Notes on measured and perceived sweetness in brews

While I was thinking about how I would alter the peanut butter mead recipe I wrote about here, I came across this article from WineFolly, which displays well-known wine types alongside two scales: the absolute measurement of sugar in grams per litre as well as the corresponding human perception of sweetness. This will be useful if — oh, all right, when — I start planning recipes out with specific results in mind instead of accepting whatever may result after my best efforts.

Briefly speaking, the WineFolly article says:

  • Higher residual sugar levels -> higher perceived sweetness
  • Higher acid content -> lower perceived sweetness
  • Higher concentration of bittering agents -> lower perceived sweetness

And for reference, here’s a summary of their chart:

Perceived sweetnessResidual sugar (g/L)Example beverage
Dry0 - 9Most table wines
Off-dry9 - 18Extra-dry champagne
Medium dry/semi-sweet18 - 50Sweet Riesling (drier range), dry champagne, demi-sec champagne
Medium sweet50 - 120Sweet Riesling (sweeter range), port, Madeira, Coca-Cola
Sweet120+Rutherglen Muscat, Tokaji, Sauternes

I was curious also about how alcohol content affected perceived sweetness. According to this paper, higher concentrations of ethanol lead to higher perceived bitterness or sourness, which implies lower perceived sweetness. The paper also says the ABV ceiling for a dry wine to still be perceived as well-balanced is about 14%. I have had a tendency toward making my wines and meads with yeasts that are tolerant to at least 14%, so I’ll try for some slightly gentler brews in the future.

Now for the numbers. When calculating sugar content, most home winemakers use the Brix scale, which is one of several measurement scales concerning dissolved sucrose in a solution of water. One degree Brix denotes a 1% by mass sucrose solution. Most sugars brewers use are not pure sucrose, but if the Wikipedia article on Brix is to be believed, other sugars behave similarly enough in hydrometer readings and during fermentation that for the purpose of home winemaking and recordkeeping, this level of precision is good enough.

Most hydrometers include a Brix measurement scale alongside the specific gravity and potential alcohol scales, but in case yours is missing any of these, here are a couple of formulas for your reference (source):

  • Brix = 220×(specific gravity – 1) + 1.6
    • Note that 1 Brix = 1% residual sugar, or 10 grams per litre
  • Potential alcohol = 0.6 × Brix – 1

So where does that leave us now? Still not with a lot of solid answers, to be honest. All of the above is just a set of guidelines that doesn’t guarantee that results will be exactly as you calculate, mainly because this is a living process. Off the top of my head, here are the things most likely to affect your finished sweetness:

  • Incomplete fermentation due to suboptimal yeast conditions (nutrient availability and balance, temperature, dead yeast, contaminants, gremlins)
  • Overenthusiastic yeast fermenting to a higher alcohol tolerance than anticipated
  • Tricky yeast reactivating after bottling and giving you a semi-sweet carbonated beverage instead of a sweet, still one
  • Human error in measurements, calculations, and recordkeeping

Considering all of the above, it’s probably best to take a Bob Ross approach to your zymurgic experimentation. As long as you’ve done your best to think and work sensibly through your process, you should get something that, while it may not be exactly what you had in mind, will still result in a happy accident.

I grabbed a screenshot of brSquared.org’s chart on relationships between hydrometer readings, sugar concentrations, and potential alcohol, since I find myself frequently referring to it. The original is here.

Leave a Reply