Lightning Channel Cost Calculator: Full Lifecycle Fees

What Does a Lightning Channel Really Cost?

Operating a Lightning channel involves far more than the opening transaction fee. The full lifecycle cost includes the on-chain open, the on-chain close (cooperative or forced), ongoing rebalancing, the opportunity cost of locked capital, and watchtower fees for security. At typical 2025-2026 fee rates of 5-20 sat/vB, a single channel lifecycle costs 1,500-7,000+ sats in on-chain fees alone, before accounting for rebalancing or capital costs.

This reference breaks down each cost component with real transaction sizes from the BOLT 3 specification so you can calculate your total cost at any fee rate. Whether you are evaluating routing node profitability or comparing Lightning against other Layer 2 solutions, these numbers provide the baseline.

Opening Transaction Cost

A Lightning channel funding transaction spends from the opener's wallet into a 2-of-2 multisig output. With one P2WPKH input, one P2WSH channel output, and one P2WPKH change output, the standard size is approximately 141 vBytes.

These figures assume a single-input funding transaction. Wallets with fragmented UTXOs may require multiple inputs, adding approximately 68 vB per additional P2WPKH input. For transaction sizing details, see the Bitcoin transaction size reference.

Closing Transaction Costs

How a channel closes determines both the on-chain cost and the time before funds become spendable. There are two closure types: cooperative and force close.

Cooperative Close

A cooperative close is a single on-chain transaction signed by both parties. It spends the 2-of-2 P2WSH funding output into two settlement outputs (one per party). The typical size is approximately 170 vB.

In the best case, a channel lifecycle (open + cooperative close) costs roughly 311 vB total: 1,555 sats at 5 sat/vB or 3,110 sats at 10 sat/vB.

Force Close

A force close broadcasts the pre-signed commitment transaction unilaterally. With anchor outputs, the commitment transaction's base weight is 1,124 weight units (~281 vB) even with zero pending HTLCs. Each pending HTLC adds another ~43 vB. In a worst-case scenario with 483 pending HTLCs, the commitment transaction alone can reach ~15,000 vB.

Force closes also require additional sweep transactions. The initiator must wait out a CSV timelock (144-2,016 blocks, or roughly 1 day to 2 weeks depending on channel size) before sweeping their output. Each HTLC-timeout sweep adds ~167 vB and each HTLC-success sweep adds ~177 vB.

At 10 sat/vB, a simple force close (commitment + sweep) costs roughly 4,500 sats: 2.6x more than a cooperative close. With pending HTLCs, the cost can grow to 10-100x more.

Anchor Outputs and Their Cost Implications

Anchor outputs were introduced to solve a critical problem: commitment transactions needed their fee rates set in advance via update_fee, which could become stale by the time a force close occurred. Anchors allow either party to CPFP fee-bump the commitment transaction after broadcast.

The tradeoff is cost. Each anchor is a 330-sat P2WSH output (the dust limit), and two are required: one per party. That is 660 sats effectively burned per channel. Anchors also add ~100 vB to the commitment transaction weight (1,124 WU vs 724 WU for legacy channels). LND requires reserving at least one UTXO per anchor channel for potential fee bumping, further fragmenting your wallet.

Ephemeral Anchors and v3 Transactions

Ephemeral anchors, enabled by v3 transactions (standardized in Bitcoin Core v28.0), replace the two 330-sat anchors with a single anchor that can be zero-value. The commitment transaction itself pays zero mining fees: all fees come from the CPFP child transaction. The update_fee message is eliminated entirely. This reduces the burned sats and simplifies fee management, though v3's 1-parent-1-child constraint prevents batching multiple anchor spends.

Rebalancing Costs

Channel balance naturally drifts as payments flow in one direction. Once a channel becomes depleted on one side, it stops routing payments in that direction. Rebalancing restores capacity but adds ongoing costs that erode routing revenue.

Circular rebalancing (paying yourself through the network) costs only routing fees, typically 100-500+ ppm per rebalance. One documented operator spent 542 ppm on rebalancing while earning only 140 ppm in routing fees: a 3.87x cost-to-revenue ratio that made the node unprofitable.

Submarine swaps (Loop In/Out) involve both on-chain and off-chain fees. A Loop Out requires a 30,000-sat prepay deposit plus a service fee (~500 sats on a 300,000 sat swap) plus the on-chain sweep fee. During low-fee periods, a 500,000-sat Loop Out costs roughly 2,000-3,000 sats total. During fee spikes, on-chain costs alone can exceed that. For a walkthrough of Loop operations, see our Lightning Loop submarine swaps guide.

Total Channel Lifecycle Cost

The following table combines all cost components for a 5,000,000-sat (0.05 BTC) channel operated for 6 months at different fee rate environments. Rebalancing assumes one circular rebalance per month at 300 ppm on half the channel capacity.

These numbers exclude channel reserve (1% of capacity locked and unspendable), opportunity cost of locked capital, and the risk of a force close, which would roughly double the closing cost. For a routing node to profit, fee revenue must exceed these lifecycle costs. Use the Lightning channel calculator to model specific scenarios.

Channel Size and Routing ROI

Routing profitability depends on channel size, fee rates, and traffic volume. The median routing fee across the Lightning Network is approximately 100 ppm (0.01%), though the volume-weighted average is higher. To break even on a channel whose lifecycle costs 12,935 sats (the medium-fee scenario above), a 5,000,000-sat channel needs to route at least 129,350,000 sats (~1.29 BTC) at 100 ppm.

In practice, most independent routing nodes earn 1-4% APY on deployed capital. Block (Cash App's parent company) reported ~10% annualized returns on ~184 BTC of Lightning liquidity, but this relied on captive exchange traffic and aggressive fee rates averaging 1,053 ppm outgoing and 2,955 ppm incoming: rates far above the network median. For independent operators, revenue typically falls below the ~4.3% US Treasury yield benchmark, making Lightning routing a net negative on an opportunity-cost basis for most participants.

For fee comparisons across routing implementations, see the Lightning routing fee comparison.

Opportunity Cost of Locked Capital

Bitcoin locked in a payment channel cannot be used for lending, DeFi yield, or other productive strategies. With US Treasuries yielding ~4.3% and various stablecoin yield opportunities offering 3-8%, the opportunity cost of locking 1 BTC in Lightning for a year is roughly 43,000-80,000 sats in foregone yield (denominated in BTC terms).

Channel funds are also illiquid: withdrawing requires closing channels and paying on-chain fees. A force close can lock your capital for up to 2 weeks behind a CSV timelock. Bitcoin's price volatility adds another layer: channels are denominated in BTC, but costs and revenues are often measured in fiat terms.

Watchtower Fees

Watchtowers monitor the blockchain for revoked commitment transactions and broadcast justice transactions on your behalf. Without a watchtower, going offline while a channel partner broadcasts a stale state can result in total loss of channel funds.

Most available watchtower services are altruistic (free), including those offered by LN+ and Voltage. The Eye of Satoshi (TEOS) is an open-source watchtower that can run in altruistic mode. Commercial watchtower pricing has not standardized, but models include per-channel subscriptions and penalty-based fees (the tower takes a percentage of recovered funds). For most operators, watchtower costs are currently negligible but represent a real security dependency.

Lightning vs Spark vs On-Chain: Cost Comparison

Lightning's channel-based architecture requires upfront on-chain costs, ongoing rebalancing, and capital lockup. Spark uses a statechain-based design with FROST threshold signatures: transfers rotate key ownership off-chain without opening or closing channels. There is no inbound liquidity problem and no rebalancing requirement.

Spark-to-Spark transfers are currently free, with a small flat fee planned in the future. Spark-to-Lightning sends cost 0.25% plus Lightning routing fees. Lightning-to-Spark receives cost 0.15%. For users making frequent small payments without routing revenue to offset channel costs, Spark eliminates the fixed overhead of channel management entirely. For a deeper architectural comparison, see our research on what Spark is and how it works.

Frequently Asked Questions

How much does it cost to open a Lightning channel?

Opening a Lightning channel requires one on-chain transaction of approximately 141 vBytes. At 10 sat/vB, that costs 1,410 sats (~$1.41 at $100k BTC). At 50 sat/vB during congestion, the same transaction costs 7,050 sats (~$7.05). The actual size varies depending on your wallet's UTXO set: multiple inputs increase the transaction weight.

How much more expensive is a force close than a cooperative close?

A simple force close (no pending HTLCs, with anchor outputs) produces a commitment transaction of ~281 vB, roughly 1.6x the size of a cooperative close (~170 vB). However, force closes also require additional sweep transactions and impose a CSV timelock delay of 144-2,016 blocks (1 day to 2 weeks). With pending HTLCs, each adds ~43 vB to the commitment plus a separate ~170 vB sweep transaction, potentially making force closes 10x or more expensive than a cooperative close.

What are anchor outputs and why do they cost extra?

Anchor outputs are 330-sat outputs added to commitment transactions that allow either channel party to fee-bump the transaction via CPFP after broadcast. Two anchors (660 sats total) are burned per channel. They add ~100 vB to commitment transaction weight and require reserving a UTXO for potential fee bumping. The newer ephemeral anchor design (using v3 transactions) reduces this to a single zero-value anchor, eliminating the burned sats.

Is running a Lightning routing node profitable?

For most independent operators, no. Typical routing yields are 1-4% APY on deployed capital, which falls below the ~4.3% risk-free rate on US Treasuries. After accounting for on-chain fees (channel opens/closes), rebalancing costs, hosting, and opportunity cost, many nodes operate at break-even or a loss. Large operators with captive traffic (like exchange nodes) can achieve higher returns but rely on fee rates well above the network median.

What is the minimum channel size to be profitable?

There is no universal minimum, but the math sets a clear floor. If opening and closing a channel costs 3,110 sats (at 10 sat/vB) and you earn the median routing fee of 100 ppm, you need to route at least 31,100,000 sats (0.31 BTC) through the channel just to cover on-chain costs. This excludes rebalancing and opportunity cost. Channels under 1,000,000 sats rarely generate enough routing volume to justify their lifecycle cost.

How do Bitcoin fee spikes affect Lightning channel costs?

Fee spikes directly increase the cost of opening, closing, and force closing channels. During the July 2025 spike to 265 sat/vB, opening a channel cost over 37,000 sats (~$37). A force close at that rate would cost 74,000+ sats. Fee spikes also make HTLC resolution more expensive and can cause channel jamming scenarios where small HTLCs become uneconomical to resolve on-chain.

How does Spark eliminate Lightning channel costs?

Spark uses a statechain architecture instead of payment channels. Transfers rotate cryptographic key ownership off-chain without opening or closing channels. There is no channel to fund, no rebalancing to perform, and no inbound liquidity to manage. Users pay only a standard mining fee to deposit BTC and a withdrawal fee to return to L1. Spark-to-Spark transfers are currently free. The tradeoff is a different trust model: Spark relies on an honest-majority assumption among its operator set during transfers, while Lightning is fully trustless between channel partners.

This tool is for informational purposes only and does not constitute financial advice. Transaction sizes are based on the BOLT 3 specification and may vary with implementation details, script types, and number of inputs. Fee rates fluctuate continuously. Always verify current mempool conditions before making channel management decisions.