Keep EV Charging Payments Clearing Through Storm Season

California declared a statewide emergency this week ahead of a fall and winter that federal forecasters put at greater than a 90 percent chance of a very strong El Niño. State agencies are staging pumps, sandbags, and road-closure equipment now, weeks before the first storm. For a charging operator the equivalent question is what happens under the stall when a cell site floods, loses power, or has its fiber cut. Satellite backhaul is the answer most operators have heard of and few have designed correctly, because a dish on its own is one more single link, and a single link is the problem. This guide is for the operator, site host, or network lead, not the driver. It follows the transaction from the tap to settlement, shows where a single path breaks it, and lays out the bonded satellite architecture that keeps it clearing.

Why a Charger Goes Offline When the Charger Is Fine

Operator guides list connectivity as the leading cause of charger downtime and stop there. The mechanism matters, because the three ways a terrestrial network fails all share one trait: they happen somewhere between the site and the internet, and nothing at the site can fix them.

In each case the charger has power, the plug works, and the site shows green until a driver taps a card.

What Happens to the Payment at the Stall

A card-present payment at an unattended terminal is a round trip. The terminal reads the card, sends the authorization to the acquirer, and starts the session when the approval returns. Without a path out, the request never leaves the stall.

• The driver taps, the terminal waits, and the session is declined.
• The driver leaves. Most do not try a second time.
• The uptime report still counts the charger as available, because the charger was.

Uptime metrics count the charger, not the transaction, which is why a connectivity outage under-reports the revenue it costs. A site can post strong uptime through a storm week and settle nothing.

Why Satellite Is the Path That Survives the Region

Every terrestrial option shares infrastructure with the storm. Two carriers often ride the same tower. Two towers often ride the same fiber route. Fiber routes run along the same roads that flood and slide. A low-earth-orbit satellite path leaves the site straight up, touches none of it, and lands at a ground station outside the affected area.

• No dependence on any local tower, so a tower power loss or backhaul cut does not reach it.
• No dependence on any local fiber route, so a slide or a flooded vault does not reach it.
• No competition with the public surge, so congestion on the carriers does not reach it.

That independence is the whole case for satellite at a charging site. It is also the reason the next section exists, because independence from the region is not the same as reliability at the dish.

Why a Dish Alone Is Not the Plan

A dish connected directly to the charger is one link with one point of failure, and it fails in ways that have nothing to do with the storm around it.

• Heavy rain and wet snow attenuate the signal. The path degrades in exactly the weather the plan was built for.
• Any obstruction, a tree in leaf, a lift parked in the wrong place, breaks the view of the sky.
• Firmware restarts and satellite hand-offs cause brief drops that end an in-flight authorization.
• Latency is higher than fiber and varies, which some payment terminals handle poorly when it is the only path.

None of that is an argument against satellite. It is an argument against running it as the sole connection, which is what the next layer solves.

The Bonded Architecture: Peplink, SpeedFusion, and FusionHub

The design that holds is a bonded link, not a backup. Three components make it one.

Without FusionHub, bonding has nowhere to terminate and the router is back to choosing one link at a time. With it, the dish going through a rain cell and a tower dropping out both look like the same thing to the application: nothing.

The difference from dual-SIM failover is the difference between switching and sharing. Failover waits for the first link to die and then connects the second, dropping whatever was in flight. Bonding carries traffic on every link at all times, so there is nothing to switch to and nothing to drop.

Store-and-Forward at the Edge

Store-and-forward is the payment-side bridge for the gap that even a bonded network cannot close: the site itself losing power, or every path failing at once.

• Converts a hard decline into a deferred clear during a brief outage.
• Keeps the driver charging instead of leaving.
• Reconciles to the same ledger once the transaction clears.

• It is not connectivity. It holds the transaction; it does not send it.
• A deferred authorization can still be declined, and the operator carries that exposure.
• Limits on amount and count are set at underwriting, and a multi-day outage exceeds them.

Under a bonded satellite network, store-and-forward is rarely needed and earns its place when it is. Treated as the plan on its own, it is a way to discover a week of declines the following Monday.

What This Looks Like in Production

The bonded design above is running today at a fast-charging site Paygasus deployed for a charging operator in Southern California. The layer weights below are the production configuration.

The weighting is the design decision. The satellite and cellular paths at 5 percent each are warm: authenticated, carrying traffic, and proven that minute. When fiber drops, failover is a change in weights, not a reconnect. A cold dish that has to acquire satellites while a driver stands at the stall is the failure this avoids.

A Pre-Season Checklist for Charging Operators

Seven questions, answered per site, before the first storm.

• Which sites have one path out, and which of those carry fast-charging revenue.
• Which sites have a satellite terminal, and whether it is bonded through the router or hanging off the charger on its own.
• Whether the dish has a clear view of the sky in the direction it needs, with nothing that leafs out or gets parked in front of it.
• Which routers have battery backing, and for how long.
• Whether the two carriers at a dual-SIM site actually ride different towers.
• Whether the terminal supports store-and-forward and what its limits are.
• Whether payments, connectivity, and satellite service are one contract or three, because three vendors at a dark site means three tickets and no owner.

Where Paygasus Fits

Paygasus delivers the payment and the bonded network it travels on under one platform, one ledger, one contract.

• Starlink Authorized Solution Provider and Peplink Authorized Solution Provider: Paygasus sources the terminal and the router, configures SpeedFusion and the FusionHub endpoint, and installs at the site.
• Paygasus Connect bonds Starlink with AT&T, T-Mobile, and Verizon under one managed network, with carrier-agnostic SIM management and remote monitoring.
• Apollo and ApolloMax terminals at the stall, as exclusive North American distributor for Payter since April 2019, with store-and-forward at the edge.
• Settlement as Payment Facilitator for Fiserv and Full Liability Submitter for JPMorgan Chase, so the transaction and the network that carried it reconcile on one ledger.
• 25,000+ EV charging endpoints across 500+ locations, deployed and maintained across all 50 states, with 24/7 operating accountability behind the contract.

When a storm takes a path down, the partner that settles the transaction is the same partner that owns the dish, the router, and the SIMs the authorization rides on. There is one phone number, and it is answered.

Next in this series → Starlink for First Responders and Utility Crews: Filling the PACE Tiers Before Storm Season