Rainbow Prius Hybrid Powered by Lithium!

Theories can be tested. People are replacing their NiMH HV batteries with LiFePO4 batteries in their Prii without reprogramming the BMS.

How many have ran into issues because LiFePO4's "receive a charge differently" than NiMH's? We should have had dozens of people right now, either with damaged expensive LiFePO'4 batteries, or with fried Prius HV systems. Where are they? Is there some sort of conspiracy that is hiding all of them?

 

The question is, how long have these systems been in operation, how many cycles has the battery seen and what is the internal resistance after 1,000 cycles, 2,000 cycles and so on. At the same time, what is the capacity before the cells drop to 3v under a 0.5CA load after each 1,000 cycles?

Unlike in a house battery system or even a EV battery, hybrid batteries suffer multiple cycles every trip, so the number of days the vehicle has been driven will not indicate how many cycles the cells have been through .....

A lot more information is required to determine if the LFP cells suffer damage or not when subjected directly to the charging regime the NiMh cells are subjected to before they start to show signs of deterioration. A lot of the deterioration seen in NiMh cells/modules, is the loss of electrolyte, this can be easily replaced in the Prius modules, the electrolyte can't be replaced in an LFP cell, you simply can't get hold of electrolyte to top up the cells that need it.

T1 Terry

 

True. And I do have a certain degree of skepticism that the LiFePO4 batteries might not be worth it. This is why I haven't rushed out to get mine. There are pros and cons to everything. Will they last as long?, Will they get you as good as fuel mileage? Will they cost the same or less? Will they require special care that the NiMH batteries don't?

But that doesn't mean that they're doomed to fail or that the Prius's charging profile doesn't fit the recommended charging profile for LiFePO4 batteries just because of an old rule of thumb that you shouldn't charge batteries with a charger not made for them. So far the verdict is that they last for at least a few years, get you better fuel mileage, they are generally a little cheaper than the OEM NiMH battery, and they have no signs of degredation yet. Sure, they may all fail this upcoming year and people are going to be unhappy they paid nearly $2,000 for a battery that only lasted them 3 or 4 years, but it's looking like they're going last much longer than that.

 

Not sure where you got the idea from, but they have dramas when charged faster than 1CA, so to handle the 150amp charging current, the LFP cells would need to be a min of 150Ah.
If you doubt this is correct, best do the testing yourself and watch the cell voltage in relation to the charging current.

T1 Terry

 

The fuse or fusable link wire at each end of the cable between the 2 different chemistry packs is to protect that cable if there is short caused by the insulation coming into contact with and metal connected to the vehicle body.
The diameter of the conductor in that linking cable will limit the current flow and the closer to the voltage being the same in each cell group connected to to that cable, the less current will flow.
The fact that the NiMh cells will act as electrolysers when the voltage gets too high so that uses up the excess charge and drops the voltage, assists in the cell balancing between the 2 different chemistries.
I would still use a balancer across each 3.2v nom. LFP cell group because the cells need to be around the same voltage to be close to the same SOC ... it is only close, not perfect as far as balancing goes, but actually coulomb counting how much went into and out of each cell is just over the top battery management.

T1 Terry

 

In the 7 seat Prius v which Li battery type is being used by Toyota?...

REVVL V+ 5G ?

 

shifting this posts main response. Hope you got a chance to read the original topic :Oy:

I thought that was explained well in the first post concerning the 3p9s + 3p38s scenario above. I see many Lithium DIY building sites that offer battery components have fused plates to connect the batteries. I've seen other projects that us thin wire as fuses for the battery connection, much like what Tesla has used and is currently using in their packs. I've also seen magnets used in place of soldering the connections. Interesting stuff, even if it's not exactly the same situation as the interconnects between the OEM Nimh and the LFP pack.

I love the way the graphics work on your site. Plus the url is easy to remember even though I've bookmarked it too. I'd be mildly surprised if you've never been to 99mpg.com That's the guy who built the Grid Charger I use and collaborated with 2 others while developing the software to control it.

10 grands a lot to loose on battery testing. Staying positive about a project after those kinds of loses can be difficult, to say the least. I've had my share of setback too. So I relate.

I'm really glad I don't have to rely on my old hybrid all that much or it would be in lots worse condition than it is at 190k miles. And I still learn new stuff about how the inner working of the proprietary system functions, mostly on my own, since there are not many owners of the model I have, that are into babysitting for 10 years olds instead of junking 10 years ago..

 

What brand? What design?

Telsa is fitting LiFePO4's into their Model 3's with amperages apparently up to around 500 amps at about 3.5C (140Ah) and supposedly can charge faster on a quick charger than the Li-ion batteries in their other cars. These are large prismatic cells and are only cooled on the bottom instead of the cooling that weaves in between cylidrical cells like with their Li-ion batteries.

Headway makes LiFePO4's rated for 15C (120A) continuous and 30C (240A) intermitent.

My AA NiMH batteries can't be charged and discharged anywhere near the C rate of what the NiMH traction battery in my Prius can. But that doesn't mean a Prius battery can't do what it does. There's a similar situation between deep cycle and starting lead acid batteries. Unless you test every LiFePO4 battery out there, you can't just stereotype them all into one lump.

 

Best check the C rating as to which direction the current is travelling. Yet to see an LFP high discharge C rated cell also have a high charge rate .....well unless pulse charging is involved, then they really can handle a high charge rate, but that would be useless for regen braking without a battery built from cells that can take that high current charging as an interface between the regen current and the LFP battery. Attempting to charge above 1CA will result in rapid cell voltage rise and if the current isn't reduced, serious cell over heating will occur.
LTO chemistry is far better for the high rate discharging and recharging, quality cells can handle 30CA in each direction. That means the 55Ah LTO cells I bought for another project could handle the full regen current as well as the load current and still return an outstanding cycle life .... finding the space to fit them is the challenge

T1Terry

 

My AA NiMH batteries can't be charged and discharged anywhere near the C rate of what the NiMH traction battery in my Prius can. But that doesn't mean a Prius battery can't do what it does. There's a similar situation between deep cycle and starting lead acid batteries. Unless you test every LiFePO4 battery out there, you can't just stereotype them all into one lump.

The reason the Prius NiMh cells/modules can handle the high charge and discharge rates is they are wet cells. The electrolyte actually separates into hydrogen and oxygen but the internal pressure can be so high that very little is lost, but some is lost and that iswhen the cells/modules dry out and loose their capacity. Re-hydrating the modules can really improve their capacity.

T1 Terry

 

So you know more than Tesla does?

The Headway cells have a charge rate of 10C (80amps) and discharge of 15C (120A) But apparently you need to double them to be close to the weight of the Prius NiMH battery, so 160A charge is possible in a 2p configuration.

The surface temp will rise 25°C when charged/discharged at 10C continously between 0% and 100%. There is a maximum temp of 60°C, so without cooling you could full charge them from 0 to 100% (or discharge them from 100% to 0%) if temperatures begin at 35°C (95°F) or less. You would need to cut back charging and discharge after that (well, discharging can tolerate 75°C according to Headway). But of course you would rarely charge or discharge them between 0 and 100% at 10C the whole time (160A if in 2P) without any cooling. This is going off of Headway's specs.

The Nexcell LiFePO4 cells used for Prius conversions are said to be even higher C rated than Headway's. And seeing how the Prius's own temperature sensors are kept in there and there hasn't been any reports I've heard of of any type of overheating problems such as thermal throttling by the Prius, or battery damage, there just doesn't seem to be any reason to believe they're having the issues you describe.

 

Do you have a link to these specs and the relevant graphs from an independent lab, they sound very interesting is they are real and not just product listing B/S on a Chinese site.

T1Terry

 

I found this link, Headway 38120HP 8Ah LiFePO4 Battery Cell but the specs don't match your claims regarding the temperature the cell must maintain while being hi ratecharged.

 

Why don't they? In "Charge and discharge cell surface temperature rise" it links 10C continuous as creating at most a 25°C increase of heat. Maximum temperature for charging is 60°C, although ideally it should be 0°C to 45°C. Take away 25°C from those high numbers and you get 35°C and 20°C ambient respectively.

But again, that's not including any cooling into the equasion, as the Prius does indeed have a battery cooling fan to compensate. Again, that's not including the usually short, intermittent charactersitics of a hybrid drive train that relies mostly on the ICE so rarely, if ever, will be running at 10C contiuously until full or empty. Again, that's not including the Prius's own temperature sensing that both increases the fan and throttles back current to try to keep the battery from exceeding 50°C, which is only 5°C warmer than Headway's charging recommendation and 15°C cooler than their max temperature rating. Again, at 5 LFP cells per bank, the Prius is not going to let them go above about 3.7V, so if they do as you say they do and they have all this internal resistance that increases their voltage with current, the Prius is going to keep current cut back so that they never go over that 3.7V at most (more like 3.5V in reality). And again, that's not including whatever Nexcell's specs are, which have a 40C rating, which is much higher than the Headway cells.

 

Sounds like you know what your talking about concerning the LFP headways. I was wondering where you're getting the prius static charge discharge limit specs from, since it's the first time I've seen them referenced here at PC. Have you or someone you know already done the 5p / block LFP conversion?

 

I look up information others have done such as: https://www.energy.gov/sites/prod/files/2014/02/f8/batterygenIIIprius0462.pdf
You can see that they get maximum and minimum voltages during regen and acceleration after several tests on different batteries (a new NiMH battery and an old one that's on it's last leg). I have also done some of my own testing with the Torque app and haven't gotten quite those numbers (max and min) so they seem correct as an absolute worse case scenario, which so happens to fit into the specs of a LiFePO4 tolerances when put into blocks of 5 cells.

Note that on that paper the maximum battery voltage in a worse case scenario is 258.1V, which divided into 14 blocks and 5 cells per block (70 cells) is 3.7V. Also on that paper the minimum battery voltage in a worse case scenario is 174.7V, or on 5 blocks per cell works out to about 2.5V per cell. Of course this is at full current, not resting voltages, and in other experiments the voltage would hit 198.9V, or about 2.8V per LiFePO4 cell.

Also in that paper, note the graph on page 8 and how the Prius will hit 150A only momentarily, usually staying within 100 to 125A (blue line) depending on resistance and voltage, and in a high resistance battery will stay around 60A. So the Prius will adjust the amperage based on the voltages as the battery charges and discharges. Fitting this into a LiFePO4 battery with 5 cells per block, it needs to put out up to 60A at 2.5V acceleration and 3.7V regen at most, which fit into the limits of even a single 8Ah Headway cell with a 15C (120A) max discharge and 10C (80A) max charge.

The OP @jacktheripper has been experimenting with LiFePO4 batteries in all sorts of Toyota hybrids as a NiMH replacement for the past several years. He even sells kits on https://projectlithium.com/. So far no one has come forward, to my knowledge, with any major issues. These are rated for 260A (40C) and so far seem to outperform the NiMH batteries. The Prius is able to drive them up to 150A without going out of the 2.5V to 3.7V range, if that.

There have others who have independently done NiMH to LiFePO4 conversions in blocks of 5 even with other brands of cells such as this guy with a YouTube channel called "Lil Asian Grocery" Lil Asian Grocery LLC - YouTube. I don't know if it continued to work out for him, but the conversion did seem to work as intended in his few videos.

 

The latest lithium packs are supposedly 7000mAh and with a max of 280Ah. (Yes I subscribe to their newsletter). Eager to hear how those work out...

REVVL V+ 5G ?

 

Ummm.... not sure what specs you are reading, but the 10CA charging requires the cell to be above 10*C, the 10CA discharge the cell must be below 25*C.
If you scroll down further you can see the graph for charging and discharging @ 1CA, reference the charging voltage, 3.67v, then the discharging voltage, the majority of the capacity being drawn is at 2.57v. This is at only 1CA or 7 amps.
Maybe you can get a few for sample cells testing and watch the voltage when charging at different current levels, then the voltage at different discharge levels. The voltage parameters are 2.5v to 3.65v, you can't go outside these voltages and the cell temp must remain within the acceptable ranges for the charging and discharging tests.

I think you might have read into the specs, what you wanted to see, not what the specs are actually showing you.

T1 Terry

 

You didn't read it right. It doesn't say anywhere that the cells have to be at 25°C or below to be used at 10C. Read it again! It says temperature will rise no more than 25°C at 10C continuous use. That's a huge difference in meaning.

Also knowing the maximum and minimum voltages on the graph doesn't say much. If you'd look at the graph more closely you'll see that the voltage hovers for the most part around 3.4V when charging and 3.2V when discharging, both at about 1.1C (8800mA) until it gets very close to either 100% or 0% capacity, which is to be expected. That's barely breaking away from nominal voltage. At 10C of course there would be a greater difference of voltage as they stray farther away from nominal. But it's quite obvious that they can be discharged and charged at a much higher rate than just 1C just from looking at the graph, which blows your theory that LiFePO4's can't do more than 1C completely out of the water.

Why does Tesla say their LiFePO4 cars charge faster (at a higher C rate) than their Lithium Ion cars do? Why does Headway rate their cells at 15C and Nexcell at 40C if supposedly they can't physically put out that much current, especially when a graph clearly indicates a very small increase of voltage for every 1C increase in current?

 

These are the types of arguments that a lot of priuschat members like to read and pick sides.
I've gotten valuable info from both sides using patients while weeding through the argument.
But to say I understand why there is an argument about battery rate specs and build configurations of different types and completely different applications, is beyond my understanding. Isn't it complicated enough, with all the different types of batteries with manufactures claimed specs that are sometimes right on and other times way off. Probably a lot like this post of mine.