Total area coverage for the standard printing conditions, taken from the ICC Profile Registry, plus measured lightness values for every kind of black.
Total ink is a property of the separation, not the press
Every unattributed figure online contradicts every other one, and there is a reason. The same FOGRA39 characterisation data ships as ICC profiles with 330 per cent, 300 per cent and 260 per cent total ink. The printing condition is identical in all three cases.
Total area coverage is a decision made when the separation is built — how much of the neutral component gets replaced by black — rather than a physical property of the press. So the honest answer to "what is the ink limit" is another question: which profile are you converting through?
The ICC Profile Registry publishes total ink, maximum black and black generation for each profile. It is free, it is primary, and no competitor in this space cites it.
Why 100% black looks washed out, measured
Solid black ink alone measures a lightness of about 16 on FOGRA51 data, against paper white at 95. That is dark, but it is visibly not as dark as it can be, and next to a rich black on the same sheet it reads grey.
A rich black — black ink with cyan, magenta and yellow underneath — measures about 13.1. Registration black at 400 per cent measures 12.71. So going from a sensible 250 per cent build to a reckless 400 per cent buys a lightness difference of 0.39, which is at the very edge of what anyone can see, in exchange for 150 points of ink the press has to dry.
One number on that table surprises people: cyan, magenta and yellow together at 300 per cent, with no black at all, measure 23.3. That is lighter than black ink on its own. Three inks are worse than one.
Small text must be 100% black, and this is why
A rich black is several inks stacked up, and every one of them is registered separately. The press holds registration to about 0.10 mm, which is invisible on a large panel and very visible on a serif at 8 point — coloured fringes on one side of every stroke.
IngramSpark publishes the only hard threshold I could find: text at 24 point or below should be 100 per cent black only, and hairlines should be 0.125 point at 100 per cent black. That is a good rule to adopt generally.
The failure mode is worth naming because it is easy to cause by accident. Converting an RGB document to CMYK does not turn RGB black into 100 per cent K. Measured through three common profiles it becomes a four-colour build every time — through FOGRA51 it comes out around 85/68/51/96. Flatten an RGB layout and your body text is now rich black.
FOGRA39 to FOGRA51 was about paper, not ink
The move from ISO Coated v2 to PSO Coated v3 is usually explained as a routine update. It was mostly about optical brighteners. FOGRA51 and 52 were measured under ISO 13655 M1, which includes the ultraviolet component of D50 illumination; FOGRA39 predates the M-conditions entirely.
Modern coated papers are heavily brightened, so they fluoresce, and how blue they look depends on how much UV your instrument emits. FOGRA52 records high substrate fluorescence and a paper white with a b* of −10. That is a genuinely different white to aim at. A secondary change moved black onto the same tone-value-increase curve as the other three inks.
Asked often
What total ink limit should I use?
Whatever your printer specifies, and it depends on the profile rather than the press. 300% suits most European coated work, 320% US sheetfed, 240% newsprint.
What is a good rich black?
Something around 60/50/40/100. It measures within a fraction of registration black while using far less ink, and it is a neutral rather than a cast.
Should small text be rich black?
No. Use 100% black alone. Every extra ink is registered separately, and at small sizes a tenth of a millimetre of misregistration shows as coloured fringing.