Lightning Channels
Public payment channels observed by mempool.emzy.de
Saved observation
29,594 public channels
- Measured at
- Retrieved at
- Source
- mempool.emzy.de · Dataset used
Recent observation. The measurement is less than 72 hours old when this page was generated on .
This observation and the chart use the same history from mempool.emzy.de, an independent instance of the mempool software. They cover the public graph observed by Emzy, not all channels, users or payment volume.
History available from .
The history is replaced in full when the site updates, without adding points from a different observer. Each point’s date is shown in its tooltip. The saved observation remains readable if the chart does not load.
- Source
- mempool.emzy.de
What this measures
This chart tracks the total number of public payment channels open on the Lightning Network, according to mempool.emzy.de. A channel links 2 nodes. It is created by a Bitcoin transaction that locks funds in an output the 2 nodes control jointly, and it disappears when that output is spent, at closing. Every public channel therefore matches a real UTXO on the chain, which makes this count harder to inflate than the node count.
What the counter cannot see
Only channels announced to the rest of the network are counted. A channel can be opened without being announced: that is the case for many non-custodial mobile wallets, whose channel with the provider’s node is not used to route other people’s payments. These private channels exist on-chain, but there is no reliable way to tally them. Customers of custodial services have no channel at all: their balance is just a row in the operator’s database.
Channel count says nothing about size either. A channel can hold a few thousand satoshis or several bitcoin. To measure the funds committed, look at network capacity, and for the size of a typical channel, average channel size.
Fewer channels, not necessarily less usage
Opening or closing a channel costs a Bitcoin transaction. When fees climb, as they did during the Ordinals wave of 2023, maintaining many small channels gets expensive, and operators may prefer to merge them into larger ones. Splicing points the same way: it lets a node add or remove funds from a channel without closing it. A channel count that falls while capacity holds up can therefore signal consolidation rather than abandonment. Fee rates help put those moves in context.
Using Lightning in self-custody takes a channel, which means at least one on-chain transaction to open it and another to close it. With block space scarce, hundreds of millions of users could not all open and close their own channels on a regular basis. That is one reason protocols like Ark try to fit many users into a single Bitcoin output while staying connected to Lightning.
🪙 What is the Ark protocol? The missing piece to make Bitcoin a scalable everyday currency
This chart counts channels, not the people who control them. And the cost of a channel pushes newcomers toward operators who handle liquidity on their behalf, and their model decides how much self-custody survives on Lightning.