The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Starter Feeding Ratio

albacore's picture

Starter Feeding Ratio

I refresh my starter by feeding 1:2:2 for 5 hrs at 25C, by which time it is doubled. I then feed 1:1:1 and leave it on the counter for an hour before putting it in the fridge for storage (5 days at 9C).

(I am talking here about running starter refreshes as a separate process to levain builds - I know some people combine, which I don't do.)

This has always seemed a reasonable regime, but I chanced upon a recent comment by Doc Dough which has got me thinking:

If your post-refresh pH is too low, the pH of the resulting levain quickly gets down to 3.8 and the LAB stop replicating while the yeast continues to grow exponentially until the glucose is consumed or you use the levain and feed it some new starch.  When feeding a starter I like to be up around 3:13:16 if the weather is warm, with a little more seed when it is cooler but always at least 1:3:3 (for white flour).

So maybe I should be feeding 1:3 or 1:4? Thinking about it, I guess most non-refrigerated starters must be fed at high ratios, otherwise you'd be feeding them four times a day.

What do you reckon?



Heikjo's picture

My first reaction was the fridge at 9C. Is that an ordinary fridge with milk, meat etc. too or one for specific uses? A fridge for every use should be 0-4C for food safety, and at those temperatures the yeast activity from sourdough seems to come to a complete halt. I don’t know about 9C though.

When most people talk about putting the starter in the fridge, it’s with 0-4C, so keep that in mind.

albacore's picture

It's a special mini fridge only for starters. Storage at 8-10C is generally considered preferable as it helps to preserve the ratio of microbes.


mwilson's picture

You never really have to worry about the LAB. If during a particular feeding cycle they are hindered for whatever reason, on the next feed they are likely to bounce back as their cell numbers are always higher than the yeast.

For example, if the starting pH is low, it means the LAB produce less acid before they shut down again. And so, on the next feed cycle, because less acid was produced previously the starting pH will be higher and therefore open the door again for LAB.

Building on what I believe to be the seed of the question, then ideally feeding at high ratios consistently would be a default approach. In so much as, you can always satisfy the optimum pH range for growth of LAB. However, in doing so the TTA would remain consistently low but relative to the buffering capacity of the flour used.

A higher or persistently elevated TTA is only ever achieved by letting the cycle run past the end point for growth (e.g. traditional SF sourdough process), again relative to what the flour can buffer.

Those are my thoughts at least...

DanAyo's picture

This EXACT statement by Doc, “ If your post-refresh pH is too low, the pH of the resulting levain quickly gets down to 3.8 and the LAB stop replicating while the yeast continues to grow exponentially until the glucose is consumed or you use the levain and feed it some new starch.  When feeding a starter I like to be up around 3:13:16 if the weather is warm, with a little more seed when it is cooler but always at least 1:3:3 (for white flour).” also made a big impression on me. For at least the last year I mainly stayed with 1:1:1. The reasoning - my starter was very active and strong. There was (and is) a concern that higher feed ratios could introduce new strains of microbes and alter a good thing. BTW, very strong white flour was fed, reasoning that the gluten would break down less. 

At this time, my practice to to fully mature the starter. It is then stirred down with a bite more flour and directly refrigerated. The extra flour (IMO) benefits 2 ways. It provides additional feed for the mature starter during refrigeration and it also stiffens the starter a bit, making it somewhat stronger.

But for an unknown reason my starter started to wane a bit. So higher feed ratios coupled with ~20-30% whole rye was introduced. Starter picked up.

Micheal’s reply (“ You never really have to worry about the LAB.”), also seems to confirm what I have believed for many years. The production of LAB is super simple and their resulting acids are not difficult. I back that statement with this. Take ANY SD starter and allow it to over-ferment. Taste it if you dare. I don’t measure TTA, but you can be sure it is high.

With the above paragraph as a foundation, I believe that any SD starter (acidic or not highly acidic) can and will produce extremely sour tasting bread if fermented at either warm (lactic) or cool (acetic) temps and given enough time (important point) to build the desired acids. That bread will taste sour.

I may not have the science right, but the experience with hundreds of SFSD loaves have me convinced.

Great topic, Lance.

Thought -
Maybe starting with a lower pH starter (feeding 1:1:1) is a good thing when focusing on maximizing the yeast.


Here are my notes pertaining to Ph and my starter. It may be of interest to some.

pH Starter Stats



3.71 refrigerated starter for 7 days

4.48 after mixing 1:1:1

3.95 after 4 hr @ 82F. 3x rise (*)

4.73 after mixing 1:1:1

3.92 after 4 hr@ 82 F

- - 3.77 after leaving (*) on counter at ~72F for 4 more hours

4.66 mixed 1:1:1 starter

5.47 mixed 1:5:5 starter

3.88 after 4 hr 1:1:1 3.5x rise

4.23 after 4hr 1:5:5 1x rise

3.77 after 8hr 1:5:5 3.25 rise



3.62 starter left out on counter 24 hr



3.80 1:1:1 refrigerated since 4/26

4.57 1:1:1 remix

3.66 1:5:5 refrigerated since 4/26

5.31 1:5:5 remix 1:5:5

3.90 1:1:1 after 4 hr matured

4.13 1:5:5 after 4 hr

3.67 1:5:5 after 9 hr 

5.41 SFSD dough BF 4 hr @81F

3.91 SFSD after 2 folds 

3.89 SFSD after 1 fold

3.65 SFSD 1 fold end of 16 hr BF

3.73 SFSD 3 fold end of 16 hr BF



3.56 1:5:5 out of fridge since 5/2

3.78  1:1:1 out of fridge since 5/2

4.69 1:1:1  refreshed with bran water

4.51  1:1:1 refreshed with spring water

6.51 bran water (bran water was cold ~44F, but meter used ATC)

6.67 spring water (spring water was at RT ~70F, but meter used ATC)

Note - 1:5:5 refreshed 2 wks ago @ 5.69 - 2 wks in fridge w/hooch 3.74

4.05 1:1:1 Bran Water matured in 3hr. This is unusual, normally takes 4hr @ 82F

3.92 1:1:1 Spring Water matured at 4hr, which is the norm



3.81  1:1:1  out of fridge - since 5-9-21

4.69  1:1:1 after refresh 

3.96  1:1:1 at maturity

4.79   1:1:1 after refresh

4.07 1:1:1 at maturity dropped temp 4F

5.88   Water and flour only (autolyse)

5.98   Water and flour after 2.5 hr autolyse 



3.73   1:1:1 out of fridge

4:53  1:1:1. After refresh

3.94 1:1:1 after 4.5 hr

4.81.  1:1:1 after refresh

3.93.  1:1:1 after4hr

4.05.  1:1:1 after adding ~3g flour



3.79.  1:1:1 out of fridge 

4.55   1:1:1 refresh 30% rye

3.99.  1:1:1 at maturity 

4.68   1:1:1 after refresh all white flour

3.97.  1:1:1 at maturity 

5.09  10:16:20 refresh

3.95.  10:16:20 at maturity 



3.83 1:1:1 out of fridge

5.21 after 1:3:5 refresh

3.91 over matured 

5.17 1:3:5 refresh


4.11 1:3:5 25% whole rye at maturity 

albacore's picture

I decided to check a few pHs at different starter ratios and see how the pH changed with time.

So I took my starter out of the fridge (5 days old) and made up 3 separate refreshes at 1:2:2, 1:3:3, 1:4:4 with bottled water. Feed flour was 50% HiEx (diy) & 50% UK BF.

I checked the initial pHs and then after 5, 7 and 8 hours at 25C. Starter pH was 3.79


Starter refresh pH trial     
feed ratio0578hrs

So it looks like I need to feed at 1:4 to have a reasonably high starting pH. It also looks like 7hrs at 1:4 will get me to about the same pH as 5hrs at 1:2, so a 7hr refresh should suffice.

I'm now wondering if the extra 1:1:1 feed I do just before refrigerating is a good thing, because it is like a refresh with a low starting pH?


DanAyo's picture

Lance, allow me to think out loud.

  1. The higher the ratio of flour to seed (because of buffering), the pH is higher.
  2. LAB reproduction (not acid production) is halted once the pH is lowered to a level in the medium high threes (~3.7?).
  3. Yeast are unaffected by pH as far as reproduction or gas produced.
  4. Because of list item #1, we expect the population of LAB to increase, because it takes longer to lower the pH of the culture to the point that they cease to reproduce. And a larger population of LAB are able to produce more acid than a lower relative population.

IF we want a stronger starter (meaning a higher ratio of yeast to LAB) that is able to produce more gas, which starting pH is better?
A lower pH or higher?

These types of post are very interesting. I want to thank those that contribute as we all endeavor to learn together.
Thanks, Lance.

albacore's picture

Well, it would seem that doing the opposite of what I'm doing might produce more yeast relative to LAB!

Surely this is the principle of LM - small feed ratios, eg 1:1, to boost up the yeast? I also have a feeling that stiffer starters might produce more gas, but I don't have any proof of this.

I like the idea of stiff starters, but starter maintenance and levain creation at 60% hydration or less is a serious chore. I have done it in the past, but I don't really want to go back there.

I shall try my 1:4 feeds for a few generations and see how things go.


DanAyo's picture

Lance, years ago Debra got me started with 1:3:5 (60%) starters. I hand kneaded them a few minutes, but never considered them a problem what so ever. 5:3:5 would be just as easy.

What did you find difficult about the 60% hydration?

Most of my levains have been 100% hydration lately because I like the Lactic Acid flavor.

albacore's picture

Danny, I guess I like the idea that those few minutes are a few seconds at 100%!


Benito's picture

Lance I’m with you on this, I have tried mixing a 60% starter and didn’t enjoy the process.  I’m happier to do 100% hydration starters.