Historically, table beer (or small beer) were the result of parti-gyling (or, more correctly, successive mashing).
The same mash would be used to brew a sequence of decreasingly dense worts, with small beer typically being the result of the final, least concentrated wort.
I used the technique a while ago to brew 10 liters of 1.125 SG imperial stout using the limited capacity of my 30 liter mash tun. Using
Randy Mosher's tables, I conceived a 1.090 recipe, from which I used the first runnings (50% of the volume of the theorical original brew), which ran up to 1.120. The second runnings turned out to be 1.055. Not too far off the mark.
Had I persevered, I could have brewed a third beer from this, which, if my estimation and feeble excel skills can be trusted, would have amounted to a 1.027 wort. Assuming 80% attenuation, would have made a 2.8% abv small beer.
Mind, I'm glad I didn't. I can imagine that very little maltiness remains after the second runnings, and what with history being, well, in the past, I prefer even my small beer to be a little more flavourful than what I suppose this hypothetical third runnings brew would have become.
But.
Using the same data, I could just as well use the same technique (50/50 split runnings) to brew one 1.044 beer from the first runnings of a theorical OG 1.034 beer, and draw a 1.023 wort from the second runnings.
Again using 80% attenuation, I'd end up with a 4.7% APA and a 2.4% table beer. Of course, I could dilute the second runnings down to whatever I think a table beer should be by sparging longer (risking astringency from the grain bed) or simply adding water to the boil kettle.
I think I'll give this a shot next time I brew something lower-ish on the SG scale. Will of course report back with results
.