Starter use and maintenance
I'm moving some questions from this topic over here so I don't hijack that one.
My questions came after reading Doc.Dough's following posts:
1:1:1 and 12 hrs at 30°C will initially have a relatively high acidity when you refresh, and thus when you feed it you need to add enough flour and water to partially neutralize the acid and get the post-refresh pH above about 5 (>6 is better). Not doing so sets you up for the LAB to stop replicating (at about pH3.5) before they have replaced the initial population and the LAB population density goes down a little bit with each refresh which produces less acid for the next round. If you do this for very long, you lose enough of the LAB that you don't have enough to initiate a batch of sourdough. Mini Oven has coached more people than I can count about feeding at 5:10:15 to recover a failing starter, and I subscribe to that practice for recovery even though I advocate feeding at a ratio of at least 1:x:x where 2<x<20 [minimum 2 for the reason stated; maximum 20 to assure that you suppress the potential contaminants in your flour - which is not sterile]. Just my $0.02 worth.
I use a weather/seasonally adjusted X:13:15 refresh ratio where X=6 in the winter and as low as 1 in the heat of the summer (to the extent that SoCal has summer heat). Though lately I have been using a thermostatically controlled water bath at 29°C and X=~4 to get a very predictable 12hrs to the point where it has lost 2% of the weight of the added flour (which if the numbers are in grams turns out to be about 300mg). The weight loss is observing the CO2 lost to fermentation and is always approximate just because I have to thump it to get it to degas before I weigh it. Time to reach that point is a better gauge of activity than volume increase and also works across all hydration levels while giving an excellent basis for comparing multiple starters refreshed in parallel. The method was developed to measure the maturity of levain that was being mixed at 230% hydration where it does not increase in volume and bubble activity doesn't work either.
My reply:
Interesting. I haven't seen weight loss used as a measurement of activity before. How does the 2% loss line up with volume change for a 100%-ish hydration starter? Do you observe that it reaches that point before or after peak? Will water evaporation also play an effect in this? Won't the water in a just-fed starter start to evaporate and the starter lose some weight because of it during 12 hours? Being flour weight, that means if you mix a starter with 30g seed, 100g flour and 100g water, you want to use it in a dough when it has lost 2g of the flour, which would be when the total mix weigh 228g?
Could you stir the starter before weighing to get more of the gas out and a more accurate reading? Of course adding the spoon or spatula to the weight since it will have some of the starter stick to it after stirring.
What's the reason you use 29C and not something like 25-26 which I've seen many times hailed as an ideal temperature for starters?
Volume change is a strong function of hydration and flour specifics and is influenced by how the starter is mixed. For a high gluten flour and active mixing I commonly see 3x - 4x before it falls back on its own (though the 2% weight loss generally corresponds to a flat top ready to recede (after peak). I have run the evaporation test to verify that it is all CO2 loss that is driving the weight loss (easy to do with a mg scale which you can now buy for less than $60).
Your calculation of a 2g weight loss is correct, but you will have container weight and spoon losses to account for. The 2% is a guideline and not a tight spec. Less than 2% is not yet mature, but it can lose 6% before it runs out of nutrients. The methodology is robust and I found it to be quite valuable for watching refrigerated starter continue to slowly ferment sugars for a month before the weight loss curve started to flatten out.
Stirring the starter before weighing is not viable as you will lose a lot more to the spoon than to CO2 loss. You can rap it on the counter if it is close to collapse anyway and that is how I make the final measurement (because I now know that 12hrs is long enough at 29°C and I expect to see 30-60 mg of loss due to burping it). Just play with it until you figure it out. But be sure to take the lid/cover off and blow out the CO2 that has collected in the container. Even a 5 oz polypropylene cup that is half full of starter holds more CO2 than you expect (for a bowl containing 250g of starter, the head space will hold enough CO2 to make a 1g difference between flushing it out and not doing so.
I settled on 29°C because I was testing for observable differences between 30°C and 29°C and decided that it didn't make any significant difference. So I left the sous vide controller set to 29°C for a week and now have a lot of data at that temperature that I don't want repeat. And it works fine, is fast, can't tell the difference, At 20°C samples want attention at inconvenient times, so it is sort of arbitrary until somebody makes a better case that I should do something else.
Thank you, Doc!
This seems like the kind of thing I want to try. Always interested in how the starter works and what I can do to optimize it. It will be interesting to see how the weight loss of my starter corresponds to the volumetric change.
What I meant with using a spoon to stir is you tare the scale with the spoon and container on it before stirring. Or know the weight of the spoon and container, stir, weigh everything and subtract the weight of the container and spoon from the total.
Using a sous vide controller is quite clever, I hadn't even thought about using that for my starters or doughs.
Not that it's something I can just apply to my routine, but out of curiosity, how do you feed the starter to get it ready with 2% loss after 12 hours at 29C? Can't be a very large seed I expect.
Another nice thing about this method is that the shape of the container doesn't matter much. I like containers or cups that are cone shaped (upside-down) to make it easier working with compared to straight-edged ones. That means the starter will rise a bit different, possibly collapse sooner and volumetric change not as easy to measure. If going by weight instead, it doesn't matter.
I got a scale with 0.01 resolution, unknown precision and accuracy. With a typical feeding of 15-25g flour, 2% loss would be 0.3-0.5g. It is a cheap Ebay China scale, so I don't know how accurate it really is, but I can try with some larger amounts when making starter for a bread and see if the measured values there aligns with those on smaller feedings. This of course also matters when I measure out the amounts of seed, flour and water when feeding it, so I got an accurate weight before fermentation starts.
Do you close your container shut or got some kind of closed container around it again? I was thinking if some evaporation from the water bath got into the starter and affected the weight.
For some reason I missed this when it was originally posted, so apologies for a long delay.
Your inexpensive Chinese scale is quite accurate by design. You may find some temperature sensitivity, but you can check the repeatability by using a stack of coins. Accuracy requires a calibration weight (and some scales come with a cal weight). 10mg resolution is fine for the purpose.
I use 5.5oz poly food service cups with lids designed for them. I mix [x:13:16] grams of [starter:water:flour] where x=2 in the heat of summer and x=6 or 7 when it is cool in the kitchen. Enough so that it loses ~2% overnight when left to ferment in the cupboard over the refrigerator.
When I use the sous vide water bath for controlled experiments (as opposed to maintaining my starter), I start with a poly cup and drop a heavy 2" dia washer in the bottom to act as a ballast weight, then set the starter cup (with lid) on top of the washer in the cup and put the whole thing into the water bath. Keep the circulator pump on low speed.
I am looking for~300mg of weight loss (2% of 16g of flour = 320mg). You can certainly measure progress toward this goal with your scale.