Extra Battery, How to Integrate with 24kWh Traction Battery?

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
What happens with SOC, SOH and Hx values if two batteries are connected simultaneously?
Well, SOC will be wrong. It will drop like it would according to wasted energy. But then it will stop
dropping at 3-4 bars left. And then it keeps that for long. Until it starts dropping again.
SOH value will recalculate up and up. Until 103%. Which is maximum. Even though actually, there
is more kWh of energy available as there are actually two slightly worn batteries.
Hx value will recalculate to realistic value. It appears, that Hx value is the resistance of vehicle's
battery pack
. 100% is factory default value. In case of two batteries, it will skyrocket. But it will be
around the average of two separate batteries (so in case of two new batteries, it will be 200%).

SOH_Hx_Nissan_leaf.jpg
 
arnis said:
What happens with SOC, SOH and Hx values if two batteries are connected simultaneously?
Well, SOC will be wrong. It will drop like it would according to wasted energy. But then it will stop
dropping at 3-4 bars left. And then it keeps that for long. Until it starts dropping again.
SOH value will recalculate up and up. Until 103%. Which is maximum. Even though actually, there
is more kWh of energy available as there are actually two slightly worn batteries.
Hx value will recalculate to realistic value. It appears, that Hx value is the resistance of vehicle's
battery pack
. 100% is factory default value. In case of two batteries, it will skyrocket. But it will be
around the average of two separate batteries (so in case of two new batteries, it will be 200%).

Does regen improve any over stock "new" battery state? Regen seems to be limited by reduced Hx values, so I wonder if the converse is true? That's one of my biggest complaints is the super weak regen ability of the Leaf. Almost all of my driving is city driving, so I'm often having to stop for traffic and traffic lights anyways, my efficiency would go up significantly if I could utilize more regen, but the Leaf is just too timid. Compared especially to my Prius which gives you full ~20-30kW of regen at any speed, until the tiny battery is full anyways. Perhaps a second motor attached to the rear wheels plus a dedicated controller for "external" regen would be the only option for improving it significantly?
 
I'm pretty pretty sure that none of those values are going to change as they are computed and measured by the BMS, which is inside the LEAF battery pack. The car will probably just start thinking you are doing a bang up job on driving economy.

I'm skeptical that charging of an add-on can be done using the LEAF's internal charger. I find it more likely that the secondary will need it's own charger. For one thing, without that you would have no way to deal with charging if the two packs ever fell out of sync voltage-wise.

You might be able to wire it to the same inlet port like the BRUSA integrations that were done, although it will be a little trickier dealing with the case where the secondary needs to charge, but the primary is full and doesn't start the EVSE.
 
davewill said:
I'm pretty pretty sure that none of those values are going to change as they are computed and measured by the BMS, which is inside the LEAF battery pack. The car will probably just start thinking you are doing a bang up job on driving economy.

I'm skeptical that charging of an add-on can be done using the LEAF's internal charger. I find it more likely that the secondary will need it's own charger. For one thing, without that you would have no way to deal with charging if the two packs ever fell out of sync voltage-wise.

You might be able to wire it to the same inlet port like the BRUSA integrations that were done, although it will be a little trickier dealing with the case where the secondary needs to charge, but the primary is full and doesn't start the EVSE.
Agreed that an external secondary battery shouldn't affect main battery numbers because those are measured off the main battery's BMS, not the motor controller or battery charger. Basically the BMS will just think you're apparently going downhill because you're using less kWh's per mile out of the main battery.

Charging another secondary traction battery off the main charger has been done already. Yes, if they are out of voltage there could be a problem. But that would only be a problem before or after charging or running the car. During charging or driving, there is no way both batteries could "fall out of sync voltage-wise" because they would be connected in parallel. It would be like two pressure tanks hooked up at the same time, neither is going to be able to have more or less voltage than the other. However, when you shut off the car and charger both batteries will disconnect. If one has more leakage current than the other or one battery gets hotter or colder than the other it could end up with a different voltage than the other once you go to turn on or charge the car. This could cause a sudden, but temporary rush of current from one battery to the other. The solution is to put an Arduino system that measures the main bus and the auxiliary battery's voltage and connects the auxiliary traction battery only when both voltages are the same. If you get home and plug in right away both battery's should have the same voltage and both begin charging unless one got a dead cell or loose connection or something in the second battery, which you wouldn't want to charge anyway.

I've contemplated doing a second battery with LiFePO4 cells. They have a narrower voltage range than the Leaf's Li ion cells, which could actually be advantageous. That way the main battery would always charge up to a higher voltage and would likely discharge to a lower voltage than the auxiliary battery. That way the Arduino system would always start with the auxiliary battery disconnected, and after a while of driving the main bus would always end up matching the auxiliary battery's voltage. Then the auxiliary battery would connect until it's depleted. Then it would be disconnected and leave the main battery to finish discharging. Charging would be the same, only in reverse.
 
IssacZachary said:
I've contemplated doing a second battery with LiFePO4 cells. They have a narrower voltage range than the Leaf's Li ion cells, which could actually be advantageous. That way the main battery would always charge up to a higher voltage and would likely discharge to a lower voltage than the auxiliary battery. That way the Arduino system would always start with the auxiliary battery disconnected, and after a while of driving the main bus would always end up matching the auxiliary battery's voltage. Then the auxiliary battery would connect until it's depleted. Then it would be disconnected and leave the main battery to finish discharging. Charging would be the same, only in reverse.

I would be very cautious of using a second battery with a voltage curve drastically different than that of the Leaf's battery; basically the current out of the secondary battery will be very different depending on what the state of charge of the 2 batteries are at. I've done a couple such experiments on the small scale, putting a high capacity power cell in parallel with a lower capacity higher internal resistance cell with a different discharge curve, and while the experiment was done at high rate of constant discharge, it did show that the power cell (akin to the Leaf battery) had most of the load for the first half of discharge, but the lower capacity "auxiliary" battery took a majority of the load toward the end of the discharge (details here: https://secondlifestorage.com/t-Different-Size-Parallel-Cell-Experiment)
 
IssacZachary said:
... The solution is to put an Arduino system that measures the main bus and the auxiliary battery's voltage and connects the auxiliary traction battery only when both voltages are the same. If you get home and plug in right away both battery's should have the same voltage and both begin charging unless one got a dead cell or loose connection or something in the second battery, which you wouldn't want to charge anyway. ...
Yes, the voltage monitoring is assumed, I think. But if you can't plug in right away for some reason, or the Arduino decides to crash, and you do end up charging the main battery without charging the secondary, you would have no way to get the secondary charged short of draining the main down to the volatage level of the secondary. In an extreme case, it could even be impossible to get them in sync again just by draining the main. If the idea is get this reasonably foolproof, it needs a better plan than just assuming it won't happen.
 
jkenny23 said:
I would be very cautious of using a second battery with a voltage curve drastically different than that of the Leaf's battery; basically the current out of the secondary battery will be very different depending on what the state of charge of the 2 batteries are at. I've done a couple such experiments on the small scale, putting a high capacity power cell in parallel with a lower capacity higher internal resistance cell with a different discharge curve, and while the experiment was done at high rate of constant discharge, it did show that the power cell (akin to the Leaf battery) had most of the load for the first half of discharge, but the lower capacity "auxiliary" battery took a majority of the load toward the end of the discharge (details here: https://secondlifestorage.com/t-Different-Size-Parallel-Cell-Experiment)

True. The LiFePO's would have to be able to handle perhaps full current. This may be possible if the auxiliary battery is also around 24kWh or more, or made of high discharge cells. Of course the cables have to match the current.

I'd love to throw 60kWh of LiFePO's on a trailer and be able to go anywhere.

davewill said:
IssacZachary said:
...In an extreme case... the idea is get this reasonably foolproof...
Just don't discharge below 0%. Or disconnect the auxiliary battery when it gets down to 10%. If the auxiliary battery is small enough you may not need to even worry about mismatched voltages after it's installed.

On the other hand I've always wondered what if you took a second used Leaf battery and wired the car to switch between batteries. Apparently this has been done. That way mismatched voltages are of no problem whatsoever. Then I also wonder what would happen if you add bigger cells, or cells in parallel, to the cells in a Leaf battery.
 
As long as you wire entire packs in parallel, there should never be an issue with charging and discharging. I don't see how it's in any way safe to do hot-swapping (or hot-switching, I guess). What if your code glitches? What if your arduino's voltage sensing drifts? What if a cosmic ray hits your microcontroller? These kinds of things are done in automotive hardware and battery switching, but they require intrinsically safe coding practices and hardware redundancy. If you do it wrong, what you get is something close to short circuit currents running between packs.

Just put packs in parallel. It's the only safe way to go about adding capacity, outside of replacing the main battery pack.

Also, out of curiosity, IsaacZachary, why would you want to use LiFePO4s in automotive? To me that seems really strange, but I might be missing something. Here's what I know about this tech - and I consider myself fairly knowledgeable considering batteries are core to my line of work:

- LFP used to have a flatter discharge curve, but NCA pretty much has the same characteristics, yet at a higher voltage
- LFP used to have better power density, but that's... just not true anymore, it's a lot worse than any modern chemistry now.
- LFP used to have higher cycle life, but that's totally not true anymore
- LFP used to be considered more environmentally safe, arguably the cobalt in NCA still makes it a bit more toxic but with 811 NMC coming up this is moot as well
- LFP cannot thermally runaway. It technically does have this advantage, as it makes it a somewhat inherently safe battery, but this aspect is so incredibly moot these days. NCA is so close to inherently safe that you have to massively abuse it to get it to do anything strange. The higher internal resistance and lower energy density (thus higher system complexity) of LFP makes it, IMO, more dangerous than more modern chemistries.

The low energy density (~2.5x difference with NMC) also makes that you really do need a trailer to pack 60kWh of LFP (=1150 lbs!!), whereas you can probably fit 60kWh of NMC (=450lbs) in the stock battery shell and a few briefcase extenders in your trunk, likely without even increasing total vehicle weight if you replace the original pack. Much more convenient, and always available, not just when you're pulling a trailer. Not to mention the cost difference and the sheer amount of extra work. 60kWh of NMC is still somewhat affordable, LFP is... just crazy. You can buy a new 2018 Leaf for that kind of money.

LFP really only makes sense as a lead acid replacement in front, not as a main traction pack.

I don't believe that one crazy video where the guy, out of frame, pretends to switch battery packs on a Leaf. Sure it could work, but I don't consider it demonstrated or particularly useful. Also, I don't consider the Leaf pack to be particularly desirable as a range extender pack. It's really heavy for its capacity and the chemistry they use inside is really inferior to even stuff like Yuasa LEV50s. As stated before, they somehow managed to get only 24kWh into a volume that can comfortably hold 45-50kWh, and still get significant voltage sag on just 80kW of load and very high wear even in mild climates. The Leaf battery is just not very good overall, I wouldn't use it for anything tbh.
 
mux said:
Also, out of curiosity, IsaacZachary, why would you want to use LiFePO4s in automotive?

Thanks for the incouragement! (Really! I'm not being sarcastic.)

Only because

  1. I can get them cheap right now. Cheaper by the kWh than any other battery chemistry, although not by much.
  2. Because they are safer in that they don't ignite or have a spillable liquid electrolyte.

Other than that, those are the only two reasons I've contemplated using them.
 
mux said:
As long as you wire entire packs in parallel, there should never be an issue with charging and discharging. I don't see how it's in any way safe to do hot-swapping (or hot-switching, I guess). What if your code glitches? What if your arduino's voltage sensing drifts? What if a cosmic ray hits your microcontroller? These kinds of things are done in automotive hardware and battery switching, but they require intrinsically safe coding practices and hardware redundancy. If you do it wrong, what you get is something close to short circuit currents running between packs.

...

He's describing a second pack in parallel. I agree you don't need that kind of monitoring if you're inside the main pack's contactor and the cells are never disconnected from one another, but if that's not the case, then something has to verify that it's safe to close the secondary pack's contactor.
 
davewill said:
He's describing a second pack in parallel. I agree you don't need that kind of monitoring if you're inside the main pack's contactor and the cells are never disconnected from one another, but if that's not the case, then something has to verify that it's safe to close the secondary pack's contactor.

Yes. Really any battery in parallel outside of the main battery needs to have some sort of voltage comparing contactor closing device for safety.

mux said:
As stated before, they somehow managed to get only 24kWh into a volume that can comfortably hold 45-50kWh...
I really, really would want to know, what would happen if you gutted out all the Leaf cells and replaced them all with something totally different. Would the Leaf recognize Ah capacities above that of the current cells? Would it freak out if voltage drop isn't as bad as it is in the current leaf battery?

Ideally, if a guy could take a wrecked Leaf battery, sell the cells and eBay and fill it with whatever cells make the most sense, hopefully doubling the capacity, and do that at no major cost to weight, that would be awesome! An actual thermal management system could be added too when doing this.
 
IssacZachary said:
Ideally, if a guy could take a wrecked Leaf battery, sell the cells and eBay and fill it with whatever cells make the most sense, hopefully doubling the capacity, and do that at no major cost to weight, that would be awesome! An actual thermal management system could be added too when doing this.

That would be the ultimate pack replacement. If you could work in a liquid cooling system that would be even better, there's conveniently located cooling lines just above the pack which run back to the charger; the radiator in the front of the Leaf is more than capable to handle cooling the pack as well (really frustrating that they left it out for "cost". It doesn't even have to be a "real" liquid cooling system like Tesla does, but maybe a liquid to air heat exchanger and a set of fans for air circulation inside the closed pack.
 
Best current value for automotive scale batteries is ~$135/kWH for Volt modules (example from a quick eBay search, likely cheaper if buying a whole pack from a local salvage yard: https://www.ebay.com/itm/173190944540 ). Tesla modules (NCA) are pricier at $285/kWH (again on eBay, may have slightly better luck locally, but they're in high demand I think).
 
Ah, yeah, cost is a logical reason to go for something specific.

IssacZachary said:
I really, really would want to know, what would happen if you gutted out all the Leaf cells and replaced them all with something totally different. Would the Leaf recognize Ah capacities above that of the current cells? Would it freak out if voltage drop isn't as bad as it is in the current leaf battery?

Ideally, if a guy could take a wrecked Leaf battery, sell the cells and eBay and fill it with whatever cells make the most sense, hopefully doubling the capacity, and do that at no major cost to weight, that would be awesome! An actual thermal management system could be added too when doing this.

Somebody has done this before, like a dozen times: replacing the old 24kWh cells with new ones. The BMS just adjusts to the new cells. It has nothing really to freak out over. Apparently the BMS is happy to both monotonically decrease and increase capacity.

I'm almost certainly going to do this kind of mod in the future, but not this year. Probably only by the time we regularly have to drive very long stretches without good fast chargers. So.... possibly never. Also: We don't have a Dutch version of copart and Leafs are a lot less common in Europe than in the US, so it's quite a bit more expensive to get a hold of a wrecked Leaf battery here. But once one pops up at a good price, I'm going to grab it obviously.

IssacZachary said:
mux said:
... LFP ... but NCA ... but with 811 NMC coming up ...

Just a question. Where would you buy your cells? I mean, if it costs $40,000 to build a better battery, we might as well as just buy Teslas.

NKON.eu, probably the largest consumer-facing bare cell vendor in Europe, and it's like 70 miles from where I live. They have excellent prices on bulk, about 1.2-1.3x wholesale prices (typically $150-200/kWh). Although for my temporary extender pack I'm using second hand Yuasa LEV50s for about $100/kWh, but that's cheating. I don't think second hand cells are acceptable for permanent installation in a vehicle, unless you have a really solid BMS.

I'm not sure how you could ever conceivably fit $40k worth of batteries in a car :p. Batteries really aren't expensive at all anymore. If you're paying over $200/kWh, you are paying way too much. A couple grand worth of batteries and a few weeks of tinkering in your spare time should net car-sized battery capacity additions. Should be worth it for almost any BEV, as long as you enjoy the process.
 
mux said:
Somebody has done this before, like a dozen times: replacing the old 24kWh cells with new ones. The BMS just adjusts to the new cells.

I've seen where they've put 24kWh cells in a 24kWh battery. But what if the cells (or combinations of cells in parallel) contain more than 24kWh? Has this been done? Let's say you do fit 50kWh of cells in the original battery box. Are we sure it will work?

mux said:
Ah, yeah, cost is a logical reason to go for something specific... NKON.eu, probably the largest consumer-facing bare cell vendor in Europe, and it's like 70 miles from where I live. They have excellent prices on bulk, about 1.2-1.3x wholesale prices (typically $150-200/kWh). Although for my temporary extender pack I'm using second hand Yuasa LEV50s for about $100/kWh, but that's cheating. I don't think second hand cells are acceptable for permanent installation in a vehicle, unless you have a really solid BMS.

I'm glad you can find new cells at that price. Like jkenny23 mentioned
jkenny23 said:
Best current value for automotive scale batteries is ~$135/kWH for [used] Volt modules...
Here in the USA it is hard to find new cells that aren't $15 for a pack of two 18650's, which is more than $1 per Wh, or $1,000 per kWh, (so a 24kWh extender or replacement comes to about $24,000 USD), although there are those Fire cells for cheap, but we all know that they aren't worth a thing.

So used cells may have to be the way to go here in the States. However, there are overstock cells that pop up every once in a while, like the ones sold by Alarmhookup on eBay: 100 LG 18650's $150 His cells come to about $215 per kWh if you include the shipping into the price. However, many times they are in limited supply. So if you can get his cells, you might be able to build a 24kWh battery for as little as $5,500. But you can't if he only has 5kWh of cells available, meaning that again, used second hand cells end up looking like the way to go.

One thing I've even contemplated starting a hobby of harvesting old laptop batteries. I'm quite scared of using generic brand cells, even more than using used second-hand name brand cells, and by harvesting laptops you know you are getting name brand cells. If it's truely Sony, LG, Samsung or Panasonic, new or used, I'd be willing to put those into my car. But still, there are a lot of counterfits out there too. I've never heard of Yuasa brand cells or batteries. Hopefully they are rebranded name brand cells. When I see deals for $0.50 per kWh for some Fire brand cell on Alibaba.com I'm sure it's too good to be true.
 
IssacZachary said:
I've seen where they've put 24kWh cells in a 24kWh battery. But what if the cells (or combinations of cells in parallel) contain more than 24kWh? Has this been done? Let's say you do fit 50kWh of cells in the original battery box. Are we sure it will work?

I believe 30kWH cells have been fitted into a 24kWH Leaf pack, using the original BMS and swapping the original bus bars over. I don't think capacity will ever be reported correctly but the range will increase. I think there was also an issue of the old BMS only allowing a narrower range of voltage to be used compared to the newer 30kWH battery/chemistry. There is the option of WolfTronix's BMS microcontroller replacement, you would then presumably get full control of the start/stop voltages of the pack.

As for harvesting laptop cells, I think that's only acceptable for a purely hobby EV or stationary storage where if it goes offline then there's no harm and it's easy to debug. DIY powerwalls even using 80+ cells in parallel still have balance issues at 14s. Imagine a lower cell count in parallel with 96 in series, a lot easier to have problems. I think the alarmhookup new old stock type cells would be good enough, as long as you carefully test each cell. I've done some testing on Samsung ICR18650-22Ps I bought there in an 8s2p test pack, it kept balance for 30ish cycles I think but slowly drifted apart, even though I matched the cells as well as I could.
 
If there truly is no name brand bulk cell vendor in the States, this is a *massive* big gaping hole in the market. You have to be able to get batteries, right? How the hell is any small company supposed to survive otherwise? I would not be in business without NKON. Is this the same thing that's going on with solar panel and fruit prices in the US? I'm noticing a trend; anything green is hella expensive over there ;)

There really is no substitute for new cells. Modern cells have extremely well-matched initial capacity and degradation, especially when they are thermally matched in the pack. This makes even an un-BMS'ed pack stay perfectly in balance over >100 cycles. Hell, the vast majority of laptop batteries don't have a balancer, and they typically get 300+ cycles before something starts skewing, and even then it's usually not even due to cell imbalance, but excessive single cell drain by the BMS.

You throw all this away with recycled cells. Even if cells have a 1% mismatch and no drift, your BMS has to be able to continuously and reliably burn away tens of mA at typical EV charging/discharging rates per cell. An average automotive BMS frontend only has 5mA per-cell balancing, proportionally even less on something like the Leaf BMS. So you need some kind of custom, extra fortified BMS. There goes your entire financial advantage.
 
mux said:
If there truly is no name brand bulk cell vendor in the States, this is a *massive* big gaping hole in the market. You have to be able to get batteries, right? How the hell is any small company supposed to survive otherwise?

I've been told by some that on the east coast of the United States you can't even purchase your own contact cement to stick the sole of your own shoe back on, meaning you have to take it to a shoe repair shop so they can do it. If a guy makes his own company he has to be certified, which then gets him the licenses to buy what materials he needs. But if you're not certified you can't get what you need. And those that are certified can't just resell what they can get in bulk.

This is something I've mentioned before, that the EV seems to be just another step away from the DIY era. Sure, there are cheaper and better batteries out there. But the market is designed for these kinds of mods and materials to be left in the hands of aftermarket companies that will only do this if they see they'd make a profit from it. Once you add the certification and then liability all into the equation you end up with the final product being much more expensive, although with the idea it will be much safer than a DIY design.
 
Obviously we're only looking at one side of the coin here, so this is not a good representation of the market or industry as a whole, but I'm really flabbergasted here. How is anybody supposed to do innovation in such a restraining business climate? I know the 'this is America!'-trope is not much more than that, but I was kind of under the impression that the US is more free in what you can and can't do than, say, the communist gulag that's called the Netherlands.

There are no red tape barriers here, we have free market prices for almost anything electronics related. I am a one-man company designing electronics, and except for some very specialist components from the US with export restrictions, I can do whatever, whenever, however I want. CE certification generally only costs me a couple hundred, maybe 1500 euros if I need to do EMC compatibility or stuff like that. The only red tape you get to is when you start selling stuff for profit and there is some kind of safety aspect to observe.

So sure, designing and building an extender battery pack for commercial purposes is probably not going to be a worthwhile venture. But designing and building it for my own purposes? No problemo! The rules and regulations on the design are public courtesy of the EU, so I can make it compliant without paying a dime for certification or whatever else. Again, this is how innovation is supposed to work, right?

Guess it's offtopic, just had to write this to show my surprise at your story, regardless of how accurate it may or may not be with respect to these projects. Not saying you're lying obviously, just that I'm sure there is more to it.
 
Back
Top