Regarding specialty malts and scaling recipes for mash efficiencies

[pricelessbrewing]

I've seen two schools of thought on this topic, both citing well regarded books and defending their point of view vehemently. Thought I'd ask you good folks your take on the topic.

The first states that all grains should be adjusted when scaling a recipe, not just base malts, adjusting the total weight of the grain bill to hit the intended OG, while keeping the % of grain bill the same. This is how I usually see it treated in brewing software, brewing texts, and recipe databases.

However the second states that you should only adjust the base malts to hit your intended OG, while keeping the specialty malts the same lb per gallon (or kg/L) ratio. They say that neither the conversion or mash efficiency matters, and that if you scale the specialty malt with the mash efficiency, you're changing the flavor contribution and so your flavor profile will change even though you may hit the intended OG.

Hypothetical Example: basic dry stout recipe. OG 1.038. 70% mash efficiency. 5G batch size.

5 lb 2 row. 62.5%
2 lb flaked barley 25%
1 lb black barley. 12.5%

Now lets say I expect to hit 89% mash efficiency (easily doable for such a small brew). Should I

A) Keep the flaked barley and black barley at 25% and 12.5% and scale everything down to hit the intended OG?

Or

B) Keep the flaked barley at 2 lb, and black barley at 1lb, and only adjust the 2 row down to hit the intended OG?

for another discussion, where someone defends school of thought B aka "adjust base malt only". Click this link to a reddit discussion

Obviously it's not a huge issue for most folks, but often times it's basically left to "well we don't know, you just have to figure out your system." And that's all well and good, but that's not how progress to understanding is made.

Is it related to conversion, where a finer crush will give slightly more roast flavors for black barley for example? To the strike volume (lb per gallon)? To the batch size (lb per gallon)? or to the mash efficiency?

 

[BeardedBrews]

Sounds like a job for @brulosopher !

 

[pricelessbrewing]

It wouldn't be pretty difficult to do in a way that team brulosophy usually carries out their experiments by isolating a single variable in question.

You'd have to manipulate multiple variables in order to change your mash efficiency enough to make a meaningful difference. For example, using my setup I can adjust the grain absorption rate by draining the mash tun and not squeezing the bag, and additionally doing a sparge vs no sparge. In my example of the 8lb dry irish stout, that would change the mash efficiency from ~74% to ~89%. This should be enough to give some idea of it. However it's not really something Marshall and team would do as it changes multiple variables to determine a single effect.

 

[Oginme]

When I have scaled recipes, either between my two systems or from a published recipe to my system, I find that the major change in ingredients happens in the base malts. There may be some minor tweaking of the specialty grains to maintain color, but the composition is more linear to batch size than efficiency.

 

[Yooper]

I always have scaled my ingredients, all of them.

I've never seen a scientific reason not to.

I mean, whether your efficiency is 50% or 90%, using 10% of caramel 10L (for example) is still 10%. I don't understand the rationale behind not doing so, really.

 

[chickypad]

My thinking (not based on actual evidence) is that the mash yield for the majority of specialty malts and adjuncts is going to be affected by mash efficiency too. I'm talking all the alternative base malts like munich, vienna, any unmalted grains, things like biscuit/victory, and even crystal malts if you read through nilo's experiements. So it makes sense to just scale all these by percentages to me. The only thing I sometimes adjust are the very dark roasted malts. I'm not sure that the yield from roasted malts is all that dependent on efficiency, in fact seems to be similar between mashing and steeping. I have on occasion adjusted a recipe for my system and noted that it looked way too heavy on the roasted malts.

In practice I think most of the time it probably doesn't matter too much. If you run the average recipe converting from efficiencies in say the 70-80% range there's not a huge difference between the results from the 2 methods. Where it becomes an issue is if the recipe is written for a very high or very low efficiency (or for brewers who get much higher or lower than the average).

 

[dmtaylor]

Sounds like a job for @brulosopher !

I's thinking the same thing!!

 

[chickypad]

It wouldn't be pretty difficult to do in a way that team brulosophy usually carries out their experiments by isolating a single variable in question.

You'd have to manipulate multiple variables in order to change your mash efficiency enough to make a meaningful difference. For example, using my setup I can adjust the grain absorption rate by draining the mash tun and not squeezing the bag, and additionally doing a sparge vs no sparge. In my example of the 8lb dry irish stout, that would change the mash efficiency from ~74% to ~89%. This should be enough to give some idea of it. However it's not really something Marshall and team would do as it changes multiple variables to determine a single effect.

Totally agree. I've seen folks report differences in body, depth of flavor, just from doing full volume mash vs. sparge so any manipulation of mash thickness, etc. is going to affect results. Maybe manipulating only crush would be a way to get close?

 

[IslandLizard]

When scaling for conversion efficiency, only adjust the base malts, and adjuncts that need to be converted.

When scaling because of brew house efficiency, such as mash tun dead space etc. racking losses, etc., scale all malts and adjuncts proportionally.

So if you scale for both, you need to treat each respectively, and make an educated guess, what goes where. Keep an eye on color too.

 

[Yooper]

When scaling for conversion efficiency, only adjust the base malts, and adjuncts that need to be converted.

But why? That's what I don't understand. If I'm using 10% caramel malt in one recipe, why would I decrease it and use 8% if my efficiency was higher? To me, that doesn't make any sense. You'd get the same % conversion with the specialty grains, so I don't understand why it'd be reduced.

 

[BeardedBrews]

The question could be simplified given the scope of the problem.

Your example at 5g with 70% Brew-House gives me 1.039 og

Scaled to 80% Brew-House you could run

4.1# 2-Row (57.7)
2# Flaked (28.2)
1# Black (14.1)

Or

4.40# 2-Row (62.5)
1.76# Flaked (25)
.88# Black (12.5)

A starting point would be to simply brew these two beers and see if you can readily distinguish between them. If you can't tell the batches apart, then that answers most of the question. (I suspect there will be a difference)

If you can tell them apart, then you'll have to make the original beer with the target efficiency to see which beer is closer. I agree that mill settings might be the best way to keep everything else constant.

 

[IslandLizard]

But why? That's what I don't understand. If I'm using 10% caramel malt in one recipe, why would I decrease it and use 8% if my efficiency was higher? To me, that doesn't make any sense. You'd get the same % conversion with the specialty grains, so I don't understand why it'd be reduced.

Why? Very good question!

Back to the OP's example:

5 lb 2 row 62.5%
2 lb flaked barley 25%
1 lb black barley 12.5%​

These are grist percentages, not extract percentages. To replicate the beer exactly, you need to keep the extracted ingredients the same.*

IF the original recipe was correctly formulated for 70% mash efficiency AND you want to replicate the recipe exactly with your 80% mash efficiency, you need to use less base malt. This keeps the ratio (%) of extract from the specialty/roasted malts to the extract of fermentables in the grist the same.

So when we convert the recipe from 70% to 80% mash efficiency we get the following:

5 lb 2 row (62.5%)        ==> (70/80)*5 lb     = [B]4.38 lb 2 row (61.4%)[/B]
2 lb flaked barley (25%)  ==> (70/80)*2 lb     = [B]1.75 lb flaked barley (24.6%)[/B]
1 lb black barley (12.5%) ==> [no mash change] = [B]1 lb black barley (14.0%)[/B]

* This assumes other process variables to be the same and extraction from specialty/roasted malts to be identical in either the 70% or 80% efficient mash. In other words, in the same amount of water you'll have the same amount of color and flavor contribution from the same amount (weight) of those specialty/roasted malts.

For base malts that also add significant color and flavor, such as Munich, Vienna, Biscuit, etc., I would follow the specialty/roasted calculations, leaving the amounts the same, unless they form the majority of the grist. In that case using 5% to 20% more [Edit] of each maybe needed to prevent losing flavor and color. That needs some judgment.

 

[doug293cz]

I think it might make sense to scale the base malts vs. mash efficiency, and non-convertable malts vs. lauter efficiency. Base malt yield will be affected by both conversion efficiency and lauter efficiency, but non-convertable malts will only be affected by lauter efficiency.

Brew on

 

[chickypad]

I think it might make sense to scale the base malts vs. mash efficiency, and non-convertable malts vs. lauter efficiency. Base malt yield will be affected by both conversion efficiency and lauter efficiency, but non-convertable malts will only be affected by lauter efficiency.

What would be the non-convertable malts? The only malts I can think of that I'm not for sure yield additional sugars in the mash are dark roasted malts. Even the contribution of crystal malts is increased by mashing so seems like they would be subject to mash efficiency.

 

[Yooper]

What would be the non-convertable malts? The only malts I can think of that I'm not for sure yield additional sugars in the mash are dark roasted malts. Even the contribution of crystal malts is increased by mashing so seems like they would be subject to mash efficiency.

That is my understanding as well. It's possible that the dark roasted malts may be different, but perhaps not enough to matter. But I don't think there is a huge conversion difference between Vienna malt vs victory malt.

 

[Yooper]

I think it might make sense to scale the base malts vs. mash efficiency, and non-convertable malts vs. lauter efficiency. Base malt yield will be affected by both conversion efficiency and lauter efficiency, but non-convertable malts will only be affected by lauter efficiency.

Brew on

What are "non-convertable malts"? I've not heard that expression before.

 

[slym2none]

I have to scale both 5G and 5.5G batches down to 2.5G all the time. I just keep the grist percentages the same. I can't see doing any extra math for something that has little-to-no affect on my end-product. I will be brewing up an Avery's "Ellie's Brown Ale" clone soon, using the grist percentages on Avery's website.

 

[doug293cz]

What would be the non-convertable malts? The only malts I can think of that I'm not for sure yield additional sugars in the mash are dark roasted malts. Even the contribution of crystal malts is increased by mashing so seems like they would be subject to mash efficiency.

I probably should have said pre-converted and non-convertable malts. My understanding of caramel/crystal malts is that they have had a majority of their starch converted to sugar as part of their processing. This is why they can be used as steeping grains, and don't need to be mashed to extract most of what is available from them. I understand that mashing these malts can often convert more sugar, but at that point it is a fraction (of total potential) of a fraction (of the grain bill) of a fraction (conversion efficiency), and error for ignoring the additional conversion should be negligible. So, maybe the rule should be any malt that is suitable for steeping should be scaled vs. the lauter efficiency.

Brew on

 

[Yooper]

I probably should have said pre-converted and non-convertable malts. My understanding of caramel/crystal malts is that they have had a majority of their starch converted to sugar as part of their processing. This is why they can be used as steeping grains, and don't need to be malted to extract most of what is available from them. I understand that mashing these malts can often convert more sugar, but at that point it is a fraction (of total potential) of a fraction (of the grain bill) of a fraction (conversion efficiency), and error for ignoring the additional conversion should be negligible. So, maybe the rule should be any malt that is suitable for steeping should be scaled vs. the lauter efficiency.

Brew on

Gotcha. But then, the next question is "why"?

I mean- if all that is true, why is not 10% crystal 40L in a recipe the same no matter what?

If I give you the following recipe:

7% crystal 40L
93% pale malt

do you mean to say that if I brew it on my system, at say, 72% brewhouse efficiency (because I leave .5 gallon behind in the kettle), it would be different than 85% brewhouse efficiency on your system? And if so (which it seems to be what you are saying), why?

If we're talking conversion efficiency, perhaps, but that generally isn't at all what we talk about when we're talking about efficiency and converting recipes. If the conversion efficiency is less than 75%, there is a problem and it's not with the recipe.

 

[pricelessbrewing]

Why? Very good question!

Back to the OP's example:

5 lb 2 row 62.5%
2 lb flaked barley 25%
1 lb black barley 12.5%​

These are grist percentages, not extract percentages. To replicate the beer exactly, you need to keep the extracted ingredients the same.*

IF the original recipe was correctly formulated for 70% mash efficiency AND you want to replicate the recipe exactly with your 80% mash efficiency, you need to use less base malt. This keeps the ratio (%) of extract from the specialty/roasted malts to the extract of fermentables in the grist the same.

So when we convert the recipe from 70% to 80% mash efficiency we get the following:

5 lb 2 row (62.5%)        ==> (70/80)*5 lb     = [B]4.38 lb 2 row (61.4%)[/B]
2 lb flaked barley (25%)  ==> (70/80)*2 lb     = [B]1.75 lb flaked barley (24.6%)[/B]
1 lb black barley (12.5%) ==> [no mash change] = [B]1 lb black barley (14.0%)[/B]

* This assumes other process variables to be the same and extraction from specialty/roasted malts to be identical in either the 70% or 80% efficient mash. In other words, in the same amount of water you'll have the same amount of color and flavor contribution from the same amount (weight) of those specialty/roasted malts.

For base malts that also add significant color and flavor, such as Munich, Vienna, Biscuit, etc., I would follow the specialty/roasted calculations, leaving the amounts the same, unless they form the majority of the grist. In that case using 5% to 20% more [Edit] of each maybe needed to prevent losing flavor and color. That needs some judgment.

Ok but why did you not scale the black barley at all then?

Wouldn't ALL the grains and their corresponding extract vary equally with conversion and lauter efficiency (ie mash efficiency) so that would be equivalent to scaling all the grains for mash efficiency, which would maintain that the grist % is corresponding to extract %.

Lets ignore color contributions, as it's easy to add a little smooth high kiln malt like a carafa or midnight wheat or something as needed.

I have to scale both 5G and 5.5G batches down to 2.5G all the time. I just keep the grist percentages the same. I can't see doing any extra math for something that has little-to-no affect on my end-product. I will be brewing up an Avery's "Ellie's Brown Ale" clone soon, using the grist percentages on Avery's website.

Batch size is straight forward and simple. Never heard any issues there. And I've always done % scaling too. If you haven't noticed yet, extra math is kinda my thing.

Gotcha. But then, the next question is "why"?

I mean- if all that is true, why is not 10% crystal 40L in a recipe the same no matter what?

If I give you the following recipe:

7% crystal 40L
93% pale malt

do you mean to say that if I brew it on my system, at say, 72% brewhouse efficiency (because I leave .5 gallon behind in the kettle), it would be different than 85% brewhouse efficiency on your system? And if so (which it seems to be what you are saying), why?

If we're talking conversion efficiency, perhaps, but that generally isn't at all what we talk about when we're talking about efficiency and converting recipes. If the conversion efficiency is less than 75%, there is a problem and it's not with the recipe.

Re: Conversion (using Braukaisers definitions), if it's below 85% for any but a big brew, you probably have issues. Typically it's in the 90-95% range.

Re: Your recipe, definitely will be different at 72% and 85%. The topic is relating which the difference will be purely OG based (and have pretty much the flavor contributions from the crystal, ie flavor contributions from specialty malts are based on lb per gallon concentrations) or whether the 85% mash efficiency will have more flavor contributions from the crystal malt due
to the higher mash efficiency.

I've always believed the latter, and most people I talk to have always scaled their recipes that way also. But I'd really like to find out why, and if any testing has been done.

 

[IslandLizard]

Ok but why did you not scale the black barley at all then?

Wouldn't ALL the grains and their corresponding extract vary equally with conversion and lauter efficiency (ie mash efficiency) so that would be equivalent to scaling all the grains for mash efficiency, which would maintain that the grist % is corresponding to extract %.

Lets ignore color contributions, as it's easy to add a little smooth high kiln malt like a carafa or midnight wheat or something as needed.

I didn't change the black malt amount because it doesn't get converted. The mash water just extracts the sugars and color, and is not related to conversion efficiency. The color is expected to be spot on in the same volume of wort. Now I can see when the black malt was milled very coarse vs. very fine there will be extraction differences, but the premise was to keep everything else the same.

Lauter (and racking) efficiencies are across the board, affecting all malts evenly, so percentages remain the same. Simply adjust the amounts of each malt proportionally to reach the gravities at the recipe's volume or leave the amounts as they are and settle for less wort in the end.

Lautering losses from the mash tun or BIAB bag are more complex, as it really depends on sparging efficiencies. I for one don't believe pouring a couple gallons of water over a BIAB sack hanging over a bucket is good sparging. Neither is sparging with minimal volumes of water. However, adding decent amounts of sparge water to the tun or dunking the BIAB in a bucket followed by a good stir, will give you better (lautering) effciency. Following these practices, even with half a gallon of deadspace, the lautering losses after 2nd and definitely 3rd runnings are quite acceptable.

 

[pricelessbrewing]

Alright so you're stance is that lauter efficiency should apply to all grains, regardless of type. While unconverted grains should also be scaled according to conversion efficiency?

If that's true, then it's not much different separately scaling with lautering/conversion and just scaling with mash efficiency. Variations in conversion are rather small, typically like 4% or so at most, I doubt that would be a discernible difference in taste.

Re: Sparging. @Wilserbrewer stands by a simple pour over sparge, but I've never done any analysis on his brew days to see if he's getting any reasonable lautering efficiency increase from the sparge. I do a squeeze, dunk, stir, squeeze, combine style batch sparge.

 

[GrogNerd]

I thought it was scale everything for batch size AND for efficiency

but just base malt to convert to "session" or "imperial" versions

Grog.confused.now

 

[pricelessbrewing]

Grog.confused.now

Same man.

For now I'm going to continue to scale everything according to mash efficiency. So far I've seen little evidence to say otherwise, but I'm open to suggestions and additional reading material if anyone has any sources that are contrary.