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


Author Topic: Sourdough starter quantity predictive model  (Read 118485 times)

0 Members and 1 Guest are viewing this topic.

Offline [email protected]

  • Registered User
  • Posts: 19
  • Age: 47
  • Location: San Diego
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #240 on: September 05, 2017, 12:05:50 PM »
Does this give you the information you need?

https://docs.google.com/spreadsheet/ccc?key=0AuvMQbzk5INUdGZScWx6U2lYSEtZVkJuVGJiR19NaXc#gid=0

I have used this spreadsheet a few times for multiple fermentation stages and it works great.   However, I tried to use it yesterday and there seems to be a calculation error in one of the cells as I am getting a #REF! error.    The cells are locked so I am unable to see where the error is.   

Any help will be appreciated.

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #241 on: September 05, 2017, 01:22:43 PM »
Try it now.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #242 on: September 05, 2017, 01:24:43 PM »
I always use the table: https://www.pizzamaking.com/forum/index.php?topic=22649.0

Any yahoo can edit that public spreadsheet, so use it at your own risk. The smart thing to do is to make a copy that only you can use.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline [email protected]

  • Registered User
  • Posts: 19
  • Age: 47
  • Location: San Diego
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #243 on: September 05, 2017, 05:14:36 PM »
I always use the table: https://www.pizzamaking.com/forum/index.php?topic=22649.0

Any yahoo can edit that public spreadsheet, so use it at your own risk. The smart thing to do is to make a copy that only you can use.

Thanks for the fix, I downloaded it this time.    I typically use the tables, but was trying the spreadsheet the last few bakes with multiple stages.    Thanks again.

Offline asmdsr

  • Registered User
  • Posts: 5
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #244 on: September 11, 2017, 03:14:37 PM »
In the chart, is the "Starter %" relative to the raw flour or total flour (ie including the flour in the starter)?

For example, if I'm trying to do 30% starter and 70% hydration, assuming starter is 50% hydration, are these numbers correct?

Flour    630
Water 410
Starter 190

And now I realize another question, what is the starter hydration assumed in the model?
« Last Edit: September 11, 2017, 03:16:16 PM by asmdsr »

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


Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #245 on: September 11, 2017, 03:26:07 PM »
In the chart, is the "Starter %" relative to the raw flour or total flour (ie including the flour in the starter)?

For example, if I'm trying to do 30% starter and 70% hydration, assuming starter is 50% hydration, are these numbers correct?

Flour    630
Water 410
Starter 190

And now I realize another question, what is the starter hydration assumed in the model?

The raw flour, but you may need to adjust your formula water for massive amounts of starter.

The starter assumption is ~100%HR.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline asmdsr

  • Registered User
  • Posts: 5
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #246 on: September 11, 2017, 03:57:12 PM »
Thanks Craig.

By the way I meant to say assuming starter is 100% hydration.

It appears that Ken Forkish's Overnight levain pizza dough recipe from Elements of Pizza has some mistakes?  The starter is 66%, and the proof time is about 8-12 hours, which is way off compared to the chart.  When I made it it came out way over-proofed.

Ken's proportions:
Flour 375
Water 225
Starter 250

My proportions I wrote in my earlier post were from my second attempt, where I dialed back to 30% starter.  It came out quite nicely, still a little over-proofed, but pretty good.

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #247 on: September 11, 2017, 04:49:51 PM »
66% sourdough or 66% baker's yeast preferment?

In my experience, that much SD will dissolve your dough. In any case, after a couple hours it would be overblown - let alone 8-12 hours. Something is wrong or missing?

Can you take a picture of the recipe and post it.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline asmdsr

  • Registered User
  • Posts: 5
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #248 on: September 11, 2017, 10:29:56 PM »
Hi Craig, here is the recipe: https://imgur.com/a/rV31d (click photo to enlarge)
« Last Edit: September 12, 2017, 09:48:38 AM by Pete-zza »

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #249 on: September 11, 2017, 11:07:24 PM »
I don't know what he's thinking with that recipe.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

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


Offline asmdsr

  • Registered User
  • Posts: 5
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #250 on: September 12, 2017, 12:36:37 AM »
Thanks for the confirmation, I though I was going crazy!  Your chart is very helpful, as I said I was able to adapt this recipe quite well using it, and will refine it further.

I saw some pictures of your pies and they look fantastic as does your whole setup.  Is https://www.pizzamaking.com/forum/index.php?topic=20477.0 your current recipe?  I notice the thread is 5 years old and 27 pages long - let me know if you have a newer version.

Regards,
Tim

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #251 on: September 12, 2017, 08:53:13 AM »
It's pretty much the same. The main changes are:

1) 2.8% salt.
2) 1.9 - 2.1% culture depending on the time of year (I've settled in on a dough that is a bit more fermented, and also my culture seems to have gotten weaker over time).
3) 36+12 hour fermentation with balls in wood boxes. Sometimes I use plastic bags in which case I do 24+24 hours.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline Andreas_1972

  • Registered User
  • Posts: 12
  • Location: Germany
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #252 on: October 05, 2017, 05:38:10 PM »
Hi Craig,

I read the question from "asmdsr" and come to a (perhaps silly) question: In an imaginary receipe there is 1kg of bread flour. From your table I get the information to use 10% starter to get the dough ready to bake in my preferred time. How much of the starter does it take then?
I think the starter itself is seen as additional ingredient, disregarding the flour in it.
So in my opinion I have to take 100g of my starter at 100% hydration. Is that correct?

On the other hand: Do I have to subtract any flour from the 1kg unfermented part, so that all flour in sum is 1kg again?

This becomes particularly evident, if I use 40% starter instead of only small portions...

Best regards,
Andreas
« Last Edit: October 06, 2017, 09:47:09 AM by Andreas_1972 »

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #253 on: October 06, 2017, 08:09:55 AM »
Andreas,

Theoretically, no, you wouldn't adjust your flour or water - just add the indicated about of starter. As you note, however if your starter is more hydrated that what you for a finished dough, at some point you might need to adjust back the water to lower the hydration of the final dough to a workable point.

The #1 caveat of this table (and the baker's yeast table) is it is intended to help you find a starting place rather than to be a be-all-end-all. It's particularly true with this table for the reasons you note. With the baker's yeast table, it's a guide for the direct method use of yeast across the table. With this table, at low levels it's direct method and at higher levels it's indirect method (a preferment).
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline Andreas_1972

  • Registered User
  • Posts: 12
  • Location: Germany
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #254 on: October 06, 2017, 09:45:42 AM »
Andreas,

Theoretically, no, you wouldn't adjust your flour or water - just add the indicated about of starter. As you note, however if your starter is more hydrated that what you for a finished dough, at some point you might need to adjust back the water to lower the hydration of the final dough to a workable point.

The #1 caveat of this table (and the baker's yeast table) is it is intended to help you find a starting place rather than to be a be-all-end-all. It's particularly true with this table for the reasons you note. With the baker's yeast table, it's a guide for the direct method use of yeast across the table. With this table, at low levels it's direct method and at higher levels it's indirect method (a preferment).

So in conclusion you would go along with me to add 100g starter (composed of 50g flour and 50g water) to 1kg flour, if the table predicts 10% starter?

Best regards,
Andreas

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


Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #255 on: October 06, 2017, 09:51:45 AM »
Yes.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

Offline dciolek

  • Registered User
  • Posts: 21
  • Location: Ohio
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #256 on: October 14, 2017, 09:43:40 AM »
Wow -- there is amazing math behind something as natural as yeast and bacteria helping dough to rise and sour!

Just confirming after some study, but in a nutshell -- it looks like the math in the google docs spreadsheet is summing each time/temp step progress to add up to an exponent of 2 (that is 2^x) which results to a multiplication factor of the original sourdough starter percentage such that it reaches some "target" completion factor -- which is a constant in the spreadsheet.  (i.e. -- how many times does the preferment have to double its concentration in order to grow to some target % value). 

Just wondering if this target completion factor is chosen relative to 100% -- which might define the state where there is no more potential yeast/bacteria activity available because they used up all the sugar/carbs.  The premise for setting it below the 100% level is that it would be over-fermented.

If that is correct (or even if not) you can use that step math and the periodic output of the Raspberry Pi w/temp probes to come up with a "Percent to Predicted Completion" based on actual temps in the dough ball over the length of the fermentation.  And you can let the computer do the work to measure at any interval (hourly, every 15 minutes, every 5 minutes, etc) to get better progress resolution. 

If you make the "completion factor" a variable, in case it needed to be adjusted for any reason due to type of dough or conditions or preferences, the rest of the math in your spreadsheet still applies.  A test run is attached:
« Last Edit: October 15, 2017, 07:04:28 AM by dciolek »

Offline plainslicer

  • Registered User
  • Posts: 79
  • Location: DC
Re: Sourdough starter quantity predictive model
« Reply #257 on: October 19, 2017, 09:44:51 AM »
I have nothing to add, but I would like to say thanks for creating this and the baker's yeast models. They have both been a huge help in improving my doughs in the past few months.

Offline mikeoz

  • Registered User
  • Posts: 2
  • Location: Melbourne, Australia
  • I Love Pizza!
Re: Sourdough starter quantity predictive model
« Reply #258 on: October 28, 2017, 06:55:04 AM »
Kudos to you Craig, this is really good info for newbies and more experience tweakers too.
I knocked out a couple of pies in my home oven last night -- 5% starter for 48 hours. My cooler is a little smaller and the dough was running at about 15.5 C (~60F). They came out a little pale, as it's a crappy home oven and 70% H2O, but damn they were tasty. Really creamy, light and fluffy cornicione, and a great balance of acid and funk in the crust. Sorry no pics, couldn't get the phone out fast enough.
I'll be playing around with the cooler to try and get the temp up a few degrees for the 17C sweet spot.

Curiously, as a reasonably competent bread baker, I thought I'd make use of the extra cooler space to get a loaf or two going. Usually I'd bulk at room temp (18-20C) for 4-10 hours, then retard in the fridge until bake. We're talking magnitudes of difference in starter % here, as I usually use 20%, but thought the results are interesting. It was ready in about 80% of the chart's predicted time. Of course readiness might be a different beast with a loaf, because you aren't stretching it out, rather trying to maintain structure. Also trying to maintain tighter gluten networks.
Regardless, the results spoke volumes -- much more complex sour flavour, and no need for the fridge. I've found that the fridge is great for 8-12 hrs retarding to fit around a schedule, but any more and the dough starts to degrade, and the crust becomes chewy.

Right now I've got a 2 loaf (almost 2kg) batch of bread dough bulking in the cooler. I was aiming for 24hrs bulk + a well monitored final proof, and dropped the starter to 13%. Interestingly (perhaps because it's a larger 'ball' of dough) the dough temp is bang on 17C.
Next thing to try is the other extreme, throwing in 1% starter and letting it bulk ferment for a couple of days.
I'll keep experimenting with bread between my pizza batches (unlike me, the wife doesn't think of pizza as an every night meal), and once I can get some consistency in the cooler temp, it will be interesting to map my results against the pizza charts.

Thanks as well to Mitch for the starter calculator, that's a really cool tool. I'm also curious how those 4th order polynomials in column D were arrived at...is it possible to step us through?

THanks again!

Offline TXCraig1

  • Supporting Member
  • *
  • Posts: 22969
  • Location: Houston, TX
  • Pizza is not bread.
    • Craig's Neapolitan Garage
Re: Sourdough starter quantity predictive model
« Reply #259 on: October 28, 2017, 08:32:37 AM »
Thanks as well to Mitch for the starter calculator, that's a really cool tool. I'm also curious how those 4th order polynomials in column D were arrived at...is it possible to step us through?

They are simply curves fit to the data in my original table.
"We make great pizza, with sourdough when we can, baker's yeast when we must, but always great pizza."  
Craig's Neapolitan Garage

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