Index / PR63

PR63

PR63 — zinc air, 1.4 V, 5.8 x 2.1 mm. Not listed by Energizer, and what still fits does not match the circuit: 1191SO drops straight in, and runs 1.55 V — +10.7%, which reads high.

What fits the compartment, and what does not

Fit answers whether it enters the compartment. Voltage answers whether the circuit will notice. A cell that fits is not always a safe substitute.

One cell is the same size, and it is not the same voltage: 1191SO runs +10.7% against this one, high enough that a calibrated instrument reads above true. Whether that matters is a question about the device, not about the cell.

Also stamped 7012ZD, AC5

Unlisted, and what fits runs at the wrong volts

Energizer no longer lists PR63 (also 7012ZD), and what fits the compartment does not match the circuit: 1191SO drops straight in but runs 10.7% off, high enough that a calibrated instrument reads above true. In a meter or a camera calibrated for this cell, that shows up as a reading, not as a failure.

What the chemistry changes

Everything above answers size and voltage. What it cannot answer is chemistry, and 1 other kind is in play here beside the zinc air this cell uses.

• Zinc air takes oxygen from the air through the holes in its face. It starts discharging the moment the tab comes off and runs down in weeks whether the device is used or not, and a sealed compartment starves it.
• Silver oxide holds its voltage almost flat until it is spent, which is why watches and light meters were built around it.

A cell that starts working when the tab comes off

Zinc air takes its oxygen from the room. The sealed tab on the face is what keeps the cell asleep, and pulling it starts the clock whether the aid is switched on or not. Energizer publishes 33 mAh for PR63, and that figure assumes air reaching the holes.

A sealed compartment starves it. 1191SO in this envelope does not need air, which is why PR63 is the wrong swap into a watch case that closes on a gasket. Nominal 1.4 V, which is below every alkaline and silver oxide cell of this envelope and is what the aid's amplifier is tuned to. The nearest is 1191SO: it drops straight in and runs 1.55 V, reads off.

• Hearing aids are medical devices. Their manufacturers name the cell on the case and in the manual, and that name wins over any table, including this one.
• 5 cells measure close enough to compare against it in the Energizer data this site reads, and 5 of them are still listed.
• By stored energy it stands 1 of 2 in its envelope, at 46 milliwatt-hours.
• Energizer files it for North America, which is where the part is sold rather than where the cell will work.

Drawn to scale

Every published figure

11 figures are published, and one is not: impedance. The row is absent rather than estimated. It reached shelves as AC5.

2 rows in the table are not the maker's but ours: stored energy and energy per cubic centimeter. Everything else is reproduced without change.

Where the capacity figure stops counting

A capacity figure means nothing without the voltage it was counted to. For PR63 the maker counts down to 0.9 V, and 2 published figures sit beside it.

The part occupies 0.05 cubic centimeters and weighs 0.2 g. Capacity and voltage together put 46 milliwatt-hours in that space. Energizer files it for North America.

Cutoff voltage

0.9 V. The published capacity is counted down to this voltage; a device that quits higher up sees less than the whole figure.

Mass against volume

0.2 g filling 0.05 cubic centimeters. When the printing has worn off, a kitchen scale separates two cells of one envelope faster than a caliper does.

Where the part is filed

North America. That is a statement about the maker's distribution and not about where the cell works.

Is that a lot of energy?

Nothing else measured in this shell holds more. At 46 milliwatt-hours it leads 1 other cell, and the next one down carries 49% of what it does. Per unit of volume that is 924 mWh per cubic centimeter, denser than 99% of everything measured here.

More energy per cubic centimeter than most

PR63 packs more into its volume than most of what is measured here: 924 milliwatt-hours per cubic centimeter, above 99 percent of the corpus.

5 of those ones are still listed. Its 924 milliwatt-hours per cubic centimeter sit above 99 percent of everything here that publishes both a capacity and a volume. 14 of the 82 coins share its 1.4 V exactly.

• 317 — the next step down by volume: 0.04 cubic centimeters against this cell's 0.05.
• 339 — the next step up: 0.053 cubic centimeters, 106% of this one.
• 1191SO — 5.8 x 2.1 mm, 1.55 V, silver oxide.
• PR70 — 5.8 x 3.6 mm, 1.45 V, zinc air.

The maker's own part numbers

PR63 carries 2 names beyond the designation, and every one of them is a manufacturer part number: a catalog line, not a size and not a standard.

• 7012ZD — a manufacturer part number for PR63, not a standard designation.
• AC5 — a manufacturer part number for PR63, not a standard designation.

The cell that differs only in depth

The code says the size, so the cells that get confused with PR63 are the ones that keep its diameter and change its depth. 4 cells share this diameter at depths from 1.2 to 3.6 mm, and every one of them enters the same opening.

It stands 1 of 2 in this envelope by stored energy. On a scale the difference shows before it shows on a caliper: this one weighs 0.2 g.

• 335 — same diameter, 1.2 mm tall against this cell's 2.1: 0.9 mm shallower, which neither designation carries.
• 317 — same diameter, 1.6 mm tall against this cell's 2.1: 0.5 mm shallower, which neither designation carries.
• 319 — same diameter, 2.7 mm tall against this cell's 2.1: 0.6 mm deeper, which neither designation carries.
• 1191SO — the same 5.8 x 2.1 mm in silver oxide at 1.55 V. Identical on a caliper, different in the circuit.
• 164 — 6.8 mm across at 1.4 V against 5.8 mm here, 1 mm apart: too wide to enter in this holder.

What this page cannot tell you

2 of the 7 fields this site reads are empty for PR63: impedance and operating temperature. Everything computed from them is absent from this page rather than estimated.

The blanks are the same in all 1 record behind this designation, so they are the catalog's silence and not a merge losing a value. What is not missing is the envelope: 5.8 x 2.1 mm, published and checked against the designation itself.

• No impedance — so this page cannot tell you whether the cell can drive a pulse, which is the whole question in a watch or a sensor.
• No operating temperature — so this page cannot tell you whether the maker claims anything at all below freezing.

Before you swap anything

Everything here is computed from published dimensions and nominal voltages. That answers whether a cell enters the compartment and whether the voltage matches on paper. It does not answer what your device does with the difference, and it cannot: a cell that fits can still be the wrong cell.

One more limit, and it is the biggest. This site reads a single manufacturer’s catalog. When a page says a designation is not listed, that means Energizer no longer offers a part under it — not that nobody makes the cell. Other makers may still sell it, and for common sizes they usually do.

A voltage difference is one relation between two cells, and this page prints it from the cell it is about — in every table on the page, including the one listing cells no longer made. A figure always belongs to the cell whose voltage is printed beside it: a cell that runs lower than this one carries a minus here, one that runs higher carries a plus, and two cells at the same nominal voltage carry zero. Meet the same pair on the other cell’s page and the sign is the other one, with the figure taken against that cell instead of this one; where the two run at the same voltage, both pages print zero. The class beside the figure is not read off that figure directly. It is decided on the gap itself, taken against the higher of the two nominal voltages, so one pair of cells gets one class whichever page you meet it on: counted from this side the boundary of the class that still runs falls at 20.0% below and 25.0% above, which is the same gap divided once by the larger nominal and once by the smaller. What genuinely differs between the two directions is the consequence, and the words carry it: a cell that runs low makes a calibrated instrument read below true, and a cell that runs high makes it read above.

3 things this site cannot see. Whether the contacts reach a shorter cell. Whether the device was calibrated for a chemistry that holds its voltage flat, which silver oxide does and alkaline does not. Whether the equipment tolerates the higher current a different chemistry can deliver. Rechargeable cells are never a drop-in for primary cells regardless of size.

Cells measured next to this one

1 cell sits at exactly this size, and 4 more share the diameter at other depths. The first group is interchangeable by measurement; the second is not.

• 1191SO — 5.8 x 2.1 mm, 1.55 V
• 335 — 5.8 x 1.2 mm, 1.55 V
• 317 — 5.8 x 1.6 mm, 1.55 V
• 319 — 5.8 x 2.7 mm, 1.55 V
• PR70 — 5.8 x 3.6 mm, 1.45 V

Where these numbers come from

Dimensions, voltages, chemistries and capacities are published by the manufacturer and reproduced without change. Everything else on the page is ours, computed from those figures: whether one cell fits where another sat, how far apart their voltages are, stored energy and energy per cubic centimeter, the rank within a shell, and the check of the designation against the measured size.

Fit is decided within 0.3 mm on every axis, which is the order of a contact spring's travel. Voltage classes are decided by consequence, not by roundness, and they are decided on the gap between two nominal voltages taken against the higher of the two: 2.0% or less of it is the same working voltage, 6.0% starts to matter to a calibrated instrument, and past 20.0% it is a different class altogether. Measuring the gap against the higher figure is what makes the verdict the same from either cell’s page, while the printed percentage stays counted from the cell you are reading about. Cells with identical stored energy share a rank instead of being ordered arbitrarily.

Ranking uses capacity times nominal voltage, because capacity alone is not comparable across chemistries: 150 mAh at 3 V is twice the energy of 150 mAh at 1.5 V. Neighbors are chosen by measurement and never alphabetically, because two cells filed next to each other by name usually have nothing in common. A blank in any table means the maker publishes no figure, which is not the same as zero.

Source: Energizer technical data, retrieved 1 September 2026.

The record behind this page

Every figure on this page comes from one catalog record, number 729, and it is linked below. The sheet is the document the dimensions, voltage and capacity were read out of.

1. [1] Energizer catalog record 729, listed as AC5, obsolete in the snapshot. Technical data sheet: ac5.pdf.

Every figure above is reproduced from those records without change; the comparisons are ours and are described on the method page. Snapshot: Energizer technical data, retrieved 1 September 2026.

Putting this on a forum or a repair page? The same answer without the site around it — fit, voltage, chemistry and the limits that go with them — is at /embed/pr63/. It loads nothing from anywhere else. Paste this:

<iframe src="https://batterycross.com/embed/pr63/" width="100%" height="520" loading="lazy" title="PR63 replacements"></iframe>