Different math on yeast calculator

Yes, I've noticed this big difference as well - I really dug into the equations since I have my own brewing spreadsheet and wanted to know how to incorporate a starter calculator into it. Mr. Malty (Jamil's calculator) seems to be the most popular, but you're right that the starters are enormous (with my 2L flasks it always required either multiple flasks or stepped starters)... I went with Kai's stir plate equation in writing my spreadsheet because it seemed more reasonable in terms of starter sizes - it also matches BeerSmith's yeast starter calculator very closely. Despite Kai's and BeerSmith's smaller starter requirements I've gotten full attenuation every time, so I believe you're ok using these.

 

Are you adjusting the Mr. Malty calculator for the correct settings? Mr. Malty's seems to error on the side of slightly more yeast. It also is much more adjustable. This means that, if not adjusted correctly, it may give you wildly different numbers. I've found it to be only slightly larger than BeerSmith most of the time.

 

In all honesty, neither of them are going to be accurate, but they can sort of give you a ballpark idea.

 

When I had a homemade stir plate I found Zainesheff's figures (typically lower) most closely approximated my results. When I bought a commercial stirstarter it tilted more toward Troester's results (higher). The commercial stir plate had a little more "ommph" going on . . . I concluded the extra action did improve results. This is based on my personal experience doing multi starters over a period of years and measuring the actual quantity of yeast produced.

The yeastcalculator site gives the option of using data based on two different researchers. Obviously the two models have different results which is probably to be expected in real world testing.

More anomalies: If you use Brewer's Friends yeast calculator it also has two different stir plate growth options (White/Braukaiser) which are both different from yc's Zain/Tro . . . I draw the same conclusion that different models were tweaked/measured such that the results are different again. They even have a pretty good discussion on the models used. The errors in all of the above are magnified at lower innoculation rates (i.e. large starters). In a small starter (high innoc rate) the four models are pretty similar.

Which one is correct? I believe @honkey hits the nail on the head . . . probably none are truly accurate. But after multi starters I've narrowed it all down to what produces the most accurate count for me, basically through a trail and error process. So let your entire fermentation process (including taste of beer) guide you to which sets of data points work best for your brew house.

For @VikeMan , which model do you use in Brewchipher?

 

Technically, neither. BrewCipher uses my own interpretation of the growth curves, which I suspect (but don't know) are the same growth curves Jamil used for his model. So BC's answers will generally be closer to Jamil's than Kai's. BC also uses some different assumptions about age related viability.

 

FWIW, I think that using a calculator at all is A LOT better than nothing. I feel that's necessary to say after having said the calculator won't be that accurate. What I would suggest if you want to get "proper" fermentation characteristics, is to start with the calculator and make adjustments as needed until your ales are hitting terminal gravity in 3-4 days or lagers in 8-12 days. The Mr Malty calculator doesn't take into account things like gravity of starter (I find 8 degrees plato to be the best gravity for starters at my brewery for the yeast strains I know well), temperature (I typically keep starters a bit warm... 24-26 degrees celsius), or individual yeast strain... some propagate faster than others. That said, I would start with the calculator results exactly as they say and fine tune from there to get the best results.

 

I made an excel spreadsheet that uses the Monod equation with growth characteristics I've gathered from a few journal articles concerning sacch. cerv., and the numbers come out almost exactly the same as the Braukaiser numbers, so I think that is the model they use. Its a nice, simple microbial growth model, and it factors in the OG of the starter, but it doesn't integrate a lag phase or alcohol inhibition very well, so I always shoot for a 15% calculated over pitch.

That being said, every yeast strain will have slightly different growth characteristics, so you'll never EXACTLY know how much yeast you're pitching without a hemocytometer (which would be a little pricey and unnecessary for the homebrew scale).

 

Can you provide references for (or, better, links to) these journal articles? Cheers!

 

Fermentation Kinetics including product and substrate inhibitions plus biomass death: a mathematical analysis.
Mathieu Bouville
Institute of Materials Research and Engineering, Singapore (7 Feb 2008)

Growth Kinetics of Suspended Microbial Cells: From Single-Substrate-Controlled Growth to Mixed Substrate Kinetics
Karin Kovarova-Kovar and Thomas Egli
Microbiology and Molecular Biology Reviews, Sept. 1998, pg 646- 666

I also assumed perfectly spherical yeast cells with a diameter of 5 micrometers and a density of 1.1 g/cm^2.

Lots of assumptions are made in these calculations, so a 15-20% error would be safe.

I hope you like math!

 

Thanks!

 

I'm working on setting up a small yeast propagation room in my house for Brett's and sour beers for work. I'll try to remember to do a few sacch starters with various yeast strains and share the results. I'll be using a hemocytometer for cell counts.