NexPower V3 hybrid battery unveil - Sodium-ion battery

Here's an extreme example to make it simple. Let's say I'm planning to drive 400 miles on a highway that is perfectly flat and at 65 mph the entire way, no stops. Will it matter if my HV battery has a capacity of 1ah or 100ah? In this case, it's more likely the 10ah battery car will have the lower mpg due to the extra weight.

 

again...

I see a lot of words from folks without a PIP and without an upgraded battery...

 

They do, or stay closely to the same mpg. Im not arguing they dont, unless you stack a lot of them up until you start experiencing diminishing returns because of weight issues primarily.

Not that it makes sense, but pretend someone attempted to make their prius to match a tesla for battery capacity. Extra weight would reduce mpg since itll be a signficiant weight at that point. It doesnt make it cost efficient to charge a EV with any gas or diesel generator, and hybrids take advantage of electromagnetism for its alternatives to torque converters, clutches, gears, and regenerative braking. Batteries are meant to be short term pump and dump, nothing more. EVs on the other hand only fuel is battery, so it makes sense to increase its capacity. While you will most likely gain some mpg by increasing battery, its only the mpg you were losing beforehand from factors like wasted electricity that couldnt be collected in time.

There is a video on youtube that shows two toy cars both going down a ramp, one with a battery and motor, other with nothing. They both end up at the same spot just at different times. The law of physics about energy not being created or destroyed applies everywhere. Adding more batteries is more equivilent to adding a bigger fuel tank for gas. It shouldnt have any effect on mpg.

i will end by saying, the amount of batteries from a stock prius is a good balance of not too little to where you cant store enough electricity to use, and not too much where its too costly and ends up removing space and adding weight. I would personally love to have a 20 gal tank even if it does reduce mpg by a point or two. But you know, they have options like prius prime, rav4 prime, etc. for people that do want extra batteries so they can drive 40 miles on EV. However, they dont sell because people dont wanna pay $20,000 more for plug in option, so their supply is limited.

 

Something sounds off. 60k coke cans and 14 cubic feet?
Convert 60k coke cans to ounces=60k x 12 =720k ounces
Convert to gallons 720k/128 =5625 gallons of coke
A 16' x 32' swimming pool at 4.5' average depth holds ~17,000 gallons

maybe I'm missing something

 

You are correct, I probably meant to write '14k' instead.
Unfortunately PriusChat doesn't allow me to edit posts after a few hours, so I can't correct my original math.

Let me do the math again, showing more work this time.
First, I didn't just convert the volume of coke, since it's stored in a non-optimal cylindrical can.
For that reason, I used the outer box dimensions of a 12 pack of coke (15.5 x 4.75 x 5.25"), which is 0.224 cubic feet.
Therefore, QTY60000 coke cans would consume ~1120 cubic feet, which is ~8400 gallons.

Root cause:
If you compare this result to my previous answer, you will see that it's exactly twelve times less volume... last time I incorrectly used "1 can is 0.22 cubic foot", whereas the correct answer is "12 cans is 0.22 cubic foot".

Post:
Had I done the math correctly last time, I would have compared this amount of volume to something else besides swimming pools. For example, this capacitor bank has the same volume as ~QTY5 Toyota Prius vehicles.

I agree with your gallon estimate.
Using the revised number I just calculated, you would theoretically 'only' need half a typical swimming pool's worth of coke-can-sized capacitors to source enough energy from capacitors to charge 500 Wh into sodium-ion battery.

Thanks for keeping me honest. Some are likely to criticize me for making an honest math mistake on a hypothetical example... I will counter in advance that that's why I like to show my work... so that people can challenge any incorrect assertions I make.

FYI: I submitted that post at 7:30AM, after a long night with zero sleep... I wasn't functioning anywhere close to my 'A' game, as I've been rather sick for almost a week now. Not an excuse, but rather a statement on why I made a back-of-the-envelope miscalculation.

 

If the replacement batteries have higher energy density for the same size and weight, then 'more batteries' can certainly increase MPG.

And in this case, the replacement batteries not only have greater energy density, they have greater capacity, they are the same size, and actually weigh less than the stock batteries.

Unless one lives in flatland, there is going to be an MPG increase. If one lives in a somewhat hilly areas, perhaps a quite substantial MPG increase.

 

Note: This post contains numerous assumptions. Until I actually evaluate the hardware, I am merely making educated guesses.

I agree that MPG will nominally increase if the pack is:
-lighter (I'm all-but-certain that the V3 pack is lighter than an OEM NiMH pack), AND;
-has similar (or better) ESR (my calculations suggest V3 is in the same ballpark), AND;
-stores more energy (I have doubts, see below).

I'm not certain the V3 pack can actually deliver more usable energy (compared to an OEM NiMH pack). My doubt stems from sodium-ion's large voltage delta over SoC. At best, I predict that less than half the energy stored in these sodium cells is actually deliverable to the Prius.

Let's assume the non-GT V3 pack contains QTY70*** 10 Ah (nominal) sodium-ion cells. So then we nominally have a 2100 Wh pack, but only if the Prius will allow assist all the way down to 140 volts. Prius domain experts: please chime in with actual values, but for now I'll use the following ballpark figures for the Gen3 Prius:
-ECM basically disables assist below ~210 volts
-ECM basically disables regen above ~250 volts

If my estimates are close, then each cell voltage will span from ~3.00 to ~3.5 volts, which means the Prius will only consume at most 30% of the actual energy inside these sodium cells. I will modify this statement in the future, once a Prius domain expert provides more accurate numbers.

Whatever these numbers actually are, they're going to span a small portion of sodium-ion's SoC discharge curve. To get around this limitation, NexPower would need to spoof the pack voltage measurements to the Prius' BSU (or replace the BSU entirely). Maybe they're doing that; I'll know once I evaluate their hardware.

***Note: I still don't know if this is true. My initial guess (based on NexPower's video) was QTY66 cells, which I later revised to QTY70 (70 is evenly divisible by 14). I then doubled my estimate to QTY140 cells after realizing there arere actually QTY2 sub-packs, whereas the CAD model shown in the video was (maybe?) only showing QTY1 sub-pack. However, right now I once again think there are QTY70 cells.

 

The Gen 2 will engage the ICE to charge the hybrid battery when it reaches ~42% soc and typically controls in the band of 42% to ~60-62%. I don't have accurate numbers for any other models

 

Using some of mudder's preliminary estimations:

Gen 2 and 3 Prius limit SOC to between 40% and 80% = 40% maximum usable capacity of the 6.5 Ah OEM NiMH battery. (not sure if linearly translating SOC to Ah capacity is accurate, but just a rough estimation)
6.5 Ah x 0.40 = 2.6 Ah NiMH
10 Ah x 0.30 = 3.0 Ah Na ion
3.0/2.6 = 1.15 = 15% higher maximum usable capacity with a 10 Ah Na ion battery.

Although it would be only a modest usable capacity improvement, actual usable energy might be a little higher than that, due to lower internal resistance. Cell life would likely be long, due to shallow cycling. And it might be inherently safe from overcharging and over-discharging while using the OEM battery management.

 

Spray mud on it and it will smooth right out!

 

Unless V3 has a true BMS that can disable regen, calling it 'inherently safe' is a misnomer. While using a small portion of the SoC range certainly increases headroom, it doesn't account for the various failures that will eventually happen to all cells.

 

Almost there

 

Thanks, mudder. "Headroom" was the word I was looking for. In only that particular aspect, would you say that it provides a good margin of safety when combined with adequate cell failure detection?

The OEM BMS appears to disable or severely limit battery use when module block voltage differences approach nominal cell voltage. With fewer cells per measured block, and the higher nominal Na ion cell voltage, do you think the OEM BMS would be even more sensitive to defective cells than with the NiMH cells? Or is even that level of cell failure detection inadequate for Na ion batteries?

 

Thanks, Xeico. I do not have a complete understanding of all the battery math. I am a little confused by why voltage would be used for this linear calculation of Ah capacity, because the voltage curves that I have seen are very nonlinear when plotted with a constant charge/discharge current.

 

Prius basically working from 210V to 252 V.
You got 28 "blades" for standard NiMH battery.
28*7.5= 210
28*8.93= 250

One blade Has 6 NiMH cells that have working range from 1 to 1.5 V.
7.5V/6= 1.25V
8.93V/6= 1.48V

Roughly speaking, from 1.45 to 1.2 V we can say that this is an almost linear graph.

 

For NiMH? Based on the graphs I'm managing to find in a quick search, that seems to make the word 'almost' do quite a lot of work.

 

Doesn't the fact that you have 100% = 1.33V bother you at all?

 

Maybe it bothers whoever made the graph. That wasn't me.

I picked one of the simpler graphs from a quick search like this:

nimh discharge curve - Google Search

There are plenty of others, many of which appear suggest that temperature has a lot to do with the voltage at 100%.

I'm not seeing any, in a quick glance, that would make me agree "from 1.45 to 1.2 V we can say that this is an almost linear graph".

 

Here is an example, unfortunately this is not from the datasheet from Panasonic, but you should not compare ordinary batteries with batteries for the automotive industry. The Prius has high current cells and the graph will look like a red line.
Yes, there is a drop from 1.48 to 1.38, but it is not as significant as below 1.2 V.

Unfortunately I don't have a new or very good NIMH blade to test with, but I'm pretty sure that most of the charge is above 1.2V

If you can, tell us about yourself, what kind of battery do you use and are you a member of Jack’s team? Maybe I’m just wasting my time explaining?

 

I'm neither on Jack's team nor a customer; the battery in my car is the NiMH one that Toyota put there.

I merely made post #96 because I read "Roughly speaking, from 1.45 to 1.2 V we can say that this is an almost linear graph" in #95, and that didn't seem to fit what I remembered from looking at NiMH discharge curves, and so I searched a bunch of them up and it still doesn't seem to fit, and even the red trace in your latest attachment would be hard for me to describe as "almost linear" unless maybe in a narrower range like 1.28 to 1.25. In the range up to 1.45, to my eye at least, the curvature is quite noticeable.