Kempower

Kempower — field brand hub

Kempower Satellite sites run a dynamic power pool: satellites (stalls/dispensers) draw from a shared power unit. Tickets that look like “one dead gun” are often pool assignment, satellite↔power-unit link, or shared coolant — not a failed SECC. Field tickets that ignore the link between satellite and power unit waste hours blaming the wrong box.

When this OEM shows up / cabinet vs dispenser

• Power unit: modules, pool allocation, site feed interface, often the OCPP endpoint for the cluster.
• Satellite / dispenser: user interface, cable, local sensors, RFID, session start path.
• Link: fiber/copper/comms between satellite and power unit — when this drops, the satellite looks “dead” or stuck Preparing even if the power unit is healthy and siblings still charge.

Confirm whether the fault is one satellite, all satellites on one power unit, or site-wide. That single question routes the job. Stall number on the pad is not always the OCPP connector ID — verify mapping before you swap hardware. Confirm which satellite ID, which pool, and whether sibling stalls still charge before you condemn a single stall’s power stage.

How faults typically present

HMI / Satellite UI

• Kempower Satellite / C-series UI codes (KEM-SAT-*).
• Stall Unavailable while adjacent satellites still Preparing/Charging → think link or allocation, not site-wide power death.
• Link-lost banners with healthy power-unit LEDs are the classic tell.

OCPP / ChargEye

• Faulted/Unavailable with vendor strings; mid-ramp abort may show as InternalError or vendor CCS code.
• Allocation stuck can look like SuspendedEVSE / 0 A without a hard Faulted — check load/pool before isolation megger (/entries/protocol-load-01-dynamic-load-management-caps-to-0-a).
• Use /guides/ocpp-faulted-triage when status is Faulted without a clear mechanical latch.
• Power unit Heartbeat OK while satellite Authorize fails (or reverse) — trace which node owns the OCPP connection for that connector ID.

Cloud

• ChargEye + OCPP beat LED-only. Pull pool assignment, coolant pressure, link status, and module online map for the satellite ID on the ticket.

Physical cues

• Damaged link cable after landscaping, loose terminations, water in junction boxes, glycol/pressure issues at the power-unit loop.

Ranked common failure patterns

1. Satellite link to power unit lost — Fiber/copper/link hardware, connector seating, or config after service. Sibling stalls healthy is the tell. Entry: /entries/kempower-kem-sat-link-satellite-link-to-power-unit-lost.

2. Dynamic power allocation stuck — Pool budget, priority map, offline modules, or LMS/failsafe after heartbeat loss. Correlate simultaneous high demand; check module health on the power unit before cable RMA. Entry: /entries/kempower-kem-sat-dyn-dynamic-power-allocation-stuck. Fleet/CMS angle: /guides/fleet-cms-triage.

3. Cabinet coolant pressure low — Shared loop: one low-pressure event can derate multiple satellites. Entry: /entries/kempower-kem-sat-cool-cabinet-coolant-pressure-low. Procedure: /guides/liquid-cooling-loops. Airlock after service: /entries/site-generic-cool-air-airlocked-loop-after-fill.

4. CCS session abort mid-ramp — Latch/cable/vehicle/SECC; prove with known-good cable and second vehicle before power-unit blame. Entry: /entries/kempower-kem-sat-ccs-ccs-session-abort-mid-ramp. Handshake: /guides/ccs-handshake-basics, /entries/protocol-ccs-h03-cable-check-isolation-fail, /entries/protocol-ccs-h01-ccs-slac-timeout.

5. RFID reader miswired after service — Harness reversed/loose after panel work; auth fails with healthy energy path. Entry: /entries/kempower-kem-sat-rfid-rfid-reader-miswired-after-service. Related: /entries/protocol-rfid-01-rfid-reader-no-uid, /entries/protocol-ocpp-f03-authorize-rejected-invalid-idtag.

6. Site power under pool ramp — Many satellites ramping together expose feeder limits. Idle voltage at the pad lies — measure under simultaneous demand (/guides/site-power-quality-basics).

7. Network soft-offline — DNS/TLS/identity after board or SIM work (/entries/protocol-ocpp-f02-heartbeat-timeout-to-csms, /entries/protocol-net-c01-dns-resolution-failure-to-csms).

Log sources / tools

Source	What you pull
Satellite UI	Stall code, satellite ID
ChargEye	Pool map, link status, coolant pressure, FW, module online
OCPP CSMS	StatusNotification, MeterValues ramp curve, connector mapping
Known-good cable / second EV	Moves CCS abort
Coolant gauges	Pressure/flow at power unit
Link path inspection	Connectors, junction boxes, recent civil work

Field checklist

1. One satellite, one power unit’s satellites, or whole site?
2. Link status in OEM tools / LEDs before RMA.
3. Pool demand vs online module capacity.
4. Recent civil work near link routes?
5. Known-good cable if ISO/CCS on one satellite only.
6. Feeder voltage under multi-satellite ramp.
7. RFID/auth vs energy path — split the ticket.

When to escalate

Electrician / site power — Feeder/PQ issues when the whole pool drops together; physical damage to link conduit after civil work that needs site electrical.

Network / CSMS / fleet ops — Auth/RFID mapping after credential or reader changes; allocation policy / LMS caps during peak — ops, not wrench; wrong CSMS connector mapping.

Kempower OEM — Confirmed link hardware after seating/config documented; coolant pump/skid with pressure evidence and fill procedure followed; SECC/power module after cable + vehicle + pool assignment cleared; pool firmware allocation bugs with ChargEye evidence pack.

Do not RMA a satellite power path for allocation stuck or RFID miswire. Do not treat single-stall CCS abort as pool coolant failure without pressure data. Payment-only failures belong to billing (/entries/ingenico-pay-to-payment-terminal-timeout).