The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Judging levain maturity by weight loss?

Doc.Dough's picture
Doc.Dough

Judging levain maturity by weight loss?

Is anybody using weight loss as a metric to judge the extent to which the sugars in a levain have been depleted through the production of CO2 by fermentation? What other methods provide a direct measure of maturity? What factors might act as potential confusers (evaporation of water is an example)?

Ford's picture
Ford

I suspect that the actual weight loss is too small to be an accurate measure of maturity, and some of the carbon dioxide will be dissolved in the dough and some will be retained as bubbles.

However, I will be interested in reviewing your experimental research in this subject!

Ford

Doc.Dough's picture
Doc.Dough

Case 1 was simply an observation that there was not as much levain in the bowl as I expected (short by 6g out of 880g) so I started thinking about why. I am usually off by no more than 1-2g.  When I mix 480g, I expect to net 472g after some losses to the bowl and the scrapers along the way but I had never done an accurate accounting.

Case 2 was a simple attempt to just keep track of the weight after all of the mixing was done through to the levain in the bowl before it was used. Over 10 hrs @ 83°F, a covered but not sealed bowl of levain lost 6g, going from 1253g to 1247g.

As a credibility check I calculated the total amount of fermentable sugars in the 212g of flour.  After hydration and amylase enzyme activity, and according to Saunders, Ng, and Kline, there should be about 5.5% maltose and 1.7% other fermentable sugars which I calculated to be ~12g of glucose equivalent.  Since you get 2 mols of CO2 fo every mol of glucose, that pencilled out to be close to 8g of CO2. So that is in the ballpark. Scale accuracy is  better than 1g but evaporation was not accounted for or controlled. I would expect (without any calculation of physical properties) that a significant amount of the CO2 would remain dissolved in the liquid.  So I have two things to do - lookup the solubility of CO2 in 83°F water, and measure the evaporative losses from a similar case with just water in the bowl.

Ford's picture
Ford

So you alerting for about 4% weight loss for complete fermentation of the lean.  A mature levain should still have some food remaining to use in the raising of the loaf.  It seems like a small change in weight for judgement.

However, go for it!  Let us know your conclusion with the backup data.

Ford

Mini Oven's picture
Mini Oven

I would suggest...  weigh the lid/cover of the starter before pitting it down on the just fed and weighed starter.  When mature, weigh the lid/foil again and be careful of the water droplets clinging to the down side surface when lifting.  There you may find some of the missing weight.

How accurate is your scales? Mine tends to round to the nearest gram so I can easily be .4 gram + or - with each large ingredient.  I use a more accurate spoon scale for salt.  

A more direct measure of maturity would be a pH measurement.  When enough acid has accumulated in the liquid starter.

I use an aroma test but my schnoz is not certified for scientific measurement.  :)  

A taste test for sour is also relative to the taster but very convenient.   

dmsnyder's picture
dmsnyder

Personally, I would trust the MOST (Mini Oven Schnoz Test) over any scale outside of a quantitative chemistry lab type scale.

David

dmsnyder's picture
dmsnyder

Visual cues and aroma are the best parameters for assessing levain maturity. I ferment my levain in transparent containers and find the visible pattern of bubbles a helpful cue. Otherwise, the specific cues are different for firm and liquid levains. Volume expansion and dome formation for firm levains. Surface bubble formation and wrinkled surface for liquid (100% hydration) levains. For either, collapse with volume reduction means over-fermentation.

Fruity aroma means moderate fermentation. Sharper, "acidic" aroma means greater maturity. 

David

Doc.Dough's picture
Doc.Dough

@David - Some tests I ran a couple of years ago demonstrated that I could achieve significantly higher TTA with very liquid levains run long and hot, with TTA continuing to increase beyond 36 hrs and no huge reduction in yeast activity.  I think DBM has observed the same phenomenon though without the quantitative measurements.  So when I am planning to make a sour loaf I put all of the liquid into the levain and generally let it run at ~83°F for 24hrs (the TTA curve is still rising but relatively flat after that).

Case 2 was one of those but the the run time was truncated.

Doc.Dough's picture
Doc.Dough

While my mid-range scale (7Kg) resolves to 1g, I have developed a method for getting somewhat better resolution even if it does not improve the accuracy.  With pieces of foil that weigh 0.10, 0.20, 0.40, and 0.80 grams (as measured on my milligram scale), I can quickly find out how close a particular measurement is from the next round gram (to one additional digit). And my scale does not round up (it truncates) so it always reports less weight than is on the pan.

But your point about weighing the shower cap before and after is well taken and I will pay attention to it tonight when I re-run the experiment with water only.

The solubility of CO2 in water at 1 atm and 25°C is .034 mol/L and a mol of CO2 is 44g so that is ~1.5g/L and I had 636g of water in the levain so the maximum that could have been held in the liquid was a bit less than 1g.  The temperature was above 25°C and it was exposed to the atmosphere so it should diffuse off some of that 1g and retain a smaller amount.  The bubbles don't count for enough to be a concern (hydration = 288% so low viscosity and even tiny bubbles pop up and either burst or collect one-deep on the surface).

dabrownman's picture
dabrownman

I was down to my last 20 g of NMNF rye starter and the previous levain I had built took nearly 18 hours to double using sprouted 4 grain bran and HE 4 grain flour,  The NMNF starter had been in the fridge unattended for 5 - 6 months so It was time to refresh it and build a new NMNF storage rye starter.

So I took 12 g of the NMNF starter and built a 3 stage levain out of it using 12, 24 and 48 g if rye at 100% hydration.  Next thing I knew it tripled at the 10 hour mark.  It obviously liked the rye whole grain rye better than the sprouted bran and HE 4 grain flour but it had peaked but not collapsed. I stirred it down and it doubled again in 2 hours.  So when was it it at its peak?

It sure smelled better at the 12 hour mark after doubling than did at the 10 hour mark after tripling but it wasn't as vigorous as it was at the 10 hour mark either.  The levain would have performed perfectly for any bread at either point.  So I stirred it down again, added some rye flour to stiffen it up and retarded it for 24 hours to bring out the acetic acid.  Today I will split it in half and feed half two times, at 100% hydration, with sprouted rye bran to make a bran levain, also retarded for 24 hours, for Friday's Sprouted 35% Deli Rye bake.  I will feed the other half whole rye two times for my new NMNF storage starter ending up at at 66% hydration.  So will the levain be better tomorrow afternoon for Friday's bake?

I think so but this is a complicated levain build that most folks won't ever do and one I only do twice a year. Weighing it to see how many grams were lost, at what stage of the levain development over several days, really has nothing to do with determining when the levain is ready to use for what I want it to do when making bread.  

It is an interesting scientific experiment though and might have some merit it a normal levain build that mast folks use for their bread.  But the flour used will make a huge difference as to when it is ready too - not to mention what each persons starter is all about too.  My NMNF starter at 20 weeks aged in the fridge isn't the same as what it is at 2 weeks storage or what a counter maintained white starter is either and a bran levain is totally different as well since it is only 20% starch instead of the normal 75%% found in flours.

But maybe there is a percentage loss of weight for all levains that will point to when they are ready to go depending on what you want them to do.

Happy investigating and experimenting Doc - it will be interesting tp watch as usual

Doc.Dough's picture
Doc.Dough

I am most interested in being able to judge the maturity of a very liquid levain (>250% hydration) where there is little foaming going on when it is ready (as opposed to a 100% hydration version where you can look at the bubble density and make a good call). When I am going for a sour loaf and need the additional acid as well as lots of LAB, the run times are long and pH is below 3.8 (generally around 3.6) but not changing any more because of the decoupling of pH and TTA at high TTA values.  I could measure TTA and get a direct measurement, but even when I am well practiced it takes about 25 min to prepare a sample and do a good titration.  So that is not worth doing if there is a more simple way to make the call. I am confident that Mini's nose is as good as running a TTA measurement, but it is in Laos and we would need to figure out how to make it available on-line in any case.

Doc.Dough's picture
Doc.Dough

A bowl of water with the same surface area as the bowl of levain, in the same location, at the same temperature, for the same length of time ....... lost no weight (<0.5g).  There was some condensation on the shower cap that covered it, but the shower cap was in the measurement both before and after so the condensation was also included.

Next time I will add the trim weights in an attempt to get down to the 0.1g resolution.  I suspect that there is some amount of evaporation, though the shower cap, while not gas tight, was enough to keep the humidity in the bowl high enough that it could condense at 83°F.  Another test might be to double the amount of levain, or the amount of flour, and see if it loses twice as much weight.

Wild-Yeast's picture
Wild-Yeast

Doc,

I don't think there'll be any significant weight loss that is measurable. The Carbon and the Oxygen that are released in the fermentation process as CO2 are in fact still contained in the dough - captured in gluten bubbles and carbonation of the water contained in the dough. An insignificant amount of CO2 escapes from the dough into the atmosphere which decreases the weight but is a variable not directly connected with the amount of inflation.

What you want to measure is density - mass per unit volume. This is a difficult parameter to measure in bread doughs of varying grains and water content. The ideal instrument would be a hand held non-contact densitometer - much like the non-contact hand held thermometer.

Wild-Yeast

Doc.Dough's picture
Doc.Dough

This weekend I ran a batch of levain in parallel with a bowl full of water that weighed the same as the levain. Both sat in an 83°F environment for 14 hrs. Both weighed 1292g at the start. Both were covered tightly with PVDF film to slow evaporation of water but to allow CO2 to outgas when the pressure got high enough. The bowl of water lost 2 g for reasons I don't understand (I did not attempt to measure the weight to a resolution better than 1g).  The levain had a very tight PVDF film in the morning but it did not burp, and had lost about 6.2g (presumably CO2 which will diffuse through the PVDF when the partial pressure is high enough. After releasing the pressure and stirring the levain, additional bubbles were released from the soupy residual flour and water that had sunk to the bottom of the bowl. The last digit of weight loss was estimated from the trim weight required to push the scale over the next higher gram increment (measured both before and after the fermentation).  The resolution was 0.001g but the accuracy is estimated to be 0.1g. The 6g weight loss is the same as I observed last week for exactly the same levain mix so that provides some confidence that the experiment is at least repeatable.  The 2g loss for the water is interesting but at the moment remains unexplained. The H2O partial pressure difference across the PVDF membrane is insufficient to cause any significant diffusion, and any dissolved air in the water was insignificant (maximum ~6mg).

So the weight loss is both measurable and significant. And the resulting levain made a nice batch of sourdough baguettes (photo here if you are interested).

leslieruf's picture
leslieruf

Mini Oven's picture
Mini Oven

The gas bubbles in dough rise upward not downward so the gas forming in dough must be lighter than the dough itself  resisting the pull of gravity.   Scales weigh the pull of gravity.  Could the trapping of gas in the dough lighten the weight of the dough, diminishing the pressure on the scale (gravity) making the dough appear to weigh less?  

Maybe it makes more sense to use weight and decreasing weight as an indication of the amount of gas trapped in the dough? After the starter deflates, does it return to the original weight?   

Experiment:  Weigh a bottle of carbonated beverage (sealed) and a balloon.  Pull the balloon over the bottle and open the top off the bottle so that the gas can escape into the balloon but still be affixed to the bottle.  Weigh again.  Any differences?  Release the gas in the balloon by cutting a small hole in it.  Does the weight change?

Doc.Dough's picture
Doc.Dough

So it is no surprise that bubbles just pop to the top and the CO2 dissipates.  When gas bubbles form in dough, the weight of the dough does not decrease though the density does.  If the gas escapes, and the mass of the gas is mixed with the atmosphere, it no longer contributes to the weight of the dough. But if the gas is retained by a tight membrane then it stays within the control volume and is correctly weighed except to the extent that it increases the size of the control volume. The buoyancy factor is small when the fluid is air.  It would certainly be significant if the fluid was water and we were floating a dough ball while the yeast was throwing off CO2. In that case the weight could go from something to less than nothing. The calculation of Wild-Yeast below would be correct if we were dealing with rising dough instead of a fixed volume of very fluid levain.  I suppose I could pull a vacuum on the levain, but that would just release the CO2 that is dissolved in the liquid and trapped in the slurry at the bottom of the bowl which might then show a small additional weight loss.

Wild-Yeast's picture
Wild-Yeast

Closed System

What's missing is a systematic way of looking at the problem in the form of a closed system mathematically. Matter is neither created nor destroyed in Newtonian Chemistry so the system must balance out by accounting for all components in the system. The dough weighs less as it expands due to some of the CO2 gas escaping and the the fact that the loaf’s larger volume displaces air and is more buoyant (Archimedes Principle). Yes, to a very small degree the dough floats just a tiny amount on the ocean of air. Something to think about next time you bake :-).

Rough Calculation for a Batard Loaf

CO2 displaces air as the fermentation proceeds and the dough expands. At 70 °F and 14.696 psi, dry air has a density of 0.074887 lb/ft3 or 1.198192 oz./ft3.

An oval banneton of around 11" by 7.5" by 3.5" high has a cubic volume of around 165 in3. Assume that the unfermented dough fills the basket to 125 in3 and will double in bulk to 250 in3. That increase in volume displaces 125 in3 of air which weighs:

[125 in3 ÷ 1728 in3/ft3] x [1.198192 oz./ft3] = 0.08667 oz.

Or: ~ 2.457 grams

This implies that the 6 grams lost were composed of;

~ 2.5 grams buoyancy gain + 3.5 grams CO2 loss to atmosphere (water loss?)

Wild-Yeast 

Doc.Dough's picture
Doc.Dough

Again the same experiment. Weight loss = 6.2g for a control volume that includes the puffy PVDF film that covered the bowl. There was an additional weight loss of  ~0.51g (accuracy + or - .1g) when I punctured the film and allowed the CO2 (and accompanying water vapor) to escape. The plastic dome had been 9.5" in diameter and about 1" high which encloses 590cc of gas (+ or - 10%). Taking into account the density difference between CO2 (1.964 g/cc) and air (1.225 g/cc), the net bouyancy of 590cc of CO2 pencils out to be 0.43g which is within measurement error of the 0.51g without accounting for the partial pressure of H2O or the density increase due to the pressure required to stretch the PVDF and form the dome, or local barometric pressure, or the 80' of altitude above MSL, .... This also does not account for the CO2 that fills the head space between the top of the levain and the plastic film which might be an average of 8" in diameter and 2" high (in case someone wants to run the numbers).

So we now have three independent samples that all agree well within experimental error. And while the procedure is a bit tedious and of limited utility, the indication is that weight loss can be used to judge the completeness of levain fermentation under some conditions.