[dmtaylor]

I scale my recipes by percentages.  I haven't done side-by-side experiments but I get the impression that if I'm increasing the base malt, I *do* want to increase the other malts accordingly.  Call it a hunch at this point, but this is how I do my big beers.  Especially stouts or black beers where adding more doesn't seem to really hurt anything and probably only helps.

[Drewch]

I don't know if it'll give you the character you're looking for, but Pattinson has documented lots of big stouts that include 10-20% sugar (typically, various English invert sugars) to boost the ABV.

Here's an example from The Homebrewer's Guide to Vintage Beer:

1924 Barclay Perkins IBS Ex

Mild malt 12.00 lb (5.4 kg) 53.33%
Amber malt 3.75 lb (1.7 kg) 16.67%
Brown malt 2.75 lb (1.2 kg) 12.22%
Black malt 1.50 lb (680 g) 6.67%
No. 2 invert sugar 2.50 lb (1.1 kg) 11.11%

Golding 90 min 5.50 oz (156 g)
Golding 60 min 5.50 oz (156 g)
Golding 30 min 5.25 oz (149 g)

OG 1103
FG 1040
ABV 8.33

[denny]

I scale my recipes by percentages.  I haven't done side-by-side experiments but I get the impression that if I'm increasing the base malt, I *do* want to increase the other malts accordingly.  Call it a hunch at this point, but this is how I do my big beers.  Especially stouts or black beers where adding more doesn't seem to really hurt anything and probably only helps.

Same way I think, Dave. If I was formulating a recipe from scratch, I'd use %. Why not when scaling up?

[BrewBama]

When I scale up/down a recipe in my brewing software of choice it increases/decreases all the grains keeping them at the same contribution percentage.  All the weights of all the grains increase/decrease.

What I believe that technique doesn’t take into consideration is that as the grain mass increases the efficiency of extracting sugars from the mash can drop.

Since the specialty malts can contribute far less in the way of sugars but contribute more to specific flavors (toffee, dried fruit, cracker, toasted bread, roasted coffee, etc, etc…), I can see the point of increasing them — just not increasing them as much as the base to keep them in check.

So, the original percentage contribution of sugars from the base can be less due to an efficiency drop, while the flavors derived from the ‘character malts’ increases by whatever weight adjustment was made. IOW, potential is there for adding more flavor percentage to possibly less sugar percentage and could cause an imbalance.

I believe this is the reason for the caution in BCS for an ‘over the top’ beer.  I also believe this is the reason some brewers use malt extract or simple sugars post mash if all the grains were increased proportionally — to bring balance back into the beer by adding sugars to balance the increased flavor from character malts.

Of course, some people like their coffee strong. Others don’t. The OP has to figure out which direction his taste buds prefer.

[denny]

Don't you also get more flavor from the base malts, not just sugars?

[BrewBama]

I’m sure there are flavor contributions from increasing base malts as well.  However, in a Stout the more intense flavors from roast malts would probably over power the more subdued flavors from base malts.

My point is the potential imbalance between sugar and flavors if not taken into consideration by either 1) not increasing those intense flavors as much as the base or 2) adding more sugars post mash to balance out the intense flavors if all grains are increased proportionally.

As in business: "at scale, things break"

Just food for thought.

[denny]

I’m sure there are flavor contributions from increasing base malts as well.  However, in a Stout the more intense flavors from roast malts would probably over power the more subdued flavors from base malts.

My point is the potential imbalance between sugar and flavors if not taken into consideration by either 1) not increasing those intense flavors as much as the base or 2) adding more sugars post mash to balance out the intense flavors if all grains are increased proportionally.

Just food for thought.

Interesting theory

[BrewingRover]

I've had Old Rasputin on nitro, it was very smooth.

[dbeechum]

I guess like a great many things - I hedge my bets on this front. I start by increasing everything linearly when looking at the calculator and then work my way to what looks right to me from an overall perspective.

E.g. if the base recipe is like 85% base, 15% speciality, I'll run that same ratio and look at the proposed mash bill. If something has slid "out of whack" like too much roast for my preference, then I'll cut that back and sub in base malt. Then I'll think about the likely final taste and adjust from there.

It's a little mathy, a little timey-wimey hand wavy - just like any good physics.

(And if you ever want to prove to yourself how much a base malt can bring in a big beer - just do a Bass #1 inspired barleywine - aka 100% base malt)

[Drewch]

Yet another option:  You could just boil longer to increase the OG.

So far, we've identified four COAs for making your beer imperial:

1. Increase base malt.
2. Increase the entire grain bill.
3. Add sugar.
4. Boil longer.

4 is almost a duplicate of 2, but I'd bet a modest sum that they'd yield at least slightly different flavors.

Edit: and now that I think about it, 3 and 1 are also sort of just different means to the same end.

[dbeechum]

With the caveat that 3 will lend a seemingly thinner/drier thing to the beer as opposed to the extra malt.

[erockrph]

If I tweak a recipe, I usually go the other direction (scaling down a recipe to get it to the 4% ballpark), but I scale small beers up on occasion. I find that scaling all the ingredients by percentage gets in the ballpark, but depending on the recipe some of the specialty malts need some tweaking afterwards. Cara/Crystal malts are usually ok when scaling by percentage (although when scaling down, the beers can sometimes use a small increase), but certain roasted malts can be overpowering when scaling up or underwhelming when scaling down. I kind of go by feel on these, but I usually find myself splitting the difference between scaling by percentage versus just adjusting the base malt.

Example - if my original recipe calls for 8 oz of pale chocolate, and I am doubling the recipe to make an imperial version. Instead of using 16 ounces of pale chocolate in the new recipe, I'd probably start at 12 ounces or so.

[denny]

I've had Old Rasputin on nitro, it was very smooth.

It's easy to simulate that with a syringe. It's what Guinness used to do before nitro cans.

[ScallyWag]

About a year ago, I scaled up my best stout recipe to Imperialize it, everything proportionally.  It achieved 10.9% ABV, and though it finished at 1.012, it was still "too much" for what I had hoped, too concentrated.

Don't get me wrong, it was still a great beer - my wife says it's her favorite of all the ones I've made - but if I were to remake it, I'd reduce the Munich and the Pale Chocolate and the Brown malt, and up the base more.  Maybe hopping it a bit more might help. 

It wasn't too roasty, it was just too...  flavorful??  Not too sweet, but too rich I guess.  My base was Simpsons Golden Promise, so that could've been too much of a good thing, might dilute that down with something lighter, like Best-Heidelberg if I have it.

Arguably at 10.9% it was Barleywine as much as it was Imperial Stout, and as it started hitting 6 months, 9 months, I was liking it more and more.  It's at about a year old now, and the last few bottles have been the best ones.  I have 2 bottles left, so I'm going to try to age it a bit more.

Point is, with this 1 beer data point, scaling it up proportionally didn't quite hit what I wanted from it.   Maybe it's a BIG beer thing.  I've only done 2 or 3 beers total at or above the 10%ABV range

[Kevin]

Or you could just find a good recipe for an imperial stout. Here are a few I got from a quick search of Ron Pattinson's blog. I've brewed both an they are both wonderful. The 1848 RIS is the real deal. That one is in one (or more) of his books but Brewer's Friend has the recipe. I believe I have also shared it in the recipe database at homebrewtalk dot com.

https://barclayperkins.blogspot.com/2019/08/lets-brew-wednesday-1914-courage.html

https://www.brewersfriend.com/homebrew/recipe/view/828618/1848-barclay-perkins-ibst-1st-imperial-stout

And this is not a stout but Pattinson listed this as an Imperial Mild. An English Mild that isn't dark... isn't low alcohol... and isn't lightly hopped. This monster brewed by Lovibond brewery in the mid 1800's.

https://barclayperkins.blogspot.com/2019/05/lets-brew-1864-lovibond-xxxx.html