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.
I just realized it may be a moot point for me, I had my degraded 2011 cells replaced with ones from a 2013 pack (same open can style as the one I purchased). Granted they didn't officially switch chemistry until 2015.
 
Finished a full discharge and charge test of the Leaf module (in 7.4V configuration) in parallel with 2s2p Boston Power Swing 5300 cells. The discharge test was done at the max power of my tester, 60W, which means current did vary as the battery voltage dropped. The charge was done at 5A, which is a bit lower than an L2 charge (~8.8A). I performed the charge termination at roughly the same level I've seen my car stop charging at, which is about ~1-2A in CV mode. Here's the discharge graph:

7abzbCe.png


So exactly as Mux found, the extender pack (in my case using LCO chemistry) delivers most of its energy toward the end of the discharge curve, which means it is providing a majority of the power towards the end as well.

Just between turtle and dead, the extender pack took the maximum proportion, about 58.1% of the total current, meaning it would have had to provide 159A, which is way more than the 26A rating of 2 Swing 5300 cells in parallel. Basically, this limits the minimum extender size to at least 12-13p, or ~17.3 kWH usable for this model of cell! Taking the calculation for a higher drain cell, the LG HE4 (which I should be getting samples of in the next day or two), the minimum extender size drops to 8p, or ~5.4 kWH usable.

hdlwOcD.png


The charge curve is nearly the inverse, with some bumpier behavior at the start of the curve. In charging, the BP cells took a maximum of 52% of the charging current, which again would be an issue in DC fast charging, limiting the minimum pack size to 6p of the Boston Power cells, or 16p (~10.9 kWH usable) for the LG HE4 cells, in order to take the ~65A that would be coming from the charger.

Early conclusion from this data; it's better to go with hybrid or EV batteries that are already sized/designed to take such large charging/discharging powers, if using a cell chemistry other than LMO. Otherwise, you can use "standard" Li-Ion cells but you will be building a very large pack (at least 11 kWH sized), and we still have the SoC estimation problem to resolve.
 
Thanks for sharing that, jkenny23!

Just one note: all 24 kWh batteries to date have the same chemistry, Cathode Active Material is LiMn2O4 (or LMO) with a pinch of LiNiO2.
The only upgrade in "Lizard" was the electrolyte.
(to know more, https://www.nec.com/en/global/techrep/journal/g12/n01/pdf/120112.pdf )

LMO batteries have a distinctive discharge curve, with the knee bend at ~3.65V
Discharger60Ah_small.jpg


Good work, gentleman!
 
Alright, I was wrong then. Good to know, and I always wondered if I was imagining that little nickel hump on the discharge curve...
 
Now for the real extender pack (Hyundai PHEV modules)...

First test is a discharge test of the Leaf module in parallel with 2s (out of the 8s full pack) of the Hyundai PHEV module:

cD7YuGX.png


This pack fares much better with the voltage curve of the Leaf, utilizing up to 95% of the "full" capacity from 4.11 to 3.0V by the time the Leaf reports dead (3.2V/cell). Still, just over a third of the capacity is delivered between VLBW and turtle. Also of concern, at peak, the Hyundai module provided 81% of the load current, meaning it will deliver up to ~72.9 kW peak, a bit higher than the pack's 68 kW max discharge spec.

And for charging:

xk2TXOf.png


The extender takes a maximum of 68.1% of the input current, which is no concern since the pack is rated for 54 kW charging (more than CHAdeMO can even deliver in total to the Leaf).

Finally here's a "baseline" of the Leaf module by itself. Capacity is 42.15 AH from 4.11V to 3.2V (dead). With the Hyundai module in parallel, this number increased to 63.52 AH (+50.7% capacity).

qzIsRUE.png


And here's a picture of the test setup in case anyone's interested:

hLFUyoV.jpg


Currently I am running a test with all 3 in parallel (Boston Power Swing 5300 in 2s2p, Hyundai PHEV module, and Leaf Module) to see how easy it is to add capacity in smaller 18650 packs later once the initial ~10 kWH high power extender pack is installed.
 
Hey mux, how much capacity mis-match between modules do you think is tolerable? I have only tested 2 modules so far, but they are coming in at 21.79 and 22.40 AH. I've got one spare module so I can pick the best 12/13 in case there's an outlier, but since these are all from different packs/vehicles there may be some large differences between them. Have you done any manual measurements of your modules to see how the voltage drift is over time between them?
 
That is not a normal mismatch, that sounds like heavily used modules. Typically, at beginning of life cell mismatch in a pack is at most a couple tenths of a percent.

If cells were 3% apart like yours, you'd need balancing current in the order of ~0.3% of the charge current in normal use (that seems to be the rule of thumb BMS engineers have decided on, it seems to work for car and bike batteries). The Leaf has a BMS that can only do something like 15mA per cell, i.e. on the old 58Ah cells that's just 0.025% of nominal capacity.
 
mux said:
That is not a normal mismatch, that sounds like heavily used modules. Typically, at beginning of life cell mismatch in a pack is at most a couple tenths of a percent.

If cells were 3% apart like yours, you'd need balancing current in the order of ~0.3% of the charge current in normal use (that seems to be the rule of thumb BMS engineers have decided on, it seems to work for car and bike batteries). The Leaf has a BMS that can only do something like 15mA per cell, i.e. on the old 58Ah cells that's just 0.025% of nominal capacity.

I take it your modules (VW ones I think you mentioned?) are all from a single vehicle then, and thus well matched to each other? The individual cells within each 8s module are well matched to each other, but my concern is the total 12s(8s) pack, where each group of 8 cells might be a different capacity than another.
 
Nope, I have modules from 5 different cars. They're all well within a percent of each other.
 
Finally got around to processing the data from the 3 cell type parallel test; the conclusions are a bit more complicated than I was expecting. On the discharge side, even though the PHEV pack made up 69.1% of the extender capacity, it only provided up to 53.2% of the current, with the much smaller Boston Power pack providing up to 34.7% of the load. This means you would still need a very large Boston Power based pack (more than 8p or about 13.8kWH) worth, likely more, since it would also increase its current proportion as the capacity increased.

UKJxQEU.png


On the charge side, there are no concerns, both of the extender battery types can handle at least 2C. The Boston Power pack would get a max of 1.67C and the Hyundai pack a max of 2.28C.

4i4b9Us.png


Next step is to try 2s5p of the LG HE4 cells since those can handle significantly more discharge, at the tradeoff of lower charge rate.
 
I should probably just mention here that I did try on my last big trip to do a full Leafspy log, but for some reason the logs came out really bad. Lots of large gaps and mangled data. Basically unusable. I was hoping to get a nice current/voltage graph out of it, but alas.

I'm now programming my own logging tools that are directly on the CAN bus so I have basically continuous logging of arbitrary EV-CAN values. I'm not sure why my leafspy is so unreliable, might have to do with the BT ELM327.
 
I have noticed that Leaf Spy drops connection to the BT dongle and/or vehicle occasionally, likely the cause of the problem.

For my own curiosity I was wondering how the extender would behave if using a configuration other than 96s, given the LMO cells charge to a lower max voltage. Looks like you could potentially get more capacity (8.36kWH at 4.29V 92s, 8.23kWH at 4.2V 94s, 7.73kWH at 4.11V 96s), but if I'm interpreting the voltage curve correctly, it would exacerbate the "more capacity on the low end" problem we have seen:

8MMu7EJ.png


Note the second graph is just an extrapolation of the 2s discharge behavior, I don't have the means to test this in a real scenario.
 
92S would be very dangerous; you're either severely overcharging cells or you have to include special circuitry to actively limit charging to a lower top voltage and monitor top-unbalanced cells so none of them go over their nominal voltage. I bet that if you'd remove the extra charge above 4.1V, the additional capacity would be minimal.

Edit: just to be clear, there's basically no difference in the cell degradation different chemistries see above 4.1V. All lithium ion chemistries get Li+/Li2+ instability/passivation at high voltage and have drastically lower cycle life at their maximum rated voltage compared to 4.0-4.1V, even though the actual capacity in the top stretch is pretty limited. For a vehicle, where you really don't want to be changing out batteries every 100 cycles, I'd keep to 4.05-4.10V per cell at most for any chemistry, unless you know damn sure what you're doing.
 
Nope, my Leafspy fucked the logging up. Working on better logging solutions right now.
 
I was able to run a 3 cell type parallel test with the LG HE4 cells (in 5p configuration). On discharge, the LG pack provided up to 43.2% of the current, with the Hyundai pack providing up to 53.7% of the load. That puts this 5p LG pack a bit undersized, since it can only provide 100A max (20A/cell) but would source ~130A at that proportion under full throttle. Increasing to 7p should be sufficient, though further tests are needed to see how the proportion scales with pack size (might be a vicious cycle of larger pack => larger load proportion => more current per cell...)

PBoUP73.png


Charging is a bigger problem, the LG pack took up to 40.2% of the charge current, which would give a worst case of 50.25A for the pack, or 10.05A/cell, more than twice the rated max charging current of 4A. To get that down to 4A the pack size would have to be increased to at least 13p, or 11.5kWH nominal.

qeNldHc.png


Hate to say it, but common 18650 or similar form factor cells just aren't cut out for EV use. The last hope would be Sony VTC5As which I've got some of (2nd hand), which are specced at 2.5AH capacity, 6A max charging current, and 35A max discharge current (for short duration). These should be workable in a ~9p pack or ~8kWH nominal, which is still a pretty big pack and will set you back $3520 from China, plus a set of Hybrid cells for $900...

Or you can just wing it with smaller packs and hope they don't catch fire (I do not condone this).

To summarize, here are the estimated minimum pack sizes based on the limited empirical data I've taken so far:
L82nOcc.png
 
Yep, this goes back to the discussion before all this testing: you HAVE to make an 18650 pack pretty darn big to have it make sense. 15kWh+ before you can use... any available cell, 20kWh+ before you can use relatively affordable commodity cells and much larger still to use the cheapest ones.

Fortunately the cost of these cells is pretty low, so it's still much cheaper than buying anything new on the market as far as large-battery BEVs go. But it's not something you can bootstrap from a small pack.
 
Back
Top