Beersmith Color Estimates

[wsmith1625]

I use Beersmith web to design all my recipes and have everything dialed in pretty well. Most estimated values are pretty accurate including Mash temp and SG/FG. The one estimate that hasn't been accurate is color. All my beers are lighter than estimated and I'm not sure why. Is there something in the software or my brewing process that I can do to improve this? I do 5.5-gallon BIAB batches with no sparge over a propane burner.

 

[VikeMan]

The one estimate that hasn't been accurate is color. All my beers are lighter than estimated and I'm not sure why. Is there something in the software or my brewing process that I can do to improve this? I do 5.5-gallon BIAB batches with no sparge over a propane burner.

It could be a few things. Here's one.

A dirty little secret about the standard beer SRM/MCU model (the Morey equation, etc.) is that it bases color contribution (Malt Color Units (MCUs)) strictly on the amounts of grain (and their lovibond colors) and the post-boil batch size. But imagine two different setups.

1) Setup #1 has no deadspace in the mash tun and uses (say) 8 gallons of water and 11 pounds of grain to make 5 postboil gallons of 1.060 wort.

2) Setup #2 has a one gallon non-recoverable deadspace, thus decreasing lauter efficiency. So now you need to use another gallon of water to make up the volume. And to hit the same gravity, you'll also need to use proportionally more grains. But the batch size is still 5 gallons.

The standard model will say that batch in setup #2 will be darker that the batch in setup #1, simply because there are more grains and the same final volume of wort. But the model doesn't know anything about that gallon of unrecovered wort that retains a portion of the color extracted (about 11% of it in this case). And for that matter, for setup #1, it doesn't know anything about how much wort has been absorbed/blocked by the grains, which will vary based on process.

 

[hotbeer]

If your beers have stuff suspended in them then the light going through it will bounce off those proteins and stuff making it look lighter than if everything had dropped out of suspension. So if you are intentionally making a hazy, then it's going to glow and look lighter than it really is.

 

[wsmith1625]

It could be a few things. Here's one.

A dirty little secret about the standard beer SRM/MCU model (the Morey equation, etc.) is that it bases color contribution (Malt Color Units (MCUs)) strictly on the amounts of grain (and their lovibond colors) and the post-boil batch size. But imagine two different setups.

1) Setup #1 has no deadspace in the mash tun and uses (say) 8 gallons of water and 11 pounds of grain to make 5 postboil gallons of 1.060 wort.

2) Setup #2 has a one gallon non-recoverable deadspace, thus decreasing lauter efficiency. So now you need to use another gallon of water to make up the volume. And to hit the same gravity, you'll also need to use proportionally more grains. But the batch size is still 5 gallons.

The standard model will say that batch in setup #2 will be darker that the batch in setup #1, simply because there are more grains and the same final volume of wort. But the model doesn't know anything about that gallon of unrecovered wort that retains a portion of the color extracted (about 11% of it in this case). And for that matter, for setup #1, it doesn't know anything about how much wort has been absorbed/blocked by the grains, which will vary based on process.

If you entered your deadspace into Beersmith properly, wouldn't all things be equal if both scenarios had the same efficiency?

 

[wsmith1625]

If your beers have stuff suspended in them then the light going through it will bounce off those proteins and stuff making it look lighter than if everything had dropped out of suspension. So if you are intentionally making a hazy, then it's going to glow and look lighter than it really is.

I'm asking because this is something I see in all my finished beers. For most pale ales and IPAs, it's not a problem for me. When trying to brew darker beer styles, I might come in lighter that the BJCP guidelines.

 

[VikeMan]

If you entered your deadspace into Beersmith properly, wouldn't all things be equal if both scenarios had the same efficiency?

I don't use BeerSmith, but to the best of my knowlege, BeerSmith uses the standard SRM/MCU formula.

If you enter your deadspace properly, BeerSmith will tell you that you need more water, as it should. But the fact that the deadspace exists means that you cannot have the same actual lauter/mash/brewhouse efficiency as if there were no deadspace. Thus you need more grains in scenario #2 in post #2, and thus a problem with the standard SRM/MCU model.

 

[Bramling Cross]

Nailing color on the first try is really tricky—ask anyone that has tried to brew an unambiguously red beer. You’ve really, really gotta nail your efficiency and volumes to pull off a red beer. To do it consistently is one of the most difficult tasks in brewing.

Why am I writing about red beers? They’re an extreme example of the importance that accurate volume and efficiency numbers play in achieving your target color.

Since you’re consistently lighter than your target, there’s a chance that you’re either starting with too large of a pre-boil volume or you’re not boiling off enough and winding up with a too large post-boil volume.

Really, this is just expanding upon what VikeMan was discussing above, but while he was looking at the mash tun, I’m looking at the kettle. Either could be right, but it’s more likely that both of our explanations are in play.

There’s a lot more to it than that, but I’ve been behind the wheel all day, I have a big Mexican dinner in my belly and I’m tired. I’m sure others will be willing to pitch in

Hope this was somewhat helpful.

 

[wsmith1625]

Nailing color on the first try is really tricky—ask anyone that has tried to brew an unambiguously red beer. You’ve really, really gotta nail your efficiency and volumes to pull off a red beer. To do it consistently is one of the most difficult tasks in brewing.

Why am I writing about red beers? They’re an extreme example of the importance that accurate volume and efficiency numbers play in achieving your target color.

Since you’re consistently lighter than your target, there’s a chance that you’re either starting with too large of a pre-boil volume or you’re not boiling off enough and winding up with a too large post-boil volume.

Really, this is just expanding upon what VikeMan was discussing above, but while he was looking at the mash tun, I’m looking at the kettle. Either could be right, but it’s more likely that both of our explanations are in play.

There’s a lot more to it than that, but I’ve been behind the wheel all day, I have a big Mexican dinner in my belly and I’m tired. I’m sure others will be willing to pitch in

Hope this was somewhat helpful.

Thanks for the added info. I'll scrutinize my hot side and see if I can dial in my numbers a bit. My next brew is going to be an Irish Red, so it's a good test. I'm using the AHA recipe British Mats.

 

[wsmith1625]

@VikeMan and @Bramling Cross got me thinking about my hot side and I'm starting to lean towards PH. I have a PH meter but I don't use it very often. Also, I target 5.2 for all my beers and was reading that you could go higher for darker beers.

 

[Bramling Cross]

Yeah, I’ve learned that it really helps to be around pH 5.5-5.6 for properly dark beers that emphasize the roasty stuff.

 

[jdauria]

I wonder if the fact that BeerSmith lists grain color in SRM instead of the widely accepted Lovibond has anything to do with it? For example, they list CaraMunich III at 71 SRM, which using a color converter calculator is, rounding up, only 53 Lovibond. If BeerSmith's formula for wort color is using 71 SRM = 71 Lovibond, then it's obviously going to calculate a darker color beer than you would wind up with, especially for beers outside the yellow to pale range. Or is their formula converting grain SRM to Lovibond first, and then calculating the beer color. That may be a question to ask on their forums.

I will note that I use Beersmith also and have noticed that when I make dark beers, usually Schwarzbier and Czech Darks, they do tend to be lighter than I expected. Since I send beers to competitions, I also make sure to have some Sinimar on hand to use to darken a beer post fermentation.

 

[day_trippr]

fwiw, Beersmith3 allows the use of either SRM or EBC.
I use Beersmith3 and Bru'n Water, and I get comparatively close color numbers out of them for recipes...

Cheers!

 

[Oginme]

I wonder if the fact that BeerSmith lists grain color in SRM instead of the widely accepted Lovibond has anything to do with it? For example, they list CaraMunich III at 71 SRM, which using a color converter calculator is, rounding up, only 53 Lovibond. If BeerSmith's formula for wort color is using 71 SRM = 71 Lovibond, then it's obviously going to calculate a darker color beer than you would wind up with, especially for beers outside the yellow to pale range. Or is their formula converting grain SRM to Lovibond first, and then calculating the beer color. That may be a question to ask on their forums.

I will note that I use Beersmith also and have noticed that when I make dark beers, usually Schwarzbier and Czech Darks, they do tend to be lighter than I expected. Since I send beers to competitions, I also make sure to have some Sinimar on hand to use to darken a beer post fermentation.

This has long been an issue in the software property label. The numbers actually reflect the Lovibond value reported by the manufacturers. The program has it labeled as SRM, but Brad has stated that the correct labeling should be Lovibond. Inputting SRM numbers into the program really messes up the color calculations significantly (I'll never do that again -- takes too much time reediting all the entries).

 

[VikeMan]

Assigning "SRM" to a beer ingredient is a recipe for disaster. I don't know why anyone (e.g. Briess or Beersmith) continues to do it. SRM is measured by sending a 430 nm light through a cm of beer and measuring the light's strength on the back end. You can't do that with a malt. You can make a wort/beer from that malt and then measure the SRM of the wort/beer, but the answer "as is" will be useless for predicting the beer color of a recipe, given that brewing software expects Lovibond.

A similar issue in this mess is the range of syrups from Candi Syrup. They are labelled with SRM values, which in the case of Candi Syrup is not Lovibond mislabeled as SRM, but is indeed the syrup's tested SRM (using the 430 nm light method, as if the syrup itself were beer). The problem with that is that people tend to take that SRM value and enter it unchanged as a Lovibond value (wrong answer) or they use an online "SRM to Lovibond Calculator" to come up with a Lovibond value to enter (also a wrong answer). Rather than go further down that rabbit hole here, I'll link this. Candi Syrup Colors - Probably Not What You Think They Are

 

[Peebee]

There is an old battle (early 1990s) with beer colour started by a chap called Dr George Fix (deceased). He noticed beer colour followed a curved line when diluted (got disproportionately lighter). This was fiercely opposed by those following laws of physics that pointed to a proportional change of colour (straight-line). The battlelines were drawn, roughly down the Atlantic, between those supporting Dr Fix (States side) and those supporting the laws of physics (European side). There was to be no prisoners.

As the States took a more prominent role in writing homebrewing texts the Dr Fix supporters started winning.

Unfortunately, so fierce were the arguments, compromise didn't get a look in! American "calculators" often estimated beer colours too light ... and European (or is it just UK?) estimate beer colours way too dark.

Yet if you follow the laws of physics, you can pick out that colour change with dilution was only described for simple or single compounds. And follow Dr Fix you find only limited data resulting in "one curve fits all". Amazing what compromise can do! I find when working with different malt colours Beersmith (the American calculator I use) is fine with its estimates but falls apart when dealing with less complex additions such as dark sugars (predicts too light).

I'm not providing a solution! Maybe the calculations are correcting some of the inflexible stances of old? I'm only suggesting the calculators might be too inflexible at present and may need the backup of personal experience.