Aftertouch is a remainder
The hammer has to travel from its rest position to the point where the jack escapes, and that distance is the blow distance minus the let-off. The key has to move that distance divided by the action ratio to make it happen. Whatever key dip is left after that is aftertouch.
With the defaults — 1.75 in of blow, a sixteenth of let-off, a ratio of 5 — the hammer rises 1.6875 in to escapement and the key uses 0.3375 in of its travel doing it. A 0.400 in dip leaves 0.0625 in, which the page prints as 62.5 thousandths, or about 1.59 mm.
Nothing about that is adjusted directly. There is no aftertouch screw. You get it by choosing the dip, or by choosing the let-off, or by choosing the blow, and it appears as the arithmetic consequence.
The lever cuts both ways
The sensitivity table is the part worth staring at. A thousandth of extra dip gives exactly a thousandth of extra aftertouch, because the dip is measured on the same side of the ratio as the aftertouch is. A thousandth of extra let-off gives a thousandth divided by the ratio, so at 5 to 1 it gives two tenths of a thousandth.
Ten thousandths of let-off — which is a large move, visible, and well outside where most specs put it — buys two thousandths of aftertouch. Ten thousandths of dip buys ten. Anyone who has spent an afternoon opening let-off in pursuit of aftertouch and got nowhere has met this ratio without being told its name.
The ratio row is the sleeper. Nudging the ratio from 5.0 to 5.1 moves aftertouch by 6.6 thousandths at these dimensions, which is more than a ten thousandth change in let-off does. The ratio is the input people are least sure of and it has more leverage than two of the three they are confident about.
Solving for the ratio instead
The second mode inverts the whole thing. Measure the key dip with a block, measure the aftertouch, and the ratio falls out as the hammer rise divided by the difference between them. It is worth doing on a few notes across the compass before trusting a single figure anywhere.
Ratios are not constant across a piano. They are designed to change, and then they change again with wear, with where the capstan ended up, with hammer weight and with how the whippen sits under the knuckle. A number that came off a data sheet for that model and a number measured on the note in front of you can differ enough to matter, and only one of them is describing the instrument.
What is not in here
Repetition, spring tension, jack return, how far the backcheck travels, key leverage against hammer inertia, and the order the regulation steps have to be done in. All of that is regulating an action; none of it is this arithmetic.
The dimensions themselves are also not here. Blow distance, let-off, drop, check height and dip are specifications belonging to the maker of that action, and where a rebuilder has chosen different ones, to the rebuilder. This page takes whatever you enter and reports what it implies. It does not know what any of them ought to be and it makes no judgement about the set you type in.
Questions people ask
What should the aftertouch be on a piano?
That is a specification belonging to the maker of the action, or to whoever set the instrument up, and this page will not supply one. What it will do is show what the figures you already have imply, and what has to change to land on the target you enter. If you are working without a spec, solving the ratio from a note that already plays well and then applying it across the section is a more defensible starting point than adopting a number from a different instrument.
Why does opening the let-off barely change the aftertouch?
Because let-off is measured at the hammer and aftertouch at the key, and the action ratio sits between them. At 5 to 1, a thousandth of hammer movement is only two tenths of a thousandth of key movement. Dip, by contrast, is measured at the key like the aftertouch is, so it moves one for one. That is a factor of five difference in leverage between two adjustments that both look like small dimensional changes.
How do I measure the action ratio?
The second mode on this page gives you one route: measure the key dip and the aftertouch on a note, and with the blow and let-off known the ratio is the hammer rise divided by the difference between dip and aftertouch. It is a static, indirect measurement and it inherits the error in all four inputs, so it is worth doing on several notes rather than one. Direct methods that measure hammer travel against key travel exist and are more accurate; this one uses tools already on the bench.
The calculator says there is no aftertouch at all. What does that mean?
It means the four numbers entered imply the key hits the keybed before the jack escapes. It is not a verdict on the action — it is arithmetic on your inputs, and the usual cause is one of the inputs being wrong rather than the action being incapable of escaping. Let-off measured from the hammer rest rather than from the string, and dip measured on a key that was already slightly down, are the two that turn up most.
Does this work for an upright as well as a grand?
The relationship holds for anything where a key drives a hammer through a ratio and escapement takes a share of the travel, so the arithmetic applies to both. What differs is the dimensions and the way they are measured — an upright is a different geometry with different reference points, its let-off is set to a different figure, and its spec sheet says different things. Enter the numbers for the action in front of you and the arithmetic follows; do not carry a grand dimension onto an upright.