Best Bitcoin Card

On-chain vs Lightning Card Funding

Short answer: For loads under roughly $200, on-chain network fees can exceed every card fee combined; Lightning (or stablecoin rails) fixes that - if the card supports it.

How you get bitcoin onto a card decides the real cost of small top-ups. The card's fee schedule matters less than the network fee - and the difference is biggest exactly when you top up small.

On-chain funding

Direct BTC loads (Freedomia, Goblin Card, 2Fiat) pay bitcoin network fees. Those fees don't care whether you're loading $50 or $5,000 - a congested mempool can charge $2-5 flat.

The math that matters: a $3 fee on a $100 load is 3% - more than most card load fees. On a $1,000 load it's 0.3%. On-chain funding is fine for large, infrequent loads and expensive for small ones.

Lightning funding

Lightning payments cost fractions of a cent and settle in seconds. Where a card supports it (Freedomia does; Cash App operates within its own Lightning ecosystem), small top-ups become practical again.

The catch: Lightning requires the provider to run (or use) Lightning infrastructure, inbound liquidity on their side, and an invoice flow in the app. Not every card that mentions "Lightning" supports funding over it - some only support withdrawals.

What to check before assuming Lightning works

  1. Direction. Does the card accept funding over Lightning, or only pay out over it? Our comparison distinguishes the two.
  2. Limits. Lightning channels have capacity limits. Large loads may fall back to on-chain or fail.
  3. Fee display. A provider that documents Lightning funding usually documents its routing fees. One that doesn't is a guess.
  4. Invoice expiry. Lightning invoices expire in minutes. Generate the invoice when you're ready to pay, not before.

Practical guidance

  • Frequent small top-ups: Lightning, or a stablecoin-funded card (no network fee in either direction).
  • Occasional large loads: on-chain is fine - the percentage becomes negligible.
  • US-based Cash App users: the app's Lightning integration handles micro-load inside its own balance model, which is a different setup than a standalone card.

What to verify on your own card

Load $20, note the network fee, note what lands on the balance. Then load $200 and compare the percentages. If the small load lost more than 5% to fees, the card is built for larger loads - and you should fund it accordingly.

The mixed strategy

Many providers accept both rails on the same card. In practice the cleanest setup is a hybrid:

  • A standing balance funded once with a larger on-chain load, so the network fee percentage stays negligible.
  • Lightning for the frequent small top-ups that would otherwise bleed percentage fees.

If the provider supports only one rail, pick the one matching your load pattern: on-chain for monthly funding, Lightning (or a stablecoin rail) for weekly ones. A card that forces small on-chain loads is quietly taking a percentage every week.

When the network fee spikes

Bitcoin fee markets are cyclical. During congestion:

  1. Batch. One $500 load costs the same network fee as five $100 loads - at 5x the per-load percentage.
  2. Time it. Weekend and off-peak UTC hours usually clear cheaper. A provider app that lets you set the load amount and pay later lets you pick the moment.
  3. Check the fee estimate before confirming. Most wallets display the sat/vB rate; anything above ~50 sat/vB is congestion pricing.
  4. Switch rails. If the provider takes stablecoin funding (USDT/USDC on a cheap network), moving dollars to the exchange and loading from there often beats paying an on-chain BTC fee at the peak.

What this changes about card choice

If your funding pattern is small and frequent, the load rail is the most important line in the comparison - more than cashback or FX fees. A card with 0% load fees that only accepts on-chain BTC will still cost you network fees on every top-up. A card with a small load fee but a Lightning rail may be cheaper overall. Compare the whole path, not the fee table.