The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Maximizing Yeast in SD Starter

DanAyo's picture
DanAyo

Maximizing Yeast in SD Starter

Michael Wilson’s use of the highly yeasted Levito Madre (Pasta Madre) have interested me for years. Unfortunately (as far as I know) all of the definitive literature is written in Italian. Because of this I have not ventured into this endeavor.

I recently got in touch with Debra and asked her for help to develop a sourdough starter which prioritizes the yeast population over the LAB. Below is her reply, which has been posted with her permission. Hopefully others will find the information useful.
<Thanks, Debra>

You're in luck. It's not that difficult to shift the balance, and it can be done rather quickly, although changes in the organism profile will likely follow more slowly. The best way is to reduce hydration to a firm dough, 50-60% or whatever gets you to something that isn't sticky after you knead the flour in. Feed it 3x a day if you can. (If you want to see it lose all sourness, feed every 4-6 hours for a day or two.) Like always, you need to work out feeding ratios that get you to the next feed without being overripe or deflated, because that's when LAB are forging ahead. You don't need high temperatures for this, room temp works best, especially now that the weather is cooling off. The warmer it is, the more frequently it will need to be fed to keep LAB from increasing too much.

With a strong ap like King Arthur, I usually go 60% on the water and feed 1:3:5, 3:3:5, 5:3:5, or even 8:3:5 depending how long it will be going between feeds --- 12, 8, 4-6 hours or less, accordingly, or depending on temperature. You know the drill. You develop your own routine as you go. Taste along the way, because that and how much rise you get will tell you everything you need to know about the balance. You'll probably get about 4x rise once it has transitioned and stabilized, provided you have the gluten structure to support that.

So the factors that favor yeast are: white flour, low hydration, more frequent feeding, and moderate-low temperature. This kind of starter makes great pain au levain, and I'll bet it would be the perfect choice for sourdough baguettes too. I don't know if high temperature is essential for the right species profile of a good panettone starter (it could be), but if you choose that route, just remember the higher the temp, the shorter the feeding intervals need to be to keep LAB in check.

breadforfun's picture
breadforfun

Interesting that white flour would favor yeast over LAB. Why is that?

-Brad

DanAyo's picture
DanAyo

Brad, my best guess would be the minerals in the bran favor the LAB. I hope someone has an authoritative answer for us.

I plant to give Debra’s advice a try. But at 3 or 4 feedings a day at somewhat cool temps, I’m thinking 5:3:5 or 8:3:5. May even go 50% hydration (since I often do 60%) with something like 5:2.5:5.

 

Debra Wink's picture
Debra Wink

Less buffering capacity

:)

breadforfun's picture
breadforfun

...please let me walk through it to make sure. 
We are talking about TTA here and not pH. Buffering allows the pH to remain slightly higher than the pH that would kill off the LAB, so they can produce more acid, rendering the starter more acidic without lowering the pH. 

Using white flour, without the buffering capability, the acid will continue to lower the pH, reducing the LAB population, but not affecting the yeast. That, then, will increase the yeast:LAB ratio. 

Do I have that correct?

It begs another question: I’ve always understood that firm starters favor acetic over lactic producers. Do the frequent feedings minimize the acetic acid?

-Brad

For other readers who want to read more on this, there are several other TFL threads like here and here

Debra Wink's picture
Debra Wink

Hi Brad,

Using white flour, without the buffering capability, the acid will continue to lower the pH, reducing the LAB population, but not affecting the yeast. That, then, will increase the yeast:LAB ratio. 

Yes, that's the general concept. Without buffering, the pH drops faster with less acid, and LAB are more sensitive to pH than yeast are. However, the pH isn't really low enough to limit LAB in this scenario as much as the low water activity of a firm dough and frequent feeding at cooler temperatures. It was part of the list so I included it. You would be able to taste a difference in acidity between a firm white starter and a desem, which is essentially the same thing in whole wheat. 

I've always understood that firm starters favor acetic over lactic producers. Do the frequent feedings minimize the acetic acid?

The acetic and lactic producers are one and the same. They simply produce more acetic when they have the raw materials to do so, and the conditions influence that. But that doesn't really matter because reduced hydration with more frequent feedings reduce LAB in relation to yeast, and so it decreases all acid production in relation to leavening power. It's not so much about what they're producing, as how many there are to produce it. It's all relative to population sizes which is what we're trying to rebalance here.

dw

breadforfun's picture
breadforfun

Hi Debra,

Thanks for the explanation and for the clarifications. It makes a lot of sense.

I am not unhappy with my loaf volume or with the flavors my loaves have developed. Now I understand that while it is not optimal, my 67% hydration starter kept at San Francisco room temperatures of around 65˚F does, in fact, tilt towards favoring the yeast.

You are a great resource for this community, much appreciated.

-Brad

albacore's picture
albacore

I currently keep an "ersatz" lievito madre starter as well as a normal one. The LM is all BF at 50% hydration, the normal one is 25% rye/75% BF at 80% hydration.

As part of of my recent focus on pH, I checked pH on both after 4 days storage at 9C.

  • LM 4.19
  • normal 3.82

Confirmation that there definitely is more yeast/less lactics in a stiff 100% white starter.

Lance

DanAyo's picture
DanAyo

Lance, all things being equal, will the LM raise a larger bread?

What are the diffrences in the breads produced from each starter?

Debra Wink's picture
Debra Wink

With less acidity comes less proteolysis, which translates to more gas-holding potential and lighter, milder breads.

Just refreshed
Stiff levain refreshed

At peak
Stiff levain at peak

 

DanAyo's picture
DanAyo

I posted this to a new topic (strong flour for long fermented Levain), because it is pertaining to another subject. 

Debra Wink's picture
Debra Wink

I think Lancelot would be overkill for this starter, or a levain made from it for a mild bread. In general I get better flavor and crust from the lowest protein flours that are high enough to do the job. And that for me is usually a strong all-purpose. I use stronger flours only as needed. With whole grain flour, the best is usually whatever's the freshest, although luckily for me, that happens to be a strong flour.

Here is a pain au levain I made using KA all-purpose (11.7% protein winter wheat) with the firm starter pictured above. 

pain au levain

mwilson's picture
mwilson

Hi Debra, that is some lovely looking bread. Perfect ?

Debra Wink's picture
Debra Wink

Coming from TFL's panettone king, I consider that high praise :)

YenForYang's picture
YenForYang

That Pain au Levain looks fantastic. What flavor does your 60% hydration starter bring to it -- is it considerably less sour than using a 100/125% hydration starter? Any different from simply building a stiff levain (which the recipe on King Arthur does)?

I would like to know your recipe, what's your formula? Do you use any other flours (e.g. whole wheat, rye)?

Debra Wink's picture
Debra Wink

That Pain au Levain looks fantastic.

Thank you, it tasted fantastic :)

What flavor does your 60% hydration starter bring to it -- is it considerably less sour than using a 100/125% hydration starter?

That particular bread wasn't sour -- in part because of the way I maintain and prepare the starter for baking, but in equal part because of the formula and process after incorporation into the dough. It's a wonderfully mild bread that allows the more delicate and complex flavors from the wheat to shine through.

Any different from simply building a stiff levain (which the recipe on King Arthur does)?

Yes, I do find the two approaches to give different results for me, but hydration isnt the only factor. Temperature, feeding rate/frequency and maturity are important parts of the equation too. The starter that is maintained stiff has less acidity and gives me bouncier and fluffier doughs

I would like to know your recipe, what's your formula? Do you use any other flours (e.g. whole wheat, rye)?

That bread is the Pain au Levain in Jeffrey Hamelman's Bread (1st and 2nd editions). It is 95% white flour (11.7% protein), 5% medium extraction rye, 65-70% hydration and 1.8% salt. It has a same-day process because its special qualities are lost with retarding.

dw

albacore's picture
albacore

I would think so, Danny,as Debra explains - less proteolysis so you are probably nearer to a pure yeast fermentation.

I guess you could view the stiff starter in a similar vein to a biga and, yes, I would say to use a high gluten flour again as used in a Biga. Maybe something about W320 or more (Italian measurement).

Lance

Benito's picture
Benito

Doc and I were chatting about pH and the effects of pH on the LAB.  Hopefully he doens’t mind me quoting his reply to me here.

Doc said “ The LAB stop replicating at around pH 3.8 but continue to metabolize sugar and produce acid until either the sugars are depleted or the pH gets down to around 3.4 (which is really low and takes a long time - especially when refrigerated). What I found was that while the LAB are sensitive to pH, they are also sensitive to both lactate and acetate concentrations (Gänzle's paper) which means that if you run the levain at a very high hydration you dilute the lactate more than you impact the pH and you get more acid in the levain.  So a levain that is propagated at a hydration of 250% will have (2.5X) more total lactate in it than a 100% hydration levain at the same pH”

So it would make a lot of sense for the corollary to be true that a lower hydration levain will reduce the LAB population.

Rob1's picture
Rob1

Hello Everybody, Hello Debra

I would like to talk about the connection between Yeasts and Ph.

We know that yeasts are dormant in the initial phases of creating our sourdough, we must lower Ph to get the yeasts going (about Ph 3.5).

Once the yeasts  will appear can we raise the Ph of the starter? From what I read it seems affirmative...this avoiding proteolysis for example.

But if we decide to raise the Ph will the LAB still "help" the yeasts in fermenting complex sugars? And raising the Ph we will not give the bad Leuconostocs a chance to increase in number?

The big problem is when we use our starters in our dough..

We keep our starter well balanced, well feeded, we taste it, we measure (ph) it.. and so on.. BUT we MIX it in our doughs in a completely different enviroment with differen flours, different temperature, different Ph ...and the problems are unavoidable..because our starter does not recognize the new "pattern"..

Debra how can we avoid this really big problem?

 

Debra Wink's picture
Debra Wink

Once the yeasts  will appear can we raise the Ph of the starter?

Yes. It will rise naturally.

And raising the Ph we will not give the bad Leuconostocs a chance to increase in number?

They don't seem to be active where yeast are fermenting. Perhaps the alcohol is toxic to them.

The big problem is when we use our starters in our dough.

If results are evidence, it's not a problem :)

We keep our starter well balanced, well feeded, we taste it, we measure (ph) it.. and so on.. BUT we MIX it in our doughs in a completely different enviroment with differen flours, different temperature, different Ph ...and the problems are unavoidable..because our starter does not recognize the new "pattern".

Sourdough microorganisms are highly adaptable to changes in their environment. 

My best 

DanAyo's picture
DanAyo

Debra, what would be the ideal temperature for the yeast maximized starter that we are discussing?

I am working at ~69-70F. At these temps 3 feed cycles are the most I am able to squeeze in. I get about a 3-3.5x rise, varying the feed ratios according to time and temp. At the temps mentioned I am unable to mature these starter in less than 6 hr, generally 7. Reducing the seed to feed ratio won’t mature them any faster than 6hr. Normally I would raise the temp if the starter was maturing slowly, but don’t use that technique for this particular starter.

Are you familiar and have you worked with Lievito Madre and/or Pasta? I am headed in that direction and talking with Michael. Sure wished the Italians wouldn’t keep this such a secret. Almost all literature is in Italian.

Guess what? I made Ciabatta with this starter (See bake #4). The biga used 76% Pre-fermented Flour. It was my best Tasting Chibby so far!

Debra Wink's picture
Debra Wink

But bake 4 looks great, so who can argue with that?

I wouldn't say there is a specific temperature that is ideal, because it depends on all the other factors too -- hydration, flour, feeding rate and frequency. To keep yeast strong and LAB in check when the temperature is high, you reduce hydration and feed less but more frequently. When the temperature is lower like yours, you don't need to be so severe with the low, low hydration, and continuous feeding, because the temperature is already on the side of yeast and leavening. It's always a balancing act. I like mine best when temps stay in the 68-74F range, and adjust feeding frequency with the seasons. 3 or occasionally 4 times a day when the temp climbs above 75 and acidity and enzyme activity increase more than I like.

I am unable to mature these starter in less than 6 hr, generally 7. Reducing the seed to feed ratio won’t mature them any faster than 6hr.

Need to go the other way -- increase your seed to feed ratio beyond 100% starter. Try 200%. But first I would recommend loosening it up a little so that it can rise faster/higher. It looks too dry to me for 70F, and it's not going to loosen up as much on its own at that temperature. Kneading something that stiff that frequently is setting you up for carpal tunnel syndrome.

Right now my temperature is close to yours, and I'm feeding at 20g starter to 11g water and 20g flour (KA-AP). So 100% starter @ 55% hydration 2x/day. It's a medium-soft, non-sticky dough that kneads easily to a nice smooth ball. If I wanted to add a 3rd feeding, I would have to increase the starter to 200-400%. But it is very mild and strong for now, and I don't need to do that. Your mileage may vary :)

DanAyo's picture
DanAyo

For those that have not mixed a starter this dry (50% hydration), the following may help.

The starter is cut with scissors into very small pieces to make incorporation easier.

This is what the dry starter looks like after turning out of the mixing container.

The dry bits are worked into the dough.

The starter should be kneaded forcefully and with a heavy hand. Continue kneading until the dough smooths out.

For best viewing use THIS LINK.

Benito's picture
Benito

Very helpful Danny, I have yet to work with a low hydration starter/levain.  I have a 100% hydration levain build fermenting now for my first ciabatta.  We’ll see how that works out and may delve into the world of biga for a future ciabatta.

mwilson's picture
mwilson

To be read in the voice of Daniel Craig in the film "Knives Out"..

Y'all misunderstand the nature of a LM starter. This idea that it favours yeast over LAB simply ain't true.

What it favours is.. a low level of acidity. LAB will always be dominant, in all cases. For the budding microbiologists.. If one wants to know why this is the case, it is simply because the cell size of LAB are 10-100 times smaller than a yeast cell. Do those numbers ring a bell... "ratios" doth one decry.

LM is a highly active sourdough starter.

From Iginio Massari's book "Non solo Zucchero"

Check out that pH specification..

DanAyo's picture
DanAyo

Michael, if the LAB count is not diminished, but the acidity is, is that because the LAB are not producing acid as much as a normal starter due to the decreased hydration and lower temps?

Are the following statements correct?
Yeast are microbes that produce CO2.
LAB are microbes that produce acids.
As yeast is not CO2, so LAB is not acid.

When we reduce the hydration of the starter and control the temperature we are setting up an environment that slows down the metabolism of the LAB (reducing acid), yet allows the yeast to produce CO2.

I realize the statements above are elementary, but assumptions can be incorrect. I have always considered the LAB count to be relative to the amount of acids. Assumption should be challenged when warranted.

mwilson's picture
mwilson

All starters have a degree of acid load and each time you refresh or feed it, the inoculum brings with it that acid and also other fermentation by-products. The measurement of TTA can quantify the acid load and together with the inoculum percentage used this will determine the starting acidity at the point of refresh.

Take for example the traditional SFSD starter process, it has a very high acid load and so at each refresh the starting pH is very low, not above pH 4.5.

While a lievito madre starter has a very low acid load, the starting pH is typically above pH 5.

I picked these two examples to compare because they both utilise a high inoculation percentage, ~40%. Such a high inoculation has the potential to carry over a lot of acid depending on how acidic it is.

Don't forget TTA increases even after the LAB have stop growing.

It could be argued that perhaps a better way to gauge the potency of LAB is by measuring how quickly the pH drops relative to the inoculation %.

Fermentation proceeds more quickly when the environment is less acidic. The low ash flour choice used with LM and frequent refreshes help to keep the growth of both LAB and yeast more or less in an exponential phase.

mwilson's picture
mwilson

I posted this somewhere else, but I'm adding it here too since it is relevant.

What is being measured:

  • pH
  • TTA
  • Population density of yeasts
  • Population density of lactobacillus

Notes.

Yeast and the LAB grow together in step at a relatively equal rate until a pH just above 4.0
After this LAB continue to grow while the yeast slow and reach a stationary phase.
Yeast remain stationary all the way to the end of the data available while LAB can be seen to enter a death phase after a short stationary phase, which is noteworthy I would say.
TTA grows steadily as one might expect reaching a maximum just after the point where LAB have entered the stationary phase.
pH declines steadily at the beginning until a critical pH around 4.0 is reached, after this the pH declines much more slowly.

Yeasts maintain a long stationary phase compared to LAB and no death phase for yeasts are recorded. This sugests that yeasts can do better during prolonged storage. This matches my own observations.

The pH continues to fall slightly even after the maximum TTA has been reached. This demonstrates the existing acids are continuing to deprotonate.

Another Girl's picture
Another Girl

Thanks to everyone here, especially to Dan for pointing me to this thread and to Debra and Michael for their expertise and generosity. On Friday, I began converting a bit of my rye sour to a stiff white starter and yesterday (Monday) I baked Mr Hamelman's Vermont Sourdough with it. I knew it could impact the flavor profile of my bread, but I didn't realize how dramatically it could do so. This is exactly the flavor I've always wanted to achieve but it has always eluded me. I'm absolutely delighted and cannot thank you all enough! ?

-AG

Tom M's picture
Tom M

I've just posted to the geeky "advanced topics" section but wanted it to join in here as well, as it pertains to favoring yeast in sourdough: 

http://www.thefreshloaf.com/node/66462/salt-sourdough-cultures

 

--Tom