Connection reliability in outdoor battery systems – what actually fails over time?

[eweichat]

Hi all,

I’ve been reading through a lot of threads here and trying to better understand long-term reliability in battery systems, especially for outdoor or semi-exposed setups.

Most discussions seem to focus on cells, BMS, and thermal management, but I’m starting to notice that connection points might be just as critical over time.

In a small setup I’ve been experimenting with, everything works fine initially, but after some time (humidity + temperature cycles), I start seeing occasional instability — voltage drops, intermittent cut-outs, things like that. It’s not clearly a cell or BMS issue, which makes me suspect connection degradation.

I’m trying to understand what typically causes failures in real-world battery systems:

• Is vibration a major factor in loosening or degrading connections?
• How much does thermal expansion/contraction affect long-term contact reliability?
• Do oxidation or moisture ingress tend to be the bigger issue?

Also curious how people here approach connection design:

• Do you prefer bolted connections, crimped terminals, or plug-type connectors?
• Any best practices for outdoor or high-humidity environments?
• Is it more about mechanical stability, or sealing against the environment?

I’m not working on anything large-scale — just trying to better understand what actually fails first in real-world conditions and how to design around it.

Would really appreciate hearing your experience or any failure cases you’ve seen.


Thanks!

 

[DogDipstick]

It depends.

On the type of construction you are looking at.. for instance.. many 18650 or 21700 cells have steel cases.. a pouch cell is totally non ferrous...

galvanic corrosion is your biggest long term enemy when it comes to electrical contact and connector troubles.

Rust and corrosion second. .... use something like Penetrox A or AB. I also insist on ONLY nickel plated copper everywhere ( from the cell tab to the controller)... and stainless bolts.. dipped in Pentrox...

NEVER NEVER NEVER link NEG pole to ANY part of ANY case or structure of the battery downwind. Fast way to get your cell eaten.

Some cells are good.. some have a reputation for failure.

Going on the seventh year my cells have been in exsistance.


Here read this.

https://www.nasa.gov/wp-content/uploads/2024/01/battery-failure-databank.pdf?emrc=69ba7fcbdee41

Hell here is a list years ago I put up as reading material.

Li-ion Pouch Cell Designs;Performance and Issues for Crewed Vehicle Applications

Li-Ion Pouch Cell Designs; Performance and Issues for Crewed Vehicle Applications - NASA Technical Reports Server (NTRS)

Are Soft Short Tests Good Indicators of Internal Li-ion Cell Defects?

Are Soft Short Tests Good Indicators of Internal Li-ion Cell Defects? - NASA Technical Reports Server (NTRS)

Lithium-ion Battery Safety Issues for Electric and Plug-in Hybrid Vehicles

https://www.nhtsa.gov/sites/nhtsa.d...848-lithiumionsafetyhybrids_101217-v3-tag.pdf

Characterization of Commercial Li-ion Cells in Pouch Format

Characterization of Commercial Li-ion Cells in Pouch Format - NASA Technical Reports Server (NTRS)

Safety and Long-Term Performance of Lithium-ion Pouch Cells

Safety and Long-Term Performance of Lithium-ion Pouch Cells - NASA Technical Reports Server (NTRS)

Tolerance of Li-ion Pouch Cells to Varied Space Environment Pressures

Tolerance of Li-Ion Pouch Cells to Varied Space Environments Pressures - NASA Technical Reports Server (NTRS)

 

[AnthonyC]

> Do you prefer bolted connections, crimped terminals, or plug-type connectors

No experience with storage batteries or solar but...
I'd minimise the junctions in the interface as much as disconnections allow (frequent / rare / never). Example for signal and low power:

A plug might use ( flex --- crimp spade / solder spade --- terminal tag --- mating contact ) and the reverse on the other gender.
Specifying a different plug can reduce it to ( flex --- screw / solder terminal --- mating contact ) and the reverse.
Bolted connection ( flex -- solder / crimp -- ring ) and the reverse, with a higher contact pressure.
Butted connection ( flex -- screw / solder / crimp ) and the reverse, with no intervening contact.
Is direct connection possible? (subsystem flex connection point --- flex --- subsystem flex connection point )

The protection requirements vary for each junction type and in different environments so their reliabilities are hard to determine, hence reducing junction count as a lazy optimisation.

For e-bikes I find any junction needs sleeving (or ties) to get graduated stiffness to cope with vibration; solder wicks and needs longer sleeving; screw terminals don't survive vibration well with or without ferrules. After stability I tackle weather sealing, this has to be flawless or it's counterproductive. In more corrosive conditions there's no substitute for maintenance: cleaning and grease renewal. Aside from flex termination any good connection joins similar metals by design so I don't find thermal expansion an issue. My 20A batteries use crimped ring terminals, glue filled heat shrink plus hot glue touch-up. I tried sealing 'waterproof' connector terminations with hot glue but that gets expensive if you need to rework.