Lithium replacement pack nearing market. Sounds great

Doing some math, it looks like you'd need 7 x LTO cells in series to fit within the range of the Prius charging system. Voltages over 18V do happen often as far as I can tell, and 6 cells would be only 16.8V. 7 cells at 2.8V max would be 19.6V. And at the lower end it would be 1.6V x 7 = 11.2, still within limits.

So 7 x 14 = 98 cells, at $35 a piece from Alibaba (not including shipping, taxes or tariffs) is nearly $3,500. And I'm going off of the 28Ah Yinlong LTO cells.

 

Ummm...... 19.6v x 14 = 274.4v and let me assure you that the original traction battery doesn't get even close to that before it explodes ..... speaking from rather expensive experience yesterday afternoon ........ Even at the 18v in each pair of modules through all 14 prs equals 252v, the computer said the battery wasn't at 100% SOC reading the scan gauge (82%) so I thought I'd give it a few more mins charging from the solar ..... it was only charging a few amps at this point ........
They go bang, flames shoot out the battery air vent and things get rather exciting because one good bang seems to deserve another ..... Wife panicked and called the local volunteer fire brigade ... and then the real circus started. Fortunately I already had most of it disassembled before they got here, I'd already put the fire out using the car's dry powder extinguisher so they weren't needed, but they thought different and now I have one hell of a mess to clean up after they emptied a big dry powder extinguisher in there as well, then complained they couldn't get a heat reading on the modules to see if they were cooling down or not ...... I sure hope this triggered a lot more training for these guys because they really didn't have a clue, didn't know how to trip the battery fuse or even where it was, they thought it was under the bonnet.

So, now it seems I'll be testing the LFP traction battery replacement first to determine just how well that will work and just how high the total and module group voltages can go before a fault is tripped.
My first attempt will be 5 x 14 cell groups, total of 70 cells. There are 76 x 40Ah cells in the extended range battery now so just a matter of extending wires and cables and bypassing 6 cells, then try driving direct from those as the traction battery ..... well after I get all this dry powder out of the car and the very enlarged traction battery out, tore the bolts out of the nylon end plates that hold the pack together it went off that big. I'll take some photos and upload them in the next few days. That steel cover over the battery pack ... well at least it funnelled the flames out through the gap and left a spot to aim the extinguisher into, it ballooned up big time.
One thing to note, do not use the scan gauge as the guide for a 100% SOC traction battery, it's not even close to accurate.

T1 Terry

 

I don't quite understand what you're getting at and also think you didn't quite understand me. And your max voltage numbers are off.

You had mentioned that 6 LTO cells per block would work. But they won't because the maximum voltage (as far as I understand) is 2.8V per cell, or 16.8V per block, or 235.2V for the entire battery. Prius regen can hit at least 285.1V and up to 1.54V max cell voltage with the original NiMH traction battery (referrence: 2010 Toyota Prius VIN 0462 Hybrid Electric Vehicle Battery Test Results (energy.gov)) This may be off from your numbers if you're slow charging the battery. From what I understand the Prius system measures amperage and resistance and adjusts voltage from there. So a fully charged battery will be lower voltage than one that's half charged under full regen amperage.

Sorry to hear about your battery. That's the original NiMH battery, right?

What brand of LFP's are you using? What do you think of the @jacktheripper 's LiFePO4 aftermarket battery?
Rainbow Prius Hybrid Powered by Lithium! | PriusChat

 

Just to clear up a few misconseptions. The 285v is measured at the terminals, not the total of the modules added together. These extreme voltages are the result of the chemistry not being able to absorb the high current, (internal resistance) LFP suffer the same problem but to a lesser degree when new, but high rate charging and discharging degrades that ability rapidly.
In an EV, the cycle life is each full charge to full discharge, in a hybrid, there are a lot of full to empty and returned every drive. In an EV the average for cycle life of a properly powered battery pack where the discharge rate never exceeds 3CA and more around the 1CA discharge rate throughout the discharge cycle, is around 2,000 cycles or 5.5 yrs, no idea how many cycles a hybrid battery would see on the average daily drive, but I can't see the same 2,000 cycles lasting for 5.5yrs and certainly not the 10 yrs the average NiMh traction battery seems to last .....

The NiMh traction battery that I destroyed was one of the latest generation apparently. I scored it from a vehicle dismantler (we call then wreckers) for $400 outright, no return carcass required. It was complete with the latest version battery computer etc and worked great, really pissed with myself for destroying it but ya gotta "break a few eggs to make a good omelette" .... or the saying went something like that .......

The LFP cells in my extended range battery are prismatic GBS 40Ah cells, I won't recommend them to anybody. They leak electrolyte vapour something fierce when recharging even at a slow rate (5 amps) and when discharged at 120 amps when topping up the traction pack each time. The voltage sag is pretty bad as well, under heavy acceleration the contactor will stay closed between the traction pack and the extended range pack because it won't flow enough current to equal the discharge rate of the traction battery because the voltage drops from 270vdc (fully charged) down to around 220vdc and it seems the inverter will take as much current as it can get at that voltage.
The LTO cell pack will sort that problem so I can see just how much current the inverter demands under full load because the voltage won't sag under high discharge rates.

The LTO cells can be charged to over 3v per cell without much cycle life loss and down to 1.5v per cell but that will never happen because the Prius computer won't allow the battery voltage to drop that far.
84 x 3v = 252v actually at the cells, not at the terminals, 84 x 2v per cell = 168v and the Prius computer won't allow it to go that low ..... well maybe if those test results in that report can be trusted.
90 cells would work better, but that can't be divided by 14.
7 cells per module makes a total of 98 cells and a near fully charged pack with 98 cells would add up to 280vdc and I'm not sure the Prius system could handle that voltage. The low end Prius cut out is around 180vdc looking at the scan gauge .... but again, that report claims 170vdc so .... again, that will only be known when the tests are done with cells that don't have voltage sag.

The proof will be actually doing the tests using each cell chemistry to get clearer results, the internal resistance of the cells puts all the figures off so the true capabilities of the Prius is an unknown until that problem is sorted.

T1 Terry

 

You seem to know a lot about battery tech so I'm surprised that you didn't seem to know that the Prius battery management system shuts off charging at around 80% and then shows the battery as full bars on the dash. I have a scan gauge and even down the longest hill charging will stop at 80%. So, it is purposeful to preserve the life of the battery, not inaccurate.

Did I misunderstand you or did you know that and decide to go beyond it anyway?

 

yeah... I have a hard time following everything he says due to so much missing information... A battery pack at 285 volts would put each 8volt module over 10volts which is impossible unless you're measuring the charge you're sending into the module that you're deliberately trying to destroy. And as you point out the ECU isn't going to allow more than 80% of full charge. Seems like whatever device/method the voltage is being measured is incorrect.

What's more NiMH don't catch on fire when they fail... They do smoke alot and can melt stuff and in very rare cases that melting can get so bad that it looks like a fire, but in general it's hard to make sense of what he's making up or not clearly explaining in a way I can understand.

 

Yes, I was trying to find what the Prius system considered was 100% SOC or even 99%, the scan gauge showed 82% so I mistakenly assumed I still 18% more I could put into the traction battery before it was at 100% SOC. I was looking for the battery voltage the Prius considered 100% SOC so I knew if it could still work with that voltage on the system.
The Prius uses 80% SOC to stop charging to allow for the fact their module balancing leaves a lot to be desired and even if it did work well, one module out of the pr monitored could be much higher than the other and I think hat is what happened.

I''ve seen 280v on the scan gauge a few times, but that was with the old failing traction battery, it didn't show that voltage for long so can the system handle that voltage long term?
I've let enough smoke out already, I'm not keen on letting the smoke out of the various computers and the inverter ....

T1 Terry

 

Not sure where you got that miss information from, but let me assure you, they most definitely catch fire. The flames coming out the fan vent were real and it took a number of attempts to keep the flames from reigniting.
It makes it difficult to answer the other queries you put up because you seem to have charging and voltage and SOC all mixed up.

It was suggested by Isaac roughly 8 posts back that the batteries often see over 18v per module and suggested that 7 x LTO cells would be a better option and I was just pointing out that dividing what ever voltage seen o a scan gauge by 14 modules doe not indicate the modules are actually at the voltage.
I needed to know what the traction pack voltage would be when the Prius considered it was charged to 100% SOC, the flames and smoke came out long before the 100% SOC was shown ...... So I still don't know the answer to that question

T1 Terry

 

Hey guys, I did some experiment before and LTO * 7 would work with proper balance circuit, this is confirmed.

My problem was quality LTO cell was too expensive to acquire and really not that scale-able. LTO is usually more expensive than LFP not to mention you need 7 instead of 5 in this case. Bad quality LTO would easily puff for no special reason (still work but puffy).

For me, the point of doing Lithium replacement is to beat Toyota NiMH price and able to deploy it at large scale with good quality control, it would work for you if you have time to search and money is not an issue.

hope this helps,
Jack

 

Thanks Jack. When you say "puffy" are you referring to pouch cells or cylindrical cells or prismatic cells?
If you are using 5 x LFP pouch cells then that explains how you did the 24hr @ 22v "torture test". I guess the purpose of this test was to ensure there was no fire etc and nothing to do with just how well that 5 cell group worked at the end of the test .... still impressive though all the same. I assume the 5 cell pack was already fully charged because it only took roughly 1 min to drop from the 5 amps to milli amps to maintain the 22v. 4.4v per cell across 4 cells and they didn't gas is impressive, but the internal resistance would be so high at the end of that torture test the cells would be virtually useless. Still it did show the containment vessel is strong enough to hold them compressed.
Not sure why you say 7 cells rather than 6 cells. 7 cells puts the fully charged voltage at roughly 20v per module group, so in reality they would never get fully charged. 6 cells fully charged is 17v, 17 x 14 = 238v, a tad short of the 240vdc to 250vdc but the cells can handle being over voltage charged with very little damage to cycle life.
The cycle life is the bit that separates LFP and LTO and a hybrid battery sees a lot of cycles over a short period, so longevity will be the telling point.
Anyway, I wish you all the luck in the world with your new venture, if you can come in under the OEM price and still get the cycle life the OEM battery achieves then you are on a winner.
After my traction battery went bang and caught fire I'm not so keen on having it positioned where it is, it took a bit to access it to ensure the fire was out and the remaining mess was actually cooling rather than still cooking away.

T1 Terry

 

@2k1Toaster is. Don't think anyone else is.

 

Well, I do it for a living actually working with LYP and LFP cells building batteries. Been at it for more than 10 yrs now and the only problems we've had with system we have installed is where the owner overrides the control system and murders cells by draining them below 0v ...... The early systems are over 10 yr old now and still hold more than 100% of the advertised capacity when tested at the C2 rate used by the manufacturer .... to me that says I know what I'm talking about. So maybe I fit into the developer category, never heard of a battery engineer as such, but maybe that is a USA thing, not necessarily recognised in the rest of the world as an official degree.
We have a writer over here that claims to be a "Research Engineer" there is no such thing, just a made up self appointed title.

T1 Terry

 

We have "sanitation engineers" too, it's hard to keep up with what I call title inflation. Just like all these CEOs of startups that are sole proprietorships. You're an owner, the owner. Not the CEO!

There are battery specialists and depending on what you focus in it's either chemistry, chemical engineering, electrical engineering, or a sprinkling of 2 or 3. That's on the design side. Applications side is different again. And can also be combinatorial.

 

:lol: Had to Google the last word to make sure it didn't belong in the "sanitation Engineer" category, but does actually appear, along with a string of highly unlikely words :lol:

I am in the process of negotiating with the minster of war and finance to allow me to spend the $$ to buy 180 x 55Ah LTO cells. If I have a win (not a certainty) I'll start a new thread regarding the build of the all in one traction battery and range extender battery. I'm aiming for a 160km plus range (100 mile) so the battery assist should still be strong even in hybrid mode yet have the battery capacity and the ability to accept the high regen current so I can use the regen all the way down the Adelaide hills.
It sould also allow me to set up for recharging from one of the high output DC charging stations that are appearing around the area.

T1 Terry

 

Then that makes 2. Didn't mean any disrespect. Lots of tinkerers, not a lot of engineers.

 

Looks like I might have had a win with the minister or war and finance (the wife) and I've contacted the factory in China to send an invoice to her business email .... one step closer

T1 Terry

 

After a few false starts, I think we might be actually underway. Can't believe the hurdles we had to jump to actually import from China .... seems the cooling of relations between Aust and China has made trading a lot more difficult. Probably could have do it as a retail purchase, but as a business in the lithium battery trade ...... free trade agreements and customs clearances to bring these "explosive devices" (that are safer than a mobile phone battery) have "lithium" in the description so all thrown in the one category as "Dangerous Goods" ......
So now we wait to see just how long it takes to get here by boat ..... far to dangerous to put on a plane
Mean while the battle with the insurance company continues, need a mechanical report, no one local has high voltage certification, I have an automotive mechanic license and the training and qualification to work on high voltage DC .... but I can't write my own report ....... So it appears I have to drag it 100kms to the big city to find someone who is qualified to say, yes, the battery is toast. The only problem there is, the dry powder chemical in the fire extinguisher the local volunteers pumped into the battery area, is highly corrosive, so all the electronics in there will now be stuffed as well ....

T1 Terry

 

Well, the money has arrived at the China factory, apparently they have a holiday week end so nothing will happen until Tuesday, then the nail biting bit continues with them reaching the boat, then reaching Adelaide wharf, then the tangled path through Australian Customs ......
Apparently there is a "Free trade agreement" in place that should cover this shipment, so that might at least save $500 or so.

T1 Terry

 

The cells arrived a while ago and today, the Prius finally went on the tilt tray truck to the repairers/assessors to see if I get an insurance claim and if the write it off and if I can buy it back at salvage price.
Conflicting advice regarding the maximum voltage the Prius will accept before the computer flags a fault. One said their Prius surged and carried on until the voltage dropped below 232vdc, another has a system that is running happily at 260vdc.
That puts my LTO battery anywhere from 84 cells and 98 cells,
6 cells x 2.65vdc (95% SOC) x 14 (module prs) = 84 cells and 223v and an upper window of 260vdc (3v per cell)
7 cells x 2.65v x 14 (module prs) = 98 cells and 259vdc @ 95% SOC and an upper window of 294v. That would give me a minimum voltage of 147v and it appears the Prius isn't happy below 180vdc or 1.83v per cell.
The 98 cell pack would give me roughly a 14kWh battery and replace the original 1.3kWh NiMh battery.
I can fit the 84 cells without that much modification, the 98 cells is another story ....

T1 Terry

 

I'd need to look it up again the maximum and minimum amounts of voltage the Prius will deal with and how.

There are three (maybe four) types of maximum voltage and at least two types of minimum voltages.
Maximum voltages:

1. Lower maximum. At rest (neither charging nor discharging) up to this voltage.
2. Too high at rest maximum. Prius will run fine but will try to burn off some voltage by spinning the engine when yo normally would have the engine off.
3. Charging maximum. This can be much higher than 1. or 2. but is allowed by the computer as long as the battery takes on enough current. As the current at these higher voltages sags it realizes the battery is filling up and it starts cutting voltage.
4. There may be a maximum voltage that causes a code to be thrown or even the car to be shut off or it just frys everything. This shouldn't be a problem if you don't charge the battery up more than voltages 1 through 3.

Minimum voltages should act similarly.