Yeast Starter Stepping Calculator

[ajfranke]

This past week I was met with an inevitable challenge: I wanted to brew a Northern English Brown, but the only appropriate yeast that I had on hand was old. Its manufacture date in early April left it with an estimated viability of ~10%, according to mrmalty.com. Obviously, I would need to make a starter at the least, and take that starter through a few steps at best.

Not yet having a copy of Yeast by Jamil Z. and Chris W., the best source of information I could find on starter stepping was BillyBroas' well-crafted and informative post and video on the topic. He uses tabulated data from Yeast to calculate what a starter's estimated final cell count will be based on starter size and initial cell count. Unfortunately, the cell counts that he addresses start at 100B, and I had only ~10B on hand. While Mr. Malty is good for estimating total required yeast, I find that it becomes cumbersome with very low initial cell counts.

Thus, I decided to automate BillyBroas' method in an Excel sheet, and add some math (linear interpolation) to accommodate different starter sizes and varying cell counts.

Starter Stepping Caclulator

Please feel free to use it, and give any feedback. Of course, the numbers that it produces are just estimates, and YMMV. The interface is still a bit lacking of polish, but it seems to get the job done.

Citation: data from Data from Chris White and Jamil Zainasheff, Yeast: The Practical Guide to Beer Fermentation. Boulder: Brewers Publications, 2009.

 

[goldmind31]

An online version of this spreadsheet would be nice. Especially as part of the Mr.Malty Calculator. You should contact them and make it happen. I know I would use it.

 

[thegerm]

I like what you've done here.

 

[Clementine]

I like the concept of what you are trying to do, I read this while I was at work yesterday but did not have time to reply, sorry for the late response. I struggled when I started slanting however from my reading I have some points of view you might find helpful but in the end they are just my opinions. I now have a system that works for me and I have confidence in it and in the end that is the key is that your system works for you. Anyway my two cents worth:

1) Most people who are doing stepped starters with low cell count are using slants or glycerol frozen banks or some other storage method and to do significant cell count generation requires multiple steps and a stir plate is a major advantage. So any stepped starter calculator that does not accommodate figures for stir plates is of limited use.

2) Eventual cell count in a starter is limited by three things, O2, sugar and nutrient levels provided you stay within recommended step sizes. So when making starters if you account for Oxygenation method you can approximate your cell count just from starter volume for a given SG and tested nutrient level. This the basis of Mr Malty and the Maltose Falcon article, using a set wort material and different aeration methods they have conducted cell counts to test the difference.

So to calculate how I have to do my starters (from slants) I figure out the eventual number of cells required to make my beer. Using the (1-OG*1000)/48*pitch rate * ml of wort formula. Then I translate this into size of the final starter required to make that number of cells (on a stir plate I calculate 200 Million Cells per ml of starter). Then I know my starting step from a slant is 10mls and I step up 5 fold to reach the finial volume described above. I also then cross reference this with a calculation made from using volume of yeast solids of a cake to double check I have enough yeast. I want to start doing some tests with weighing yeast that I saw another member (Cidahmastah) using and now another guy in NZ (Honeybadger) is trying but I don't have data to add on that yet. I made a spreadsheet on my basic method to help with these calculations but I don't know how you posted your the file attachment here does not work for .xls or .xlsx files

Clem

 

[ajfranke]

1) Most people who are doing stepped starters with low cell count are using slants or glycerol frozen banks or some other storage method and to do significant cell count generation requires multiple steps and a stir plate is a major advantage. So any stepped starter calculator that does not accommodate figures for stir plates is of limited use.

I tried to accommodate use of stir plates and oxygen with the "Using Oxygen?" pull-down menu and the "O2 Factor". In his video, Billy says that the use of either increases the cell growth in a starter by about a factor of two. My implementation isn't very precise, but the tool is meant to be for estimation purposes.

2) Eventual cell count in a starter is limited by three things, O2, sugar and nutrient levels provided you stay within recommended step sizes. So when making starters if you account for Oxygenation method you can approximate your cell count just from starter volume for a given SG and tested nutrient level. This the basis of Mr Malty and the Maltose Falcon article, using a set wort material and different aeration methods they have conducted cell counts to test the difference.

Perhaps future iterations of the spreadsheet will account for starter OG and nutrient additions. I would need to do more research, first.

I also then cross reference this with a calculation made from using volume of yeast solids of a cake to double check I have enough yeast. I want to start doing some tests with weighing yeast that I saw another member (Cidahmastah) using and now another guy in NZ (Honeybadger) is trying but I don't have data to add on that yet. I made a spreadsheet on my basic method to help with these calculations but I don't know how you posted your the file attachment here does not work for .xls or .xlsx files

I'm still very bad at estimating the sizes/compactness/trub fraction/etc. of yeast cakes, but that is certainly a valid method. The spreadsheet that I posted is a linked to hosted space on Dropbox.

 

[diS]

Great job.
Thanks!

 

[Phunhog]

Wow that is awesome!! I was trying to find something like that!

 

[ekjohns]

This is nice but without knowing the initial yeast count are you SOL?

 

[stratslinger]

Very cool looking tool! This should be super useful for any future big brews.

But ekjohns makes a good point - and it goes beyond just this particular calculator... If you're using a known quantity of yeast, and can estimate its viability, a tool like this is of great use. But if you wash your yeast and re-use it, how do you figure out starter sizes? If you use yeast slants, how do you figure out starter sizes? In either case, I don't see a way to know your starting cell count with any level of reliability.

 

[diS]

I am trying to figure how many steps I need to get appropriate pitching rate from 10ml of frozen yeast (1.5 B/ml).

I need ~230 B cells and I start with 300ml starter:

1st step- 300ml (26 B)
2nd step- 1000ml (58 B)
3rd step- 2000ml (175 B)

Even with 3 steps I cant get right pitching rate.
One solution that is on my mind is to repeat 2000ml step twice:

1st step- 300ml (26 B)
2nd step- 2000ml (78 B)
3rd step- 2000ml (233 B)

This way I am getting right cells number, but I am not sure about repeating steps..
What do you think about repeating same size step (if I decant 2000ml starter and repeat it at same size)?
.. or maybe someone have better idea how to solve it?

 

[diS]

At first sight I can say that it looks really great.
Thanks for sharing!

One question: why it says "Exceeded Max. Inoculation Rate" when I set 0.35 L or below for 1st step?

I dont have time to better introduce it right now since I am going to work, but I"m looking to test it as soon as I get more time.

 

[el_caro]

If it is correct in its calculations I think it is a wonderful effort.

 

[diS]

I am comparing results from these two calculators, and either I am doing something wrong or one of them is missing something.

Starting point is 35 ml White Labs vial with 100 B cells and production date at 7/20/2011.
Viability is 47% or 47 B cells. If we are making 1.3 L starter each calculator gives different number of finished cells: 116 B vs 185 B.
This difference increases even more if we are splitting this amount of yeast into 7 x 9 ml vials and then propagate it to pitching rate from that point.

In each vial we will have 16 B or 26 B yeast, and if we make 3 steps (0.3 L / 1 L / 2 L) from one vial, cells number at 3rd step differs considerably, 186 B vs 304 B.

Please correct me if I am missing something.

 

[ajfranke]

Here is my version of a stepped starter calculator, feedback is much appreciated. Thanks.


http://dl.dropbox.com/u/32177750/Yeast%20Cell%20Calcs.xlsx

Looks really nice. And I like that it includes the "max pitching rate for growth" consideration, which mine doesn't. Some useful stuff, coming out of this thread.

One "bug" report: I tried opening this spreadsheet on an OpenOffice.org 3.1.0 and the DATEDIF() function call in cell C16 was not recognized. The spreadsheet appeared to work when I replaced that call with a DAYS() call. I don't know if support for DATEDIF() has been added in later versions of O (currently at 3.3.0) or LibreOffice. In Excel 2003, DATEDIF() isn't supported (probably not a big problem since .xlsx files barely are) but there is a DAYS360() function.

 

[diS]

My 1st assumption that calculators are too good to be true (to avoid cell counting via density, microscope, weight..) unfortunately came true. I"m not saying that it surprises me but I was hoping to get +-10% results..
I am in the middle of White"s book and I finally realize that yeast is living organism and it doesn't care about our formulas and multiplier factors (although it would be nice). They will do what they want and we cannot 100% rely on our calculations, they are just guidelines to show us the way how they react in starter.

Right now I have 2 L starter on stir plate (2nd step from 0.3 L), I though to make another one with same wort amount but I think I"ll pass it and pitch this 2 L starter in my next batch.
By trials and errors I"ll find what best suits with me, and I have to cope with it.

Thanks again to both of you guys for sharing your works!!

 

[Sulli]

BTW diS, I would love to hear how your ferment goes. Please keep me posted.

 

[diS]

Sure, I planed to brew in next couple of days.
Starter is on stirplate for 12 hrs right now and I have one extra day (since I am skipping 3rd step). I"ll keep it on stirplate 24 hrs more, take OG reading and if everything goes nice put it in the fridge.
It is WLP300 and I am not planing to decant it due to low flocculation, I want every cell in my batch so I"m gonna just shake it and pitch it.

 

[theredben]

Edit: After re-reading the post a second time, it seems that the OP's idea was to start with commercial yeast packages, so this should instead be read as a warning to people using it as a calculator for their starters that begin with home cultures.


The problem with making an internet calculator like this is that there is no way to standardize the inital propogation step. There is no way for a homebrewer to determine the initial cell count regardless of inoculation method (loopfuls from a slant/portions of glycerol suspension) since the viability of the yeast culture is not known. Since this first step is always going to vary, the final cell numbers that people find with these calculators are just guesses.

It is not that I have anything against these kind of calculations (I sure don't perform viability tests on my yeast!). In fact, they can be quite reassuring to someone new to yeast culturing. It is just that everyone needs to determine their own "in-house" method that works for them. There are guidelines to be found on the internet, but you always need to tailor the results for your own processes.

For example, my steps are usually: 2 loopfuls from a slant ->10mL ->100mL ->500mL ->1500+ mL (depending on the OG of the beer being brewed). This is a process that works for me, and the actual cell count does not make any difference at all. On a homebrew scale the Quality Control that level that includes cell counts is usually overkill. Many breweries do not bother with routine cell counts, they use a known volume of yeast based on previous experience and just pitch based on that. While it may not produce "optimum" results - if it ain't broke don't fix it!

 

[diS]

BTW diS, I would love to hear how your ferment goes. Please keep me posted.

WLP300 is pitched yesterday and airlock is pretty active. I brewed hefeweizen with 55% wheat and 45% pilsner- single step infusion 60 mins @ 152F.
We"ll see how well will it attenuate, so far so good.

 

[ajfranke]

The problem with making an internet calculator like this is that there is no way to standardize the inital propogation step. There is no way for a homebrewer to determine the initial cell count regardless of inoculation method (loopfuls from a slant/portions of glycerol suspension) since the viability of the yeast culture is not known. Since this first step is always going to vary, the final cell numbers that people find with these calculators are just guesses.

The scientific term is "estimate" but I agree completely. Even when starting with commercially-packaged yeasts, the initial cell count and viability are only estimates. Mr. Malty may say "67% viable", but the calculator can't account for hot/cold shipping, bad QA, mishaps while pouring into the starter, etc. Hence the "numbers are estimates, and YMMV" disclaimer in my original post; you and others have given fantastic explanations of the reasons for why the calculator produces estimates.

Many breweries do not bother with routine cell counts, they use a known volume of yeast based on previous experience and just pitch based on that. While it may not produce "optimum" results - if it ain't broke don't fix it!

This is an important point that needed to be made. Thanks.

 

[Phunhog]

Even when you use Mr. Malty it is just a good estimate at best. Mr. Malty mentions that a WL vial can contain anywhere from 70-130 billion cells based on the yeast strain. It defaults to 100 billion as an "average". BUT your particular vial could contain 30% less or more yeast which obviously throw your calcs way off.