BYD BMS Hackability

[Scrappyjoe]

Hello,

After a long, long time missing out on the DIY EV party because my country doesn't have any, the first BYDs are arriving in our salvage yards. I have found a 2025 BYD Sealion 6 PHEV at a salvage yard with a perfect pack for my application.

There is very little information online about these PHEV packs, or BYD packs in general, and I assumed I would have to replace its BMS. I have an Orion 2 waiting on my shelf so I figured I could use that.

But then I found a YouTube video of a BYD Shark PHEV battery here -



- and, while the pack looks really well constructed and tightly put together, it also looks like getting cell taps to the terminals will be virtually impossible without basically grinding the cells off of the metal. I was hoping that maybe the BYD cell taps would route to the BMS section of the battery but it seems that this isn't the case, and that cell balancing is done locally by daughter boards that are impossible to get access to.

So, do I have any other options here? Does anyone know if the BYD BMS is amenable to hacking? Has anyone torn one of these down and replaced the BMS with a different one? Is there any chance at all that I will be able to avoid cutting the cells out of the pack if I want to use the battery?

My best case scenario would be that someone is willing to take on getting these BYD BMSes to work while outside of their cars, so that they can be integrated into the Zombieverter ecosystem, but I don't know if there's anyone willing to take on that work.

Any guidance, links, resources or further discussion welcomed. I have an acquaintance with a BYD Seal if that helps. I could perhaps persuade him to let me capture some CAN logs if it can be done in an obviously non invasive way.

[Scrappyjoe]

Here is a video of a BYD Atto battery dismantling, which shows it has significantly more accessibility to the terminals. So, not all hope is lost. But, if I was a betting man, I'd say the Sealion PHEV is more likely to look like the Shark pack than the Atto pack. I guess I should bite the bullet and see.

[Scrappyjoe]

After various searches on the wiki, the forum, and the internet, I came across the Battery-Emulator project by dalathegreat on GitHub.

https://github.com/dalathegreat/Battery-Emulator

This is a very interesting project that aims to make it possible to use salvaged various EV batteries with solar inverters from various manufacturers. From what I can gather it provides a hardware emulation layer that enables the two systems to talk to each other.

Scrolling through the issues, I discovered this issue which was just added last week, which adds experimental Zombieverter support.

https://github.com/dalathegreat/Battery ... /pull/1917

From what I can gather, what it does is it basically creates a Nissan Leaf BMS 'inverter target'. So you can take any supported EV battery, and pass it through the hardware emulator to spit out Nissan Leaf BMS can messages. The Zombieverter, in turn, can understand these.

My battery is not supported, but I'll join the discord server and see if I can get it supported, and then I'll participate in the experiment to hook it up with the Zombie

[Scrappyjoe]

Still waiting for my Sealion 6 battery to arrive; in the meantime I went to the BYD dealership to try obtain specs for the battery. The techs there couldn't give me anything - not nominal battery voltage, number of cells, nothing. Prospects are good for junkyards full of vehicles nobody knows what to do with in 5 years.

Anyway, I crawled under the car to get some rough measurements. I didn't have a tape measure so I just used my arm span and then checked with a tape measure when I got home - so these could be off by up to 10cm.

BYD Sealion 6 2025 AWD - 18.3kwh pack

Width 70cm
Length 115cm
Height 10cm

The pack has a foam sheet about 2cm thick between the top of the pack and the floor of the car, and underneath it has a plastic looking cover. The whole underside had a slightly sticky oily feel - perhaps anti rust. It looked pretty easy to remove. The bolt holes were easy to see.

For actual pack specs, I will have to fall back on some guesswork.

Google tells me via https://www.carexpert.com.au/byd/sealio ... erformance that the nominal voltage is 384V, pack size is 18.3kwh, and chemistry is LFP.

A single LFP blade nominal voltage is 3.2V, so that implies 120 cells in series. Since the voltage range of LFP goes from ~2.5V to ~3.65V, this implies a voltage range of 300v to 438v.

This is not good for me, as I am using an Outlander stack, which maxes out at 400v. If I do use the pack, each cell will only be able to be charged to ~3.33v. This looks to be the voltage at about 90% SOC or so. A bit of a pity to lose that last 1.8kwh but I suppose I will extend the pack life significantly.

[johu]

The Outlander charger will go above 400V (not sure how much)
It's only the Outlander DC/DC that is limited to 399V

So at least the charge end voltage can be 3.4V or so and once the cells settle they should be low enough for the DC/DC to come back on

[Scrappyjoe]

Thanks for the clarification. I was actually thinking primarily of the Outlander rear drive unit inverter - I seem to recall it only goes as high as 400v or so.

[arber333]

I had some twitching from rear inverter when using zombie vcu from oem commands in regen with battery at 400vdc.
but power stage is using 600V igbts so it should be ok up to 500v if using Openinverter brain.