The Harley Bike Battery and BMS

Been down a bit of a rabbit hole lately thinking about repurposing old Harley Davidson bike batteries for small off-grid builds 🤔

From what I've gathered the OEM Harley AGM units are decent quality but fairly small capacity — fine if you're just running a few 12v bits on a narrowboat or van but not exactly a powerhouse. The lithium ones from newer models are more interesting tbh.

Main thing I keep coming back to is the BMS situation. The factory BMS on these is locked pretty tight to the bike's charging system. Anyone actually tried pulling one and running it with say a Victron MPPT? Curious whether the BMS plays nicely or just throws a fit.

I've seen a few YouTube builds (mostly American) where lads have bypassed the OEM BMS entirely and slapped in an external one — JBD units seem popular for that. Not sure I'd trust that approach without proper cell testing first though.

Also worth noting — sourcing replacement cells if anything goes wrong could be a headache. At least with a Fogstar or similar pack you've got some backup.

Few questions I'd love to get some thoughts on:

• Has anyone actually bench tested a Harley lithium pack here in the UK?
• What capacity are we realistically looking at?
• Is the bypass BMS route sensible or asking for trouble?

Probably more faff than it's worth for most builds but the price of used ones on eBay has caught my eye lately 😅 Could be a nice cheap starting point for something small.

@IvyLes the AGM angle is interesting but I'd steer clear honestly. Motorcycle AGMs just don't have the cycle life for daily off-grid use — they're designed for float/standby duty really, not deep cycling. You'll hammer them flat in a season.

If you're after cheap repurposed batteries, lithium forklift or EV cells are a much better shout. Fogstar do decent refurbed cells and at least you're getting proper cycle counts out of them.

The BMS side of things with motorcycle batteries is also a headache — AGMs don't have onboard BMS so you'd need to manage charge/discharge yourself, which adds complexity for not much gain.

What's the actual build size you're planning? That'd help narrow down whether repurposed cells even make sense vs just buying new LiFePO4 outright.

@IvyLes worth distinguishing which Harley batteries you're looking at. The OEM Yuasa/Deka AGMs are decent build quality but rated for shallow cycling — primarily cranking batteries, so expect maybe 100-200 deep cycles before capacity tanks noticeably.

If you're after motorcycle-adjacent chemistry, the lithium units (Antigravity, Shorai etc.) are LiFePO4 and actually more interesting for small builds — proper cycle life, lightweight. Problem is the integrated BMS is sized purely for cranking loads, not sustained discharge management.

For a small emergency backup I'd honestly just grab a Fogstar Drift cell or similar and build proper. Less faff than reverse-engineering proprietary BMS logic on a bike battery.

What's the actual use case — trickle load or something with meaningful draw? That changes the answer significantly.

@IvyLes had a similar thought a while back when a mate offered me a pulled Harley Lithium (the newer Screamin' Eagle units) for next to nothing. Ended up passing on it because the BMS is completely proprietary and locked to Harley's own charging profiles — getting it to play nicely with a Victron MPPT was more hassle than it was worth.

The AGM point @LindaJones69 raises stands too, but I think the lithium ones are actually the more tempting trap. Decent cells underneath but the battery management is basically a black box.

For a garden office or small off-grid setup you're genuinely better off with a proper leisure cell from the start — even a budget Fogstar Drift works out cheaper once you factor in the integration headaches.

Good shout from @IslandExplorer on the Screamin' Eagle units — the lithium angle is far more interesting than the AGMs for this application.

Key thing nobody's mentioned: those factory Harley lithium packs use proprietary BMS communication that makes integration genuinely awkward. You'll likely need to bypass or replace the BMS entirely, which introduces real unknowns around cell balancing tolerances.

For my van build I went down a similar rabbit hole with salvage EV cells before eventually giving up and going Fogstar Drift LiFePO4. The time cost of integrating mystery BMS hardware just wasn't worth it.

If you're committed to the Harley route, I'd strongly recommend:

• Proper cell-level voltage testing before assuming pack health
• A known Victron-compatible BMS replacement
• Conservative C-rate assumptions until you've logged several cycles

For emergency backup specifically, the risk profile seems unnecessarily high when new LiFePO4 is so affordable now.