Taproot vs Legacy Lightning Channels Compared
Compare Taproot Lightning channels to legacy P2WSH channels on privacy, fees, capacity, and wallet support. See real fee savings data.
Taproot Channels vs Legacy Channels Overview
Lightning channels have traditionally used P2WSH 2-of-2 multisig outputs for their funding transactions. This design works, but it leaves a clear on-chain fingerprint: anyone running a block explorer can identify these outputs as Lightning channels. Taproot channels replace that structure with a P2TR output that uses MuSig2 key aggregation, making cooperative closes indistinguishable from ordinary single-signature spends.
The specification for simple taproot channels (BOLT PR #995) merged in May 2026. LND shipped production support in v0.21.0 (June 2026), and Eclair followed with v0.14.0 (May 2026). The upgrade delivers measurable savings in on-chain fees and a significant privacy improvement for cooperative channel operations.
| Property | Legacy (P2WSH 2-of-2) | Taproot (P2TR MuSig2) |
|---|---|---|
| Funding output type | P2WSH | P2TR |
| Funding output size | 43 bytes | 43 bytes |
| Cooperative close input | ~96 vB | ~57.5 vB |
| Cooperative close witness | ~219 bytes (2 ECDSA sigs + redeem script) | ~66 bytes (1 Schnorr sig) |
| Cooperative close savings | Baseline | ~40% smaller input |
| On-chain fingerprint | Identifiable as 2-of-2 multisig | Looks like single-sig spend |
| Signature scheme | 2x ECDSA (DER-encoded, ~71 bytes each) | 1x Schnorr MuSig2 (fixed 64 bytes) |
| HTLC routing | Yes | Yes (PTLCs not yet available) |
| Public channel support | Yes | Not yet (pending gossip v2) |
| Specification | Original BOLT 2/3 | BOLT PR #995 (merged May 2026) |
On-Chain Footprint and Fee Savings
The funding output itself costs the same in both designs: a P2WSH output and a P2TR output are both 43 bytes. The savings appear when the channel is spent, specifically during cooperative closes. A legacy cooperative close input requires approximately 96 vB because the witness must include two DER-encoded ECDSA signatures (each roughly 71 bytes) plus the redeem script revealing the 2-of-2 multisig structure. That adds up to about 219 witness bytes.
A Taproot cooperative close uses a key-path spend with a single Schnorr signature produced via MuSig2. That signature is a fixed 64 bytes plus a sighash flag byte, totaling about 66 witness bytes and roughly 57.5 vB for the input. The result is approximately 40% fewer virtual bytes per cooperative close input.
To estimate concrete savings for your channels, use the Taproot fee savings calculator. The table below shows approximate costs at various fee rates for a single cooperative close input.
| Fee Rate (sat/vB) | Legacy Close Input Cost | Taproot Close Input Cost | Savings per Close |
|---|---|---|---|
| 5 | ~480 sats | ~288 sats | ~192 sats |
| 10 | ~960 sats | ~575 sats | ~385 sats |
| 25 | ~2,400 sats | ~1,438 sats | ~962 sats |
| 50 | ~4,800 sats | ~2,875 sats | ~1,925 sats |
| 100 | ~9,600 sats | ~5,750 sats | ~3,850 sats |
These figures represent only the spending input. A complete close transaction also includes the transaction header, outputs, and any change. The percentage savings on the total transaction are smaller than 40% because the input is only part of the transaction weight, but the absolute savings scale linearly with the fee rate.
Privacy Improvements
Legacy Lightning channels are identifiable on-chain in two ways. First, the funding transaction creates a P2WSH output with a known value pattern that chain analysis firms can flag as a probable channel open. Second, when the channel closes cooperatively, the witness data reveals both individual public keys and the 2-of-2 multisig redeem script, confirming it was a Lightning channel.
Taproot channels improve both dimensions. The funding output is a standard P2TR output, indistinguishable from any other Taproot wallet output. When the channel closes cooperatively via the key path, the on-chain transaction shows only a single aggregated public key and a single Schnorr signature. There is no way to determine from the transaction alone that two parties were involved.
The privacy gain has an important caveat: force closes still reveal the internal script structure. When a channel is force-closed, the commitment transaction must use a script-path spend, exposing the Taproot leaf script and making the channel identifiable. Additionally, in-flight HTLCs remain visible in both channel types. Full payment-level privacy requires a future upgrade to PTLCs, which is a separate protocol change not yet deployed.
Force Close Behavior
Force closes work differently between the two channel types at the script level, though the user experience is similar. In legacy channels, the commitment transaction spends the P2WSH funding output using OP_CHECKMULTISIG with the two channel keys. In Taproot channels, the commitment transaction uses a script-path spend through the Taproot tree, revealing the specific leaf that encodes the revocation and delayed-payment conditions.
Taproot channels with anchor outputs use ephemeral anchors and fee bumping via CPFP. LND v0.21.0 introduced RBF cooperative close for Taproot channels, allowing you to bump the fee on a stuck, unconfirmed cooperative close transaction. This is a practical improvement over legacy channels where a stuck close required waiting or using CPFP with an anchor output.
Force close costs remain comparable between the two types because the commitment transaction must include the script-path witness, HTLC outputs, and sweep transactions regardless of the channel type. The savings from Taproot are concentrated in the cooperative close path, which is the common case for well-managed channels.
Implementation Support
Simple taproot channels are defined by BOLT PR #995, which merged into the Lightning specification in May 2026. Support varies across implementations.
| Implementation | Taproot Channel Support | Version | Status |
|---|---|---|---|
| LND | Yes | Experimental in v0.17 (Oct 2023), production in v0.21 (Jun 2026) | Feature bits 80/81 finalized |
| Eclair | Yes | Initial in v0.13.0, production in v0.14.0 (May 2026) | Enabled by default |
| Core Lightning (CLN) | No | N/A | Not yet implemented as of mid-2026 |
| LDK | In progress | N/A | Development ongoing |
To open a Taproot channel with LND, use lncli openchannel --channel_type=taproot. Both peers must support the feature. If your counterpart runs CLN or an older LND version, the channel will fall back to legacy P2WSH. For a full comparison of Lightning implementations, see the Lightning implementation comparison.
Taproot channels are currently limited to unannounced (private) channels. Public channel support requires updates to the Lightning gossip protocol (gossip v2), which is still under development. This means routing nodes that rely on publicly announced channels cannot yet use Taproot channels for their public graph presence.
Should You Close Legacy Channels to Open Taproot Ones?
In most cases, no. Closing a legacy channel and opening a new Taproot channel requires two on-chain transactions: one to close the old channel and one to fund the new one. The fee savings from Taproot cooperative closes are real but modest in absolute terms, especially at low fee rates. You would need many future open/close cycles to recoup the cost of the migration itself.
The calculus changes if you plan to close the channel anyway, or if privacy is a priority. When you naturally close a legacy channel and open a new one, choosing a Taproot channel for the replacement costs nothing extra and delivers both privacy and fee benefits going forward.
A future protocol upgrade may allow in-place channel type upgrades without closing and reopening. The splicing mechanism could potentially be extended to change the funding output type, but this is not yet specified or implemented. For now, the practical advice is to open new channels as Taproot and let legacy channels close naturally.
Does My Wallet Need to Update?
If you use a non-custodial Lightning wallet that connects to your own node, you need to ensure your node software supports Taproot channels. For LND, upgrade to v0.21.0 or later. For Eclair, upgrade to v0.14.0 or later.
If you use a mobile wallet that connects to a Lightning Service Provider (LSP), the wallet vendor handles the channel management. You benefit from Taproot channels automatically once your LSP upgrades. Check with your wallet provider for their current channel type support. For a comparison of wallet capabilities, see the Lightning wallet comparison.
Layer 2 protocols built on top of Lightning also benefit. Spark, for example, operates as a Bitcoin Layer 2 that abstracts away channel management entirely. Users on Spark do not need to manage channels directly, and the protocol can adopt Taproot channels at the infrastructure level to reduce costs and improve privacy for all participants.
What Taproot Channels Do Not Change
Taproot channels are a meaningful upgrade, but they do not solve every Lightning privacy or cost problem. Several limitations remain unchanged:
- Payment routing still uses HTLCs, not PTLCs. The payment hash is the same across all hops, enabling correlation by routing nodes.
- Force closes still reveal internal channel scripts, making them identifiable on-chain regardless of channel type.
- Channel capacity is still visible during force closes, which leaks balance information.
- Public (announced) Taproot channels are not yet supported, so routing nodes cannot advertise Taproot channels to the network graph.
- Taproot channels do not change the Lightning routing fee structure or improve payment success rates.
For a broader look at Lightning privacy, including blinded paths and onion routing, see our research on Lightning Network privacy.
Frequently Asked Questions
What is a simple taproot channel?
A simple taproot channel is a Lightning payment channel that uses a P2TR funding output with MuSig2 key aggregation instead of the traditional P2WSH 2-of-2 multisig. The "simple" prefix distinguishes it from more complex future designs that might use additional Taproot script paths. Simple taproot channels use the key-path spend for cooperative closes and a script-path spend for force closes, keeping the design minimal while delivering privacy and fee improvements.
How much do Taproot channels save on fees?
The cooperative close input shrinks from roughly 96 vB to roughly 57.5 vB, a reduction of about 40%. At a fee rate of 50 sat/vB, that saves approximately 1,925 sats per close. The total transaction savings are somewhat less than 40% because the input is only one component of the transaction weight. Force close costs are comparable between the two types. Use the Taproot fee savings calculator for estimates at current fee rates.
Can I open a Taproot channel with any Lightning node?
Both peers must support simple taproot channels. If you run LND v0.21+ and your counterpart runs CLN or an older LND version, the channel will negotiate down to a legacy P2WSH type. Currently, LND and Eclair support Taproot channels in production. Core Lightning does not yet support them. You can check your peer's feature bits to determine compatibility before opening a channel.
Are Taproot channels more private than legacy channels?
Yes, for cooperative closes. A Taproot cooperative close looks like an ordinary single-signature transaction on-chain, while a legacy close reveals the 2-of-2 multisig structure. However, force closes still expose the channel's internal script tree, and payment routing still uses HTLCs with the same hash across hops. Full privacy improvements require future protocol upgrades including PTLCs and gossip v2.
Should I migrate my existing channels to Taproot?
Generally no. Migrating requires closing the legacy channel and opening a new one, which costs two on-chain transactions. The fee savings from Taproot closes are real but would take many channel lifecycles to offset the migration cost. The recommended approach is to open new channels as Taproot and let legacy channels close on their own schedule.
Do Taproot channels support public (announced) channels?
Not yet. Simple taproot channels are currently limited to unannounced (private) channels. Public channel support depends on updates to the Lightning gossip protocol (gossip v2), which is still being developed. Routing nodes that need publicly announced channels must continue using legacy channel types for their graph presence.
What is the difference between Taproot channels and PTLCs?
Taproot channels change the funding and commitment transaction structure to use P2TR outputs and Schnorr signatures. PTLCs (Point Time-Locked Contracts) replace the HTLC payment routing mechanism with adaptor signatures, eliminating the shared payment hash that enables cross-hop correlation. These are separate upgrades: Taproot channels improve on-chain privacy and fees, while PTLCs improve payment-level routing privacy. Taproot channels are deployed today; PTLCs are not.
This tool is for informational purposes only and does not constitute financial advice. Transaction size estimates are approximate and based on standard witness structures. Actual sizes may vary depending on the number of outputs, pending HTLCs, and implementation-specific details. Always verify current implementation support and fee rates before making channel management decisions.
Build with Spark
Integrate bitcoin, Lightning, and stablecoins into your app with a few lines of code.
Read the docs →
