Rager or Tinseth? which formula is better? what are the differences?
Rager or Tinseth??
- Thread starter sgraham602
- Start date
Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:
I don't find either particularly useful. My club just did a barley wine which, by the Tinseth numbers with the parameters adjusted to best fit what actually happens in my brewery and the markings on the hops bags should have produced 67 IBU actually came in at 37. This kind of variation (46%) is perhaps larger than usual but certainly 20 and 30% variations are not.
I use the Tinseth formula because it has the two parameters and because if you have actual bitterness data you can adjust those parameters to best fit your equipment/practices as I have done but until you start to get reliable alpha acid data on the hops you buy you can't really expect any model to give very good predictions.
I use the Tinseth formula because it has the two parameters and because if you have actual bitterness data you can adjust those parameters to best fit your equipment/practices as I have done but until you start to get reliable alpha acid data on the hops you buy you can't really expect any model to give very good predictions.
There rarely is a better or best in brewing...it's not a contest. It's just a matter of preference, all that is really important is to stick with one..... But in terms of accuracy, they're all accurate, you might think of it simply being that they're in different languages....as long as you stay consistant in using one over any other it will be right.
But in reality it's all arbitrary anyway...they're just numbers.
If you ever listened to Palmer's basic brewing interview shortly after he attended a professional conference on hops and brewing, where he admits he got it all screwed up, you'll realize none of it really matters...
But in reality it's all arbitrary anyway...they're just numbers.
If you ever listened to Palmer's basic brewing interview shortly after he attended a professional conference on hops and brewing, where he admits he got it all screwed up, you'll realize none of it really matters...
Warthaug
Well-Known Member
- Joined
- Apr 8, 2011
- Messages
- 624
- Reaction score
- 212
ASBC MOA (Method of Analysis) Beer 23A. The bittering principal is extracted into essentially very pure gasoline by putting a sample, some hydrochloric acid and the 'gas' in a tube and shaking vigorously. The clear organic phase is then transferred to a cuvet and the absorption (A) at 275 nm in 1 cm measured. IBU = 50*A
Tinseth is more accurate, but I find that many recipes are formulated via Rager. I find that if I calculate hop additions from an existing recipe and use Tinseth, the bitterness will be higher than expected.
In my opinion, it makes little difference which formula you use. But you have to calibrate your bittering expectations with the formula used. I've stuck with Rager just because that formula meets my bittering expectations in the finished beer.
In my opinion, it makes little difference which formula you use. But you have to calibrate your bittering expectations with the formula used. I've stuck with Rager just because that formula meets my bittering expectations in the finished beer.
I'm especially interested where the formulae fail in underestimation. I recently brewed a Lil Sumpin Sumpin (Lagunitas) clone where Rager estimated 40s, and Tinseth 30s, but Lagunitas claims 65, and Jamil translated hop additions straight from their brewers.
Beer color can be decently collapsed to one dimension like SRM. Limiting IBUs to hops and not other factors like minerals really 'dilutes' the value of the metric.
I realize experiments try to measure iso-alpha acid amounts, but I still don't have any clear rules of thumb of their stability or utilization from whirlpool, or just exactly how dry-hopping affects perceived bitterness...
If commercial breweries are looking elsewhere, I'm excited!
Beer color can be decently collapsed to one dimension like SRM. Limiting IBUs to hops and not other factors like minerals really 'dilutes' the value of the metric.
I realize experiments try to measure iso-alpha acid amounts, but I still don't have any clear rules of thumb of their stability or utilization from whirlpool, or just exactly how dry-hopping affects perceived bitterness...
If commercial breweries are looking elsewhere, I'm excited!
Kaiser
Well-Known Member
lockwom,
commecial brewers don't obsess as much about estimating IBUs as much as home brewers do. That's mainly because there are many factors that affect IBU's that are not captured by the IBU estimation formulas and because measuring IBUs and adjusting the recipe based on that measurement is a more reliable and practicable approach for home brewers. That's also why all these IBU estimation formulas have been developed by home brewers.
I agree with Martin's position, that you should stick with one formula and calibrate your bitterness expectations. In case of your Sumpin clone, adjust the hops until it tastes like the commercial example. If that means that your estimated IBUs will be 40 while the measured IBUs of the commercial beer is 65, then that's what it is. Next time you want to brew a beer with a similar bitterness, shoot for an IBU estimate of 40.
That's one of the first things I learned in home brewing when I brewed a SNPA clone, aimed for 39 IBUs and found that the beer was way too bitter.
Kai
commecial brewers don't obsess as much about estimating IBUs as much as home brewers do. That's mainly because there are many factors that affect IBU's that are not captured by the IBU estimation formulas and because measuring IBUs and adjusting the recipe based on that measurement is a more reliable and practicable approach for home brewers. That's also why all these IBU estimation formulas have been developed by home brewers.
I agree with Martin's position, that you should stick with one formula and calibrate your bitterness expectations. In case of your Sumpin clone, adjust the hops until it tastes like the commercial example. If that means that your estimated IBUs will be 40 while the measured IBUs of the commercial beer is 65, then that's what it is. Next time you want to brew a beer with a similar bitterness, shoot for an IBU estimate of 40.
That's one of the first things I learned in home brewing when I brewed a SNPA clone, aimed for 39 IBUs and found that the beer was way too bitter.
Kai
Indeed, I liked the position enough to bestow my first forum 'like'!
I hope to keg the LSS clone soon to see how far off the mark it is. But my next beer will be a MO/Cascade SMaSH, and I will use it to calibrate many things. Added to the list will now be an IBU formula preference.
Kaiser
Well-Known Member
It could either be a formula preference or a simple scaling factor. E.g. if you want the bitterness of a commercial beer with Y IBU, aim for X = 80% * Y.
Kai
I don't think it's the formulae. I think it's that you don't have good data to put into the formulae.
I'd be interested in hearing how you came to that conclusion. The statement is perfectly true but not many people are aware of this (AFAIK).
I offer without comment the first paragraph of the relevant MOA:
"Reports of the Subcommittee on Determination of
Isohumulones in Beer for 1967 and 1968 (Ref. 1) indicate
that bitterness units (BU), as determined in Method
A below, express the bitter flavor of beer satisfactorily,
regardless of whether the beer was made with fresh or
old hops. The European Brewery Convention has
adopted the E.B.C. Bitterness Units, determined in a
similar way, as a uniform method that best expresses the
true bitter flavor value of beer."
Most of what I've read about SRM/Lovibond was in Daniels's book. I didn't feel like I knew enough on the subject to unilaterally claim the measure's perfection, but I am happy to learn.
As for IBU's intent vs my criticism: I respect the measure's repeatability, but ultimately bitterness is perceived. So my claim relates more to the fact that a heavily hopped recipe made with Burton style water would taste different with purer water like Pilsen's.
Kai, I really respect your work, especially your awareness of German homebrew techniques. Do they use different approaches?
There are some parallels and some differences in color and bitterness. Both are, of course, perceptions and both are expressed with a single number and the single number, while it does give some indication as to what one will experience is far from the whole story. To convert SRM (which represents the optical absorption of the beer at one wavelength i.e. a repeatable physical measurement) to a perceived color one must convert the SRM to the absorbtion spectrum of the beer. Noting that a kriek and a lambic may have the same SRM it is clear that the SRM alone doesn't tell the whole story. A couple of 'spectral deviation coefficients' complete the spectrum picture at which point we have to consider the color of the illuminating light, the width of the glass....
With bitterness the IBU number, also based on a physical mesurement - the concentration of isolhumulone - gives a picture of bitterness but not a complete one. There is the fine bitterness of Saaz and the coarse bitterness of the high alpha cultivars. And then the differences in perception that come, again along the fine/coarse 'axis' with respect to the presence of sulfate.
The big difference between the two is that there is lots of science available to describe the spectral difference between kriek and brown ale of the same SRM and how to convert SRM plus SDCs (spectral deviation coefficients) to visible color taking viewing light quality and path into account. Similar science does not, AFAIK, exist for bitterness but there are people working on it.
I guess I feel that beer color, IBU readings and pH should be used in the same way as OG, FG, TE, ABV and attenuation numbers and that is as checks that you made the beer you were trying to make. If all those numbers come out close to what you have obtained other times you brewed the same beer then there is a pretty good chance that this beer will meet your expectations. If one or more of them is way off there is a pretty good chance that it won't.
As most home brewers and small craft brewers do not measure any of those parameters except OG and FG the value of calculating IBU and color based on formulae seems to have limited value as the formulae for neither are particularly good predictors and as I have noted before I think that is not so much because the formulae are wrong as it is that you don't have good data available to plug into them. FWIW over an ensemble of 23 beers with IBU's ranging from 12 to 36 I found rms error of 5.1 IBU (including a bias of 2.2 IBU) from the Tinseth formula with the time dependence parameter adjusted for minumum IBU (not time and OG parameters as I indicated in an earlier post). Those of you familiar with error analysis will be saying to yourselves 'Why didn't he back the bias out?' I could, of course, do that but it would add another parameter to the formula, I don't, based on 23 beers, have much confidence that 2.2 is the right bias number and if I do back it out of these beers it only drops the rmse to 4.6 IBU.
With bitterness the IBU number, also based on a physical mesurement - the concentration of isolhumulone - gives a picture of bitterness but not a complete one. There is the fine bitterness of Saaz and the coarse bitterness of the high alpha cultivars. And then the differences in perception that come, again along the fine/coarse 'axis' with respect to the presence of sulfate.
The big difference between the two is that there is lots of science available to describe the spectral difference between kriek and brown ale of the same SRM and how to convert SRM plus SDCs (spectral deviation coefficients) to visible color taking viewing light quality and path into account. Similar science does not, AFAIK, exist for bitterness but there are people working on it.
I guess I feel that beer color, IBU readings and pH should be used in the same way as OG, FG, TE, ABV and attenuation numbers and that is as checks that you made the beer you were trying to make. If all those numbers come out close to what you have obtained other times you brewed the same beer then there is a pretty good chance that this beer will meet your expectations. If one or more of them is way off there is a pretty good chance that it won't.
As most home brewers and small craft brewers do not measure any of those parameters except OG and FG the value of calculating IBU and color based on formulae seems to have limited value as the formulae for neither are particularly good predictors and as I have noted before I think that is not so much because the formulae are wrong as it is that you don't have good data available to plug into them. FWIW over an ensemble of 23 beers with IBU's ranging from 12 to 36 I found rms error of 5.1 IBU (including a bias of 2.2 IBU) from the Tinseth formula with the time dependence parameter adjusted for minumum IBU (not time and OG parameters as I indicated in an earlier post). Those of you familiar with error analysis will be saying to yourselves 'Why didn't he back the bias out?' I could, of course, do that but it would add another parameter to the formula, I don't, based on 23 beers, have much confidence that 2.2 is the right bias number and if I do back it out of these beers it only drops the rmse to 4.6 IBU.
Similar threads
