Free tool

PCB Trace Width Calculator — IPC-2152 vs IPC-2221

Enter current and allowed temperature rise. You get the IPC-2221 answer that almost every free calculator gives, alongside an IPC-2152-informed estimate — and an explanation of why they differ, which on internal layers is by a lot.

IPC-2221 width
mil — the 1998 formula
IPC-2152 estimate
mil — adjusted, see below
Cross-section
mils² at the 2221 width
Your width
CurrentIPC-2221 widthIPC-2152 estimateDifference

Runs entirely in your browser. Nothing is uploaded, stored, or sent anywhere.

Most free trace-width calculators are running a 1998 formula

The equation almost every online calculator uses — I = k · ΔT0.44 · A0.725 — comes from IPC-2221, and its data is older than that. IPC-2152 superseded it in 2009 after a far more thorough test programme, and the results were different enough that designing to the old formula now routinely gives the wrong answer.

Most calculators do not tell you which standard they are using. This one shows both, and says where they disagree.

The internal-versus-external inversion

This is the big one. IPC-2221 applies a factor of two derating to internal traces: a buried trace is assumed to carry roughly half the current of an identical surface trace, on the reasoning that it cannot convect heat to the air.

IPC-2152's testing found that backwards. A buried trace is surrounded by laminate, and FR-4 conducts heat considerably better than still air does. In many configurations an internal trace runs cooler than the same trace on the surface. The old 2× derating is not conservative in an interesting way — it is simply wrong, and it costs real copper and real routing space on every inner layer.

Switch the layer selector and watch the two columns diverge. That gap is the cost of using an obsolete standard.

Nearby copper is the other missing variable

IPC-2221 knows about width, thickness, current and temperature rise. It does not know whether your trace runs alone in the middle of nowhere or beside a solid ground plane — and that changes the answer substantially, because the plane spreads heat away from the conductor. IPC-2152 addresses this with correction factors for exactly that situation.

On a real board almost every trace has copper near it, which is one reason IPC-2221 tends to be conservative on outer layers even before the internal-layer problem is considered.

What is and is not implemented here. The IPC-2221 column is the published formula, which is public and widely reproduced. IPC-2152's nomographs and correction-factor tables are copyrighted and are not reproduced — the second column is our own estimate, applying two of the standard's well-published findings to the 2221 baseline, and it is labelled as an estimate throughout. For a design that has to be signed off against IPC-2152, use the standard itself.

Temperature rise is a choice, not a limit

A 10 °C rise is the usual default and it is a design decision rather than a physical boundary. What matters is where the trace ends up: 10 °C on a board already sitting at 85 °C inside an enclosure is a very different proposition from 10 °C on the bench. And copper's resistance rises about 0.39% per degree, so a hot trace also drops more voltage — which is frequently the real constraint on a low-voltage rail. The trace IR drop calculator works that side of it.

What neither standard covers

More free tools

Each of these runs entirely in your browser. Nothing is uploaded, stored or sent anywhere, and none of them needs an email address.

PDN target impedance →
Ripple and transient current to a target impedance, decap count and bulk capacitance.
Decoupling impedance plot →
Plots the anti-resonance peaks between capacitor self-resonances that flat models hide.
Buck converter design →
Inductor, output and input capacitance from the operating point. Vendor-neutral.
Interposer thermal (2.5D) →
Flip-chip on interposer: both heat paths resolved, every layer ranked, hot spot.

See all 49 engineering tools →

Stay in touch

New tools and notes, once a month at most

We add tools here fairly often and write up the things worth writing up — a stackup that behaved oddly, a standard that turned out to be obsolete, a calculator that was quietly wrong. Join and you get told when something new lands.

One email a month at the very most, and usually less. No drip sequence, no sales cadence, no sharing your address with anyone. Unsubscribe whenever you like.

Join the list →

Talk to us about a board design review →