• Welcome to the new forum! We upgraded our forum software with a host of new boards, capabilities and features. It is also more secure.
    Jump in and join the conversation! You can learn more about the upgrade and new features here.

Yeast Starter Calculator Accuracy

sockmonkey

Apprentice
Joined
Oct 5, 2011
Messages
3
Reaction score
0
Before BS2, I've always used mrmalty.com for calculating yeast starter size. Today I ran my next batch through both BS2 and mrmalty and the results for yeast starter size were significantly different.

Both BS2 and MM calculated the same number of cells needed, and the exact same viability. However, BS2 says I need a .81Qt starter and MM says I need a 2.12Qt starter (both using a stir-plate).

Any ideas on the huge discrepancy?
 
Hi,
  Hard to say without seeing all of the numbers but are both recommending the same number of yeast packets?  This can make a huge difference.

Brad
 
Yes, both are based on 1 yeast packet (WLP001)

1.082 SG, 5.5 US Gal, Ale yeast, 76% viablity, stir plate.

Yeast cells needed:  308B (MM)    308B (BS2)
Vials or packs w/starter:  1 (MM)    1 (BS2)
Recomm. starter size:  2.91qt (MM)  0.81qt (BS2)

All values in "Yeast Starter Settings" are default.
 
Hi,
  I know it is probably heresy to say so, but I think BeerSmith might be most accurate here.  I've been meaning to talk with Jamil about
this as I think he may be rolling in the stir plate correction factor too late in the calculation:

  Here's some of the math:
    - About 75 billion cells available from the yeast (based on viability)
    - Need 308 billion total for the batch
    - Growth factor needed is 308/75 which is 4.1 overall growth factor for the starter
    - The stir plate growth multiplier is 2.66 so we get 4.1/2.66 = 1.54 growth needed (equiv without stir plate)
    - To get a 1.54 growth factor, you need an innoculation rate of roughly 100 million cells/ml (from a table lookup) - its not a large growth
    - Dropping 75 billion cells into 0.81 quarts is 75 billion cells/0.77 liters which is very close to 100 million cells/ml
    - So as far as I can see the 0.81 quarts (say round it to a quart overall) is accurate if you have a good stir plate

  Honestly I could not match your numbers on Mr Malty without raising the growth factor way up - I think he is calculating the number of packs needed without factoring the stir plate in and then calculating the size of starter needed and factoring the stir plate in here.  To my way of thinking, the stir plate is a very important factor when estimating the number of packs of yeast needed, which is why I follow the outlined steps above.

Thanks,
Brad
 
Thanks Brad, that is helpful and I agree with your assessment (seeing the math always helps!)

Yes, I always move the growth factor slider until "1 pack or vial" is needed. As I think about it now, that doesn't seem logical, but that's the way I've been doing it for years. I guess this means I've been overpitching ever since I started using a stirplate!

 
Strictly from personal experience and knowing nothing of the inner workings of either calculator, I've been making 0.8 liter starters on a stir plate for quite a while.

I seldom have a smack-pack younger than 60 days, and the starters (ales) have always done the job well on 4-gallon worts between 1.055 and 1.065. 

 
Brad,

I have been digging into this myself and there may be an issue in Beersmith that needs to be investigated.

Using the Yeast Starter Tool for a 5.25 Gallon batch with an OG of 1.067 both BS2 and MrMalty calculate ~242 Billion Cells. 
Using Liquid Yeast packaged on 11/6/2011 both tools estimate ~67% Viability. 

Now when you get to the calculation for the size of the starter in BS2 without a stirplate the internal calculation of the number of packs needed without a starter appears to be incorrect.    With the above inputs BS2 suggests that 6 yeast packs are needed which it 2 times as much as is really needed

6 packs x 67 Billion cells / Pack = 402 Billion cells

The ideal pitching rate is 242 Billion cells

The calculation then suggests that 2 vials in a 3.09 Liter starter is required which is about 1 L more than necessary to achieve a 1.86 growth factor.

Would you take a look at the internal calculations in BS2?    If there is a conversion factor error or an incorrect table lookup that could impact the calculation of the starter size needed with a Stir Plate.

Cheers
 
I agree.  Something isn't working right.  or....I just cant figure out how to utilize beersmith with yeast ranching.

I've mentioned before, I farm my own yeast.  I freeze 15 ml of slurry with 15 ml of glycerol...roughly 18 billion cells per vial.  I pitch this in 100 ml of wort, then step to 500 ml, then to 1 l, then I will decant and add a second 1 l of wort to the same flask.  This gives me something like a 20:1 growth factor...limited by maximum cell densities in a 1 l flask (~220 billion cells).

So, I create inventory items for my "hampton labs" yeasts with all the relevant facts.  I change the "form" to culture, so that I can set a different age rate.  Then I use these inventory items in my recipes.

[list type=decimal]
[*]I add the yeast to my recipe, and beersmith calculates the viable cells at 14.4 B. It also tells me the number of cells needed (eg. 233 B).
[*]I use the "yeast packs to use if no starter" value as a guide for the amount of growth required.  I don't really use it for anything, but its nice to see how much growth is required.
[*]I don't really know what to do with the "recommended starter size".  The number doesn't seem right to me.  In my example, BS is saying 0.84 l.  But, I think 233 B cells is more than 1.0 l will support (density wise).  1.05 l seems closer.  I seem to be able to adjust this with "stir plate factor".  I've settled on 2.35, but that seems like a hack for BS overestimating the cell density that can be supported without a stirplate.
[*]Now, we get to "yeast cells with starter".  I can't get this number to make any sense to me.  No matter what I do, I can't get it to exceed 86.5 B in this example (a growth factor of 6.0).  I've set "growth factor limit" to 1000, so it stays out of my way.  I've tried changing the starter size, and the stir plate factor...both to no avail.
[/list] 

As I've worked through typing this, I think I've worked it out myself.  Growth factor seems to be internally limited to roughly 6.0.  For those of us who "ranch" this doesn't really work. I can't grow enough slurry to store 100 B cells per vial.  Well...I could...but I call that beer, not starter. 

The information I need is really this: 1) the optimal starter size for the initial step based on viability, and a desired pitching rate (~ 100M/ml).  2) The number of decant-step cycles required at my final volume since maximum concentration isn't generally reached when stepping from a smaller volume.  The intermediate steps are just mechanics depending on what step ratio I choose (2x, 4x, 10x, etc).

My point in (2) is that if I step from 100 ml to 1000 ml (10x step up), I will end up with about 100 M/ml density...or roughly half of maximum.  If I decant and add another 1000 ml wort, I can get to the maximum of ~220 M/ml ( or 220 B cells) without using a larger flask. 






 
Both calculators use a stir-plate multiplier of 2.66, but in different places in the equation.  BS2 divides the growth factor by the stir-plate multiplier to reach the growth factor used to calculate volume of the starter.  Mr. Malty calculates the volume of the starter using one package of yeast and divides that volume by 2.66 if you change the settings to "stir-plate."  Mr. Malty is using the stir-plate multiplier incorrectly.
 
Care to cite a reference or two?  Mr. Malty calculator has been around a LOT longer with multiple revisions, and has had a lot more scrutiny by VERY knowlegable users, professional brewers, and yeast scientists. 

 
Nope, no references.  I hope Brad can take a look at this and explain better than I can.  On Mr. Malty, any result for liters required with a stir-plate is going to be 2.66 times less than the amount required for a simple starter.  It doesn't matter what numbers you plug into the boxes.  That wouldn't be the case if Mr. Malty used the stir-plate factor before calculating volume.
 
Jamil did his own personal research over many years, and verified it all with Chris White (founder of White Labs), and cross referenced it with significant academic research....ultimately culminating in a well respected book.  I trust that data and those methods, in the absence of PROOF that his data and formula is incorrect.

 
sam said:
Both calculators use a stir-plate multiplier of 2.66, but in different places in the equation.  BS2 divides the growth factor by the stir-plate multiplier to reach the growth factor used to calculate volume of the starter.  Mr. Malty calculates the volume of the starter using one package of yeast and divides that volume by 2.66 if you change the settings to "stir-plate."  Mr. Malty is using the stir-plate multiplier incorrectly.

Logically, one formula should be more precise.  In the MMalty version, is the 2.66 multiplier simply saying we should get 2.66 times the cell count on a stir-plate for the well-known and accepted reasons?  That is why it's a constant?  Does the volume of liquid wort affect the stir-plate multiplier in any way? 

BS2 divides the growth factor by the stir-plate multiplier to reach the growth factor used to calculate volume of the starter.  In the BSmith version, what does this mean?  If the growth factor is also 2.66, and that is divided by the 2.66 plate multiplier, then doesn't that reduce to 1.0 and become meaningless in the formula? 

What is the inherent logic in dividing those two factors? 



 
sam said:
Both calculators use a stir-plate multiplier of 2.66, but in different places in the equation.  BS2 divides the growth factor by the stir-plate multiplier to reach the growth factor used to calculate volume of the starter.  Mr. Malty calculates the volume of the starter using one package of yeast and divides that volume by 2.66 if you change the settings to "stir-plate."  Mr. Malty is using the stir-plate multiplier incorrectly.

I'm trying to understand this better, and this seems to be the crux of it.  It seems less an error in either method, but a clear-cut choice.  JZ seems to assume you bought one package when you went shopping, and uses his calculator to figure out how big you can make it using different starter methods.  As if you might do a simple if that's all you needed. 

BeerSmith seems to assume you haven't shopped yet, and tells you how much you need given the equipment (and time) you have in your brew-lab.  For me, that works b/c I use the same stir-plate every time.  I've probably over-pitched more often than not, but I would plan multiple steps before buying two liquid yeasts. 



 
Could this be the reason there are no step calculations in BS2? Is there an assumption that if you need a larger cell count; you'll need more vials/packets?
 
Im trying to get to the bottom of this. I emailed white labs and asked them what they think. I also emailed Jamil to see what he has to say on this. I havent heard back from him but will post what he says about it.

Here is the email I typed and the response I got from white labs...


I read the thread you provided the link for and tried to figure out which method is best.  From Chris White's book, pitching into a 0.8L starter will give you about a 38% growth rate with no agitation or supplemental oxygen added; essentially a "worst case" scenario.  He also mentions that you can get two to three times as much growth when using a stir plate, so let's use the 2.66 number that both programs use and see how it comes out.

75 billion cells x 1.38 (for 38% growth) x 2.66 gives a final number of about 275 billion cells.  That's pretty close to the 308 billion that the BS2 calculator came up with, so I'd lean towards that one as being more accurate.  However, I'd highly recommend picking up Chris's book and reading the yeast propagation section for homebrewers, starting on page 132.

    Cheers!


---------- Forwarded message ----------
From:*************************
Date: Wed, Aug 22, 2012 at 7:11 AM
Subject: growth factors
To: "info@whitelabs.com" <info@whitelabs.com>


Im trying to figure out why 2 yeast calcs come up with different numbers. Specifically Mr. Malty and Beersmith 2. Both calculators use a stir-plate multiplier of 2.66, but in different places in the equation.  BS2 divides the growth factor by the multiplier to reach the growth needed for the starter.  Mr. Malty seems to calculates the volume of the starter using one package of yeast and divides that volume by 2.66 when you change the settings to stir plate. Which one is correct, or are they both correct? Or am I missing something here. There is a big dispute over this on some of the forums. Thanks for taking the time to answer your customers questions!

http://www.beersmith.com/forum/index.php/topic,5895.msg29390.html#msg29390



John Carroll
White Labs, Inc.
Sales & Customer Service
Tel: (888) 593-2785 - U.S. & Canada
      (303) 530-0469 - International
 
Back
Top