LiBSU - Redesigning the "Battery Support Unit" to Support Lithium

Yeah, the "Gen3 Prius" equivalent in Lexus world - RX450h and the 2012-2019 Highlander, for example - P/N 89892-48040 has even *less* real estate than 89892-47010/47020 (that is currently being designed around) does.

All told: I would not expect LiBSU to *ever* cover 8-cell-style NiMH. There is not enough market, and not enough stuff in common. Or, rather, a future might exist where the guts of the LiBSU product boogy their way over to the "Gen3 Prius PHY" equivalent in 240-cell land, but I would not expect to find the LiBSU brand on them.

 

Thanks for breaking it down for us laymen.

 

I can't believe you've found one of these in the UK! I've been looking for one for years... If you don't have a use for it - would you be willing to sell it to me?

I've got 2 Gen2's both quite sick... one I'm hoping Mudder's BMS will enable me to get Sodium working safely, and the other I dream of making a PHEV...

 

Hey @mudder I know you've said it'll be ready when it's ready, and I respect that, but my battery just went kaput and I'm trying to decide whether to wait for LiBSU or just replace it with NiMH for now. I'm not in a huge rush since I have another vehicle (just way less efficient), so I'm genuinely open to waiting if the timeline makes sense.

So my question is: if you had to assign a rough percentage to how complete the project is right now, what would you say?

Also thank you for taking the time to work on this. I've been wanting to go the lithium route for a while, but everything I read made it clear it wasn't safe just yet. Mad respect for your work!

 

I've been compiling info regarding the 2012-2015 PHV since a friend got a 2014 PHV that was pretty much stuck with bad Brick C & D. (Has four 14 cell Bricks termed per Techstream). 56 cells total. Here is the link. Prius Plug In P1A00, P1A61, P1A64, P1A67, P301A, P314A, P31AB | PriusChat

I envision using Prime cells in the future in one of the two HV packs that literally we just picked up in Chattanooga last weekend. Go figure.

However, for my gen 2's and possibly the gen 3, would be nice to have more cost effective options to use those Panasonic Lithium NMC prismatic cells and/or a range of other cells.

I am definitely willing to test as well if you need and I do have enough NMC cells, though still need to compress the bricks, drill out the rivets and determine if going to try to fit nuts welded on the end plates or use rivets again. Debating getting a PRIME pack next month just to have while they're seemingly cheap and I see some newer ones around locally. The 2017-2022 come with 96 cells FYI.

Looking forward to reading more and updates.

 

I've joined Prius chat, first post. Fellow owner of the 2000 Insight(qty 2).
Here's some data I gathered from my 2007 Prius about how the car treats the battery and what voltage tolerances it works within.

Background: This is data from a sick battery pack back in October 2025 when things were colder as well as from the pack when it was rebuilt with 9 modules from a junkyard pack(reinstalled 6/3/2026) where all of the replacements were better when performing both load testing and capacity than every single existing cell in the Prius. ..but the rest of the junkyard pack had problem cells within the module where they would all drop to 6.11v or less after 1Ah of discharge at only 10 amps. Worst module in the *current* pack is 2902mAh measured at a 10 amp discharge to a 5.4v termination so in my opinion it's still a rough pack but Dr. Prius thinks having less than a 3Ah pack means "Battery pack has esimated 92.89% capacity left, it is in new condition." ...IMHO, a bit of a junk tool that's a cludge to use but I imagine it's consistent enough for that one function where it has you aim for approx 6.5 amps from full and waits until the car fires up at what the car thinks is 40% SOC. My pack rebuild targeted trying to get the best IR modules to replace the worst IR modules("IR" in my case was determined by comparing voltage under load after pulling 1Ah and 2Ah out of the modules) and not care about capacity which is also what I did when rebuilding Insight packs, it seems to have brought a bunch of performance back to the Prius.

My 2nd gen Prius data(per Dr. Prius, limited testing of only 3 different days(6/3/2026 and 2 days in October 2025):

Lowest block voltage - 12.41v at 42% SOC drawing 104.1 amps with the temp sensors reading 93.9-98.8f, whole pack was 183.49v at the moment. That block has the smallest 2902mAh module(measured at 10 amps to 5.4v termination when rebuilding it), so it was hitting empty hard. ..and then it continued to discharge at a slightly lower amperage of 91.35 amps at 183.94v pack voltage with a 41% SOC where with the reduced current that block showed 12.94v. Battery thermistor temps 93.9-98.9v.

Lowest pack voltage
178.94v 121.79 amps 42.5% SOC, 93.9-98.8f. 6/3/2026.

Highest pack and block voltage(same Dr. Prius screenshot)
Highest pack voltage - 286.33v (/14=20.45v or 10.225v per module)
Highest block voltage - 20.70v while the lowest was 20.10v
This was at 59% SOC 53.94 amps, 52.6-53.9f in October 2025.

Highest discharge amperage
190.31v 143.93 amps 52.5% SOC 115.3-119.8f

Highest charge amperage
261.54v 94.37 amps 60% SOC 89.4-95.6f

Conclusions based on the data:
The 2nd gen Prius seems like it tries to keep the lowest pack voltage above 180v although it overshot to 178.94v.
The car is okay with pushing to 286.33v during regen when the pack is lukecold, but I don't know what the software limits are.
It let's the 2 module block level get down to 12.41v under a 104.1 amp load at 42% SOC, it seems to allow you to pound the pack, even near its 40% SOC cutoff.
Apparently 119.8f still allows 143.93 amps discharge, which surprised me. I also saw 66.16 amps regen with the hottest thermistor at 121.f. I was cycling the pack pretty hard multiple times after putting the pack I rebuilt back in the car, so that's why it was hot.
Even though this pack is not an awesome collection of modules and could be much better if I tore into a second junkyard pack during the rebuild, it seems like the 2nd Gen Prius will push more voltage per cell than the 1st gen Insight will and discharges with a slightly more conservative per cell value and the amperage it pounds the pack with in both directions is also impressive even at its software defined SOC extremes, these are very powerful NiMh packs.

No idea how the thresholds have changed between the 2nd and 3rd gen Prius though. I'm also willing to gather additional data.

Another thing: I have the 2nd Gen 2008 Prius BSU from the junkyard pack, if you don't have one, do you want me to mail it to you? It seems my Youtube comment got deleted from your recent RMA Youtube video. I assume these parts aren't a match between Gen 2 and 3? I know you are working on the Gen 3 first, but if having a 2nd Gen Prius BSU earlier helps, I'm happy to send it over.