'06 Prius battery upgrade to LTO 55Ah cells

If I'd been filming the pack strip down it certainly would have, with a lot blips in the commentary ..... A number of the 5mm stainless nuts seized on the threaded rod, but the first one was the most exciting. I locked two nuts together and had an battery drill spinning the rod to get it out of all the threaded rivnuts and 5mm stainless nuts. Worked on about 4 of the lengths of rod so I was getting a little too confident about the whole process. Then the nut seized, the 5mm rod buckled and shorted between 4 cells ..... lots or sparks etc. Thankfully it was only two rows of two cells in series and there weren't any links between the series strings to build a battery.
Grabbed the angle grinder and cut the threaded rod between the cell groups, that reduced it to each 2.4v cell shorting to the one beside it. They got hot and one cell actually vented its electrolyte, so that was another moment the sphincter muscles went into over drive.
In the end, 4 cells down for a cost of around AUD$300.

Plan for reassemble is to add plastic hose over the threaded rod as it travels past along the cell case ...... if it had been at the full 84 cells in series stage it would have been a tad more dramatic and it was already dramatic enough ....

Actual work is getting in the way again, so it all goes on the back burner till I get some more time to devote to the project .... maybe the "no working week ends will resume so I'll have tomorrow and Sunday to play with the battery pack.

T1 Terry

 

At least I discovered how the cells vent. There is an aluminium cap with a thin centre section that fails under pressure or I'm guessing, heat.
Also discovered the electrolyte reacts with the plastic wrap on the cells and the plastic racking cradles resulting in the two melting together ... made it kinda interesting getting the cells out of the plastic bits, but a bit of similar coloured paint can to the rescue so I could reuse the cells that were still undamaged, other than the plastic film missing at the ends.
I have built the number 2 pack, 3 cells high x 2 cells long, actually remembered to add the copper links between each cell so I had somewhere to attach the voltage sense wires and balancing wires.
As far as balancing, possibly build up 5 of these Zero Emission Vehicles Australia and use 5 of these https://zhcsolar.com/product/battery-balancer-192v/ to do the heavy lifting as far as balancing goes, the Zeva unit more for cell voltage/temp management and charger control.

T1 Terry

 

Photos:





Marks where the rod contacted the cells



Partly failed burst plug



Burst plug went bang





Pack 2 between end plates. You can see the copper link here, same on the other end and the links between the cells can be seen on the photo above .... well sort of .....

T1 Terry

 

Let's call it, checking the manufacturer isn't fudging the figures with both the cells and the balancers. I know the ZHC Solar balancers work, we have been using these on 24v and 48v systems for quite a while now, but only the 4 cell ones, didn't know they made an 8 cell and a 16 cell unit until about 12 mths ago, but all of those we have fitted have worked brilliantly.
I was a little concerned about the lower voltage the LTO cells use compared to the LFP and LYP cells, but the voltage range between fully charged and 0% SOC is very similar if not greater on the LTO cells, so here's hoping they will do the job.

A quick checl of my dodgy maths tells me it is 6 x 16 cell units I need, not 5. This will allow a cell overlap each side with the 4 mid pack units and a 2 cell overlap with unit 1 and unit 6.
The idea is to stop the charging if a cell reaches 2.85v and trigger a timer. The timer will continue to cycle until the high cell voltage is cleared, then the charging will resume until the pack reaches some where between 235vdc and 240vdc to allow enough cell voltage overhead for balancing.
The Prius regen should top out some where between 224v (8v per module) and 252v and that works out to 3v per LTO cell ...... but I very much double the regen will be enough to replace what will be used out of the battery and the ICE powered generation stops a bit lower than the 224vdc mark ...... but very careful logging of all that is going on will be the go while I learn how close my assumptions are to the reality ....

Part way through rebuilding Pack 1, this time with tabs to T off voltage sensing and balancing cables and plastic pipe over the threaded rod.
This time I'll be using pieces of 6mm ID aluminium tube as the spacers between the cell mounting blocks and I'm using garden dripper hose for the 5mm rod insulation .... each 300mm along the tube there is a plastic insert at the dripper hole and the 5mm rod cuts a nice thread into that holding it in place during assembly and it was cheaper than clear plastic hose .... I'll let you guess what made the choice initially :lol:

The copper links are what we use for the Winston 100Ah cell links when we build off grid house batteries so we have a few on hand .....

T1 Terry

 

More progress. I've rebuilt pack one that goes behind the back seat and in front of the rear strut towers. Pack one plus pack 2 (to pack that goes between the strut towers, weigh 51kg, 2 kg less than the original Prius Gen 2 traction pack .... but I've still got to build pack 3 yet and that will add another 38kg, 89kg where the original 53kg battery pack was mounted ..... and that is only half the total pack required to reach the same voltage as the original NiMh battery pack.

Probably around the 180kg mark all up, a little less than 3 times the weight of the NiMh battery, but roughly 9 times the capacity and hopefully more than 20 times the usable capacity, so it will be interesting to see what the trade off is between the added weight and stored energy that can be used as required ........

The original NiMh battery only seems to able to handle 150wh of regen before it's full and the regen cuts out, if I only use 1/4 of the LTO battery capacity getting to the top of the big hill, I should have storage capacity for around 2800wh of regen on the trip down the hill ..... I don't think over voltage from regen will be a problem

T1 Terry

 

I don't believe there is any reprogramming required because the voltages remain the same, but yet to see if the regen throws up some sort of errors.

As far as the project goes, work has been getting in the way of my playing battery builder, but I now have the 3 packs built for the part of the battery that will replace the the spot where the original traction pack lived.
I am now at the point of building a 4mm thick aluminium base plate to mount it solidly the the Prius battery mounting points. The next idea that has been doing laps in my head, is building a Lexen clear box to go over the cells. Two reasons, I can see the lights flick on or off on the cell modules and with my idea of cooling the cells using an E Fluid, I can see if any areas of the battery or cells start to boil the E Fluid indicating that part is getting hot. The other advantage of the E Fluid, no air contact on the cell terminals so hopefully no oxidisation issues ..... and it might look pretty flash as well :lol:
Lexan is strong enough to build aircraft windscreens so it should be plenty strong enough to meet any anti-intrusion/crash strength requirements they might throw at me.

T1 Terry

 

Been a while since I made much progress on this project, earning money for the wife seems to keep getting in the way of having fun .....
Finally finished the 42 cell pack that fits where the original traction battery lived .... only to discover I didn't allow for the walls of the battery box between the mounting point for the front 18 cells (3 cells long X 3 cells high X 2 cells wide) and the brace for the strut towers .....
A redesign with 9 cells (3 cells long X 3 cells high X 1 cell wide) to fit in that spot and 12 cells (2 cells long X 3 cells high X 2 cells wide) between the strut towers and the original design 18 cells (3 cells long X 3 cells high X 2 cells wide) behind that will reduce that pack to 39 cells, the rest will have to go in the remodelled spare wheel well.
This turns out to not be as big an issue as I first thought. I can place the cells longways across the vehicle and this will give me room to avoid moving the muffler. The top layer is 18 cells (2 cells long x 9 cells wide) 2nd layer is 16 cells (2 cells long x 8 cells wide) to allow clearance for the fuel filler pipes and breathers and the curve of the jacking point support (that will also support the battery box, it was originally welded to the spare wheel carrier) and the 3rd layer positioned to stay close to the rear axle and give a clean departure angle at the rear consisting of 12 cells (2 cells long x 6 cells wide) the whole pack sitting east west between the left hand chassis rail and the muffler ...... That makes a total count of 39 cells up front and 46 cell under the floor level = 85 cells, 1 cell too many .... I'll have to see where 1 less cell will make a better fitting battery pack.
On an Australian build that would be between the passenger side chassis rail and the muffler situated on the drivers side between where the spare wheel well was and the drivers side chassis rail.
I'm guessing they didn't change the floor pan between LHD and RHD models, so that would put the muffler on the passenger side and the battery pack up against the drivers side chassis rail.

T1 Terry

 

Hi Yuriy, Welcome to the forumGoogle came up with this as the translation

Hi all. I installed a homemade battery from lifepo4 70s 13ah cells. The Prius still thinks it has 6500mAh left (the capacity of a new battery). Without changing the can bus, we will not achieve a good effect on the use of high-capacity batteries.

I hope that is a rough translation of what you posted.

What do you see see when using the 13Ah cells compared to the original 6Ah NiMh modules? With a Gen2, the green on the battery display should stay green longer and the regen should take longer before it cuts out because the battery is fully charged.
The problem with using the LiFeP04 (LFP) cells is their slower acceptance rate of high current charging resulting in the voltage climbing faster than the actual capacity. Once the system senses the voltage is high, it will reduce or even cut the regen, but then the battery voltage should drop down to the actual charged voltage.
Do you have a scan type gauge connected to the OBD2 port to give you live readouts of what is actually being seen by the computer?
I hope this translates to Russian so you can understand what I'm asking ....

T1 Terry

 

Hi Yuriy, you do realise the 18.8v seen at some of the modules equals 3.76v if all the cells are perfectly balance .... which they wouldn't/couldn't be because of the high charge current on regen and the cells being so far above the fully charged voltage of 3.45v.
The 4v in some of the lithium manufactures specs is related to charging at 3C, this is to allow for the over voltage such a charge current will produce across the cell. In you case, using 13Ah cells, the 3C charge current is only 40 amps, not the 120 plus amps the cell can see on heavy regen, that is getting close to 10C and LFP cells just can't handle that sort of current. The voltage will be spiking long before the cell reaches even 50% capacity, so there is every chance you are only getting the cells charged to 50% capacity and that is why you are not seeing any improvement of the original NiMh battery, the cells just aren't getting fully charged each time.

Add to that, a 10C charge current will generate a lot of heat at the electrolyte level and cause electrolyte separation, resulting in the plates being coated with a film that will greatly increase the resistance in each cell because the lithium ions can not get in or out of each plate as quick as they could when the plates were not coated.

The result is actually sort of protecting the cells from over voltage charging damage to a degree, but it really screws up their ability to fast charge.

If you had used LTO chemistry cells, good quality LTO cells can handle a 10C charge and discharge rate, some speciality manufactured cells can handle up to 30C, but the price of those is crazy high.
If you look up OSN Power and contact Dinah, tell her T1 Terry from Australia said for you to deal with her, and find out how much Yinlong LTO cells would cost you, it might not be as bad as you think.
The cells you have will work great for a house battery, you can charge them during the day with solar and plug the car in when you get home to recharge the LTO battery ready for the next day.

T1 Terry

EDIT: Just to let any who read this know, I don't get any kick backs or otherwise from Dinah at OSN Power, she has always treated me fairly and gone out of her way to sort problems for me, so I feel I can trust her to do the same for those I recommend talking to her.