A D V E R T I S E M E N T


Author Topic: Baker's yeast quantity prediction model - please compare to your results  (Read 241661 times)

0 Members and 1 Guest are viewing this topic.

Offline BeanAnimal

  • Registered User
  • Posts: 86
  • Location: Pittsburgh, PA
    • BeanAnimal's Reef
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #560 on: August 05, 2021, 06:24:36 PM »
So I headed down a rabbit hole with this thread, data and desire to automate it for ease of use.

I transcribed the data - fit curves (FWIW power curves fit the rows, 3rd and 4th order polynomials fit the columns) - and no matter how much I tweaked things, was not happy with the complexity of the solution or the interpolated values returned by the curves.

I abandoned that effort and have built a Lagrange bicubic interpolation function into the google sheet that holds the data. Much better! I can pick arbitrary yeast concentration or temperature value, as long as they are within the bounds of the table, and get a valid return rise time.

One day next week I will try to tackle the UI portion of the sheet and hope to share it soon after, as long as I am not stepping on anyones toes here.



« Last Edit: August 05, 2021, 06:26:59 PM by BeanAnimal »

Offline ARenko

  • Registered User
  • Posts: 566
  • Location: The Woodlands, TX
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #561 on: September 12, 2021, 05:29:31 AM »
I can pick arbitrary yeast concentration or temperature value, as long as they are within the bounds of the table, and get a valid return rise time.
Don't you mean yeast and temperature value?  Also, why do you want to choose the amount of yeast?

Offline BeanAnimal

  • Registered User
  • Posts: 86
  • Location: Pittsburgh, PA
    • BeanAnimal's Reef
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #562 on: September 12, 2021, 09:15:37 AM »
Don't you mean yeast and temperature value?  Also, why do you want to choose the amount of yeast?

Technically - or one or another and a time to derive the missing parameter. As it turns out the interpolation worked well enough, but when I built my app I did end up falling back on the curve fit equations (one for each degree F). This proved to be accurate enough for data inside and outside of the table. A single best fit equation got fuzzy around the outer edges so I opted for an equation per curve.

Offline ARenko

  • Registered User
  • Posts: 566
  • Location: The Woodlands, TX
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #563 on: September 13, 2021, 07:21:22 PM »
Technically - or one or another and a time to derive the missing parameter. As it turns out the interpolation worked well enough, but when I built my app I did end up falling back on the curve fit equations (one for each degree F). This proved to be accurate enough for data inside and outside of the table. A single best fit equation got fuzzy around the outer edges so I opted for an equation per curve.
I'm still wondering why you'd want to pick a yeast amount.  Do you actually decide a specific yeast amount you want to use first, or are you just trying to cover all options in the event you find a reason to some day?

Offline BeanAnimal

  • Registered User
  • Posts: 86
  • Location: Pittsburgh, PA
    • BeanAnimal's Reef
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #564 on: September 14, 2021, 09:25:13 AM »
There are three variables - Temperature, Concentration and Time.  Choosing any two allows you to derive the third.

When solving for 2 or more fermentation stages (mathematically) one must enter a concentration (derived from the prior stage by solving for time and temp) and a new temperature to time to get the new time. That is a mouthful... But suffice to say, the equation is already there.

In a practical sense we need an equation arranged to solve for concentration and the same equation arranged to solve for time. Solving for temperature would be possible, but I don't have the ability to adjust temperature easily so why build to solve for it? It is easier to adjust the base (real) concentration to accommodate the desired time window.

So lets say that I have to meet the following real world criteria


It is Friday at 10am and I have people coming for Pizza on Sunday at 10pm
My home is 72F
My Refrigerator is 35F
Outdoor temp is going to be 75F by 8pm on Sunday

It follows that I have 72 hours until the dough needs to be ready.

I know that I want a 2 hour bulk on the counter and 2 hours on the try outside during the part (lets ignore the actual time it takes the dough balls to warm or cool at when moved).

That gives me ~68 hours bulk cold ferment.

The only way to solve this is by choosing a yeast concentration to fit that window.

Pretty easy using the chart or one of the several multi stage calculators that have been published. In my case, I have a concentration slider and hour spinners and can solve the problem is 15 seconds. I enter the number of desired hours at each stage and then slide the concentrations slider until I hit 100% fermentation (clipped on this smaller iphone).








A D V E R T I S E M E N T


Offline ARenko

  • Registered User
  • Posts: 566
  • Location: The Woodlands, TX
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #565 on: September 14, 2021, 11:30:10 AM »
What I was getting from the original post I quoted (where you said you can pick a yeast concentration) was that you are choosing yeast % (e.g. I want to use .1% yeast in this batch) and solving for one of the other two.  I guess I misunderstood.  It would make no sense to me to constrain yourself to a yeast % from the outset.
« Last Edit: September 14, 2021, 12:32:52 PM by ARenko »

Offline BeanAnimal

  • Registered User
  • Posts: 86
  • Location: Pittsburgh, PA
    • BeanAnimal's Reef
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #566 on: September 14, 2021, 01:00:55 PM »
What I was getting from the original post I quoted (where you said you can pick a yeast concentration) was that you are choosing yeast % (e.g. I want to use .1% yeast in this batch) and solving for one of the other two.  I guess I misunderstood.  It would make no sense to me to constrain yourself to a yeast % from the outset.

Well you would need to pick yeast % and Temp to solve for time and that is in fact what you do when you do a multistage ferment ;)

Offline ARenko

  • Registered User
  • Posts: 566
  • Location: The Woodlands, TX
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #567 on: September 14, 2021, 08:21:28 PM »
Well you would need to pick yeast % and Temp to solve for time and that is in fact what you do when you do a multistage ferment ;)
Sure, but as in your example (and what it looks like in your app) you adjust the yeast slider until you get 100% fermentation according to your time and temp constraints/ inputs.  Like I said - I misunderstood what you meant by choosing yeast. 

Offline BeanAnimal

  • Registered User
  • Posts: 86
  • Location: Pittsburgh, PA
    • BeanAnimal's Reef
Re: Baker's yeast quantity prediction model - please compare to your results
« Reply #568 on: September 14, 2021, 08:42:38 PM »
Correct!  You can also select a yeast percentage and adjust time or temperate (or both) for 1 or more stages  until you get to 100 (or whatever your target is). The chart works just fine… I just decided to try my hand at an IOS app and kept adding options. Feature creep will likely prevent me from ever being done. Though, I am not really enjoying IOS development and am ready to move on. Was going to have it store recipes, do conversions on the fly, push reminders for when to change stages, etc
« Last Edit: September 14, 2021, 08:47:11 PM by BeanAnimal »

A D V E R T I S E M E N T