltage lower than it's max voltage, have the power supply limit itself so it doesn't overheat, and let it charge.

The only guarantee is that the frequency will average to 60 hz. Even that varies somewhat.

Form what I understand, frequency is so consistently 60hz that clocks have run on AC based on this frequency for decades without needing adjustment. It takes things like, Texas' entire power grid failing for that to fail.

The safest way to do it is to have a BMS that monitors each cell in your battery, and reports that info over CAN a couple times a second.

That would stop a somehow random, spontaneous, catastrophic failure of a battery module, but is otherwise in my opinion absurdly paranoid. For an OEM concerned with liability and the most irresponsible owner, yes. See: Almost every ruined DIY EV has a story that goes "I lent it to someone and told them how to use it, but they ...". But for a DIYer? Not really practical.

OEM batteries don't just randomly, wildly start drifting apart while driving (or any other time). Has this ever happened, even once?

And if it does happen, the BMS isn't saving it. Best it can do is blow the canopy and eject, shut everything down. A BMS trickle isn't going to recover a failed cell.

Yeah, over an entire season and different weather conditions, many charge/discharge cycles, gradually, perhaps cells in a module could slightly drift apart, and some would need to be occasionally given a little burp of charge.

A crappy little passive BMS (not one that the whole driving circuit goes though) would keep these topped up. So, I would go with one of those, not a big vehicular-sized one if you didn't want to have to monitor them.

But like I said, on such a little vehicle, I'd almost just make a little column of panel meters to watch the voltage now and then.

A better backup, and a better sign of any looming catastrophic failure, is to have temperature sensors and cutoffs. A failing pack is a hot pack.

I personally would be very nervous to plug a dumb charger into a 10kwh bank of lithium cells and walk away.

It's up to everyone's choice. But, a good rule of thumb is to have a consistent level of paranoia, and to tackle the low hanging fruit across the entire project. At some point you'd be paranoid about not spending extra on special extra-durable tires or taking your suspension in for regular checks, etc, perhaps before you'd bother with this. Or you wouldn't drive it at all if you care about safety. What do you think the survivability is when this plastic cheese wedge gets into even a minor crash? It's going to crumple like saran wrap and tumble like dice on a craps table, it's no longer than it is tall or wide! Instead of spending $2000 to upgrade it, for $500 you can buy a used 90s Honda Civic that's, I don't think I'd exaggerate to say 10-100x as safe.

For the extra cost of a vehicular-scale BMS, I'd say just throw more batteries in and then undercharge them so there's even more overhead.

most of the lithium's energy is in the bottom half vs the top half of the voltage curve.

Mmm... no. It's almost all fairly steadily right around 3.7v. Something like 80% of it is 3.6-3.8v. Discharge curve is quite flat.

So the car will shut down at 40V but the 14S lithium pack can still be used down to 35V so there's energy left on the table.

I pretty strongly disagree with this point.

35v / 14 cells = 2.5v per cell.

That is not a smart discharge level for lithiums, nor is there any useful power to discharge that low.

40v / 14 cells = 2.85v per cell.

Even that is pretty low for lithium and most people will not let their batteries discharge that far. There is certainly all-but-zero energy left below 2.85v per cell.

42v might be a smarter discharge cutoff for most people. The difference energy between 42 and 40v is also nearly zero. It plummets fast.

Heck there's people who set their lower limit to 3.3v, (46.2v), which itself is like, 90%+ of the energy.

Also, the dash will almost certainly have a volt gauge on it that the driver would occasionally glance at, to make their own determination of their range. With such a short range vehicle, your mind is always partially on how much range you have left. The alternative being that the car would suddenly cut out mid-trip, and, you think it's best to beat the cells down to 2.5v before being forced to shut off? I don't think that's wise or useful.

Of course, no matter what you need a BMS for your lithium pack

"need" is a strong word. Lots of DIY EVers never did, most E-bike builders don't either.

He's really not using that much more battery than an E-bike or a scooter.

To travel 45mph (pegging the gauge), it would use ~92 Wh/mile. If he wants a 50 mile range that's only 4600 Wh.

To travel the speed the vehicle was built for (35mph), it would use 75 Wh/mile. 50 mile range means 3750 Wh. That's only a 4-slice toaster volume of battery. It's 375x 18650s.

I'm thinking it's going to be easier to just get a modern motor controller instead of shoehorning a modern pack to work with this system. That being said the car is wired to work like this right now, so maybe I should give it a try. Thoughts?

Your best money is probably spent on the controller, more so even than upgrading the batteries.

That said...

Half the reason most people do things is for the narrative of it.

If you just wanted the end result you'd buy... literally any other car ever made

I think a fun part of the stories you'd tell about this project, would be how awful it was to drive with the default 3-position speed controller. Don't keep the controller, but, do you want to miss out on the story of knowing what it was like to drive it the way it was built? I'd drive it for a few days at least so you can pick up some memories and situational context.

There's almost nothing to break. It's series/parallel contactors, and a resistor. That's your 3 speeds. No transistors allowed. Heck, I'd keep it around in the vehicle as a speed controller spare-tire kind of thing.

One of the nice things about EV builds is that your setup is modular and agnostic to the rest of the system. If you want to add battery capacity later, you can. If you want to upgrade the speed control, you can. If you want to upgrade the dash, you can.

You're looking at under 100 amps. Probably under 50 amps unless you're climbing a hill or accelerating. Almost any little used golf cart speed controller is going to be excessive for this. I wouldn't spend a lot of money here.

 

[brian_]

I think in all of these cell voltage discussions everyone should be careful about which lithium-ion electrode chemistry they are discussing. LFP (LiFePO4) voltages are significantly lower from most of the others.