Issue dossier

Multiple badges failed during the event. Batteries inserted with reversed polarity destroyed the AMS1117-3.3 regulator, which the board does not protect against. Other badges failed in a way consistent with thermal runaway of the IP5305T boost and charger IC. The report could not reproduce that failure, but found inductor saturation, current peaks near 3.7 A during load cycling, about 11 V switching-node ringing with no snubber, 112.8 °C on the IC case, and FET degradation, and it says the power-bank IC is not designed for Wi-Fi burst loads. It found no proven short-term fix for assembled badges.

A linkable camp-impact record with badge context, severity, evidence confidence, and source notes.

Back to issues index

power-stage failures and thermal-runaway risk · maker preliminary failure analysis · preliminary; redesign recommended

Owners of surviving badges should check battery polarity before inserting a battery and should not leave the badge running unattended on battery power. The report's causes are hypotheses, not a confirmed root cause, and it gives no count of failed badges.

Badge
BSides Lisbon 2025 Electronic Badge
Category
power-stage failures and thermal-runaway risk
Severity
major
Confidence
maker preliminary failure analysis
Status
preliminary; redesign recommended
Timeframe
BSidesLisbon 2025 event to report revision 3 (2026-02-05)
Source note
Bsides 2025 Badge Preliminary Hardware Failure Analysis Report, revision 3, by posix, in kripthor/bsideslisbon2025-badge.

Evidence

Related Resources

badge firmware repository

kripthor/bsideslisbon2025-badge

GPL-3.0 repository created February 9, 2026 and titled Bsides Lisbon 2025 badge firmware and information. It holds the BsidesLisbon2025_TheEpochalypse Arduino sketch, its snake game, easter-egg and badge PKI headers, and the preliminary hardware failure report. It has no schematic, board files, BOM, or photos.

Badge: BSides Lisbon 2025 Electronic Badge

Source trail

Evidence Sources

BSidesLisbon · retrieved 2026-10-29

BSidesLisbon 2026 speakers: Don't Drawer Your Badge

Official 2026 programme source for the ESP32-C6 identification. The abstract says the previous BSides conference badge is powered by the ESP32-C6 and describes mapping its pinout without documentation, finding the display driver, and working around hardware power constraints. It names the badge as the previous one rather than by year.

Badge: BSides Lisbon 2025 Electronic Badge

kripthor / GitHub · retrieved 2026-10-29

BsidesLisbon2025_TheEpochalypse

Primary firmware source published by kripthor, whom the official BSidesLisbon 2026 home page credits with developing the conference badge once again. It proves the ST7735 160x128 display driver, the six button pins, the keep-alive pin, ESP-NOW beacons and the RSSI badge grid, the nine exploit and defense pairs, HTTPS over-the-air updates with version comparison, micro-ecc signature checks on master-badge commands, the Y2038 and Hackerman screens, and the idle timeouts. The sketch says version 2.53, so it is a later revision than the firmware 1.0 named in the failure report.

Badge: BSides Lisbon 2025 Electronic Badge

posix / kripthor GitHub repository · retrieved 2026-10-29

Bsides 2025 Badge Preliminary Hardware Failure Analysis Report, revision 3

Primary engineering source, retrieved as a 1,354,284-byte PDF (SHA-256 ac0234dd621303e6647b515609ddd1a885c7383ab8a750eddceec765f57d6a9e). It proves PCB revision 1.0, the U1 ESP MCU, U2 IP5305T, U3 AMS1117-3.3, L1, and D2 parts, and that multiple badges failed during the event. Reversed battery polarity damaged the AMS1117 and was reproduced on the bench. The IP5305T failures looked like thermal runaway but could not be reproduced; the measurements show inductor saturation at about 680 mA load, current peaks near 3.7 A, switching-node ringing near 11 V, 112.8 °C on the IC case at a pulsed 780 mA, and signs of FET degradation. The report labels its causes as hypotheses that need more testing.

Badge: BSides Lisbon 2025 Electronic Badge