yecora rojo desem again

This one was another epoxy method desem, but I sifted the flour after grinding and maintained two separate overnight soakers in a wine fridge: (1) a salted soaker with the bulk of the main flour at 75% hydration; and (2) a wetter unsalted 100% hydration soaker with the "sandy" middilngs (#50 sieve).  Recently, I've been adding the bran to the starter, but I'm not sure that approach is as beneficial in the case of the low 60% hydration desem starter.  So many potential tradeoffs!  An initial lamination mixing was performed with the blended/rehydrated starter, and then after a first S&F the bran soaker was re-incorporated.  The idea is that the bran will benefit from additional water and delaying re-inclusion in the mix will improve things with a good early gluten development.  It did feel that way during mixing, although I can't say I notice a big improvement from the previously yecora rojo desem bakes using a single salted soaker.  I've also been using the blender mixing jar (as mentioned above) to rehydrate the dense starter, which makes mixing a lot easier.

Note: I had logged an initial extraction rate of 80% with this flour and have been grinding and sifting in larger batches.  I store the the fine flour and sand middlings separately, which allows me to process the bran in different ways before remixing a 100% whole grain flour.  I ran out of bran after my last bake while I still had a fair amount of flour left, which shouldn't have happened.  I've been pushing the second stage milling finer after a cool down in the freezer (approximately -2 on the MockMill 200 Lino).  I just measured extraction again and got something closer to 88%.  So the last 3 bakes have actually reproduced something in the spirit of dabrowman's 107% whole grain loaf.

A more hands off version with fewer stretch and folds (I believe 1x after mixing) and looser shaping.  This was slightly under-baked due in part to a late oven pre-heat cut short by the fast final proof in this dough.  The impact of this more relaxed dough on crumb structure seems noticeable and is worth repeating.  The final shaping clearly impacted crumb structure in the center portion, and it is worth trying to do this earlier, possibly paired with looser shaping.  Once the aliquot jar indicates a 1.25x rise, things are already moving pretty quickly with this dough, and a 15 or 20 minute rest after pre-shaping may push the final shape a little later than it should be.  The pre-shape could be shifted a little earlier to accommodate this.

Summary:

• same formula (see link at top) 84% hydration, 30% PFF
• epoxy method (sifted bran in starter)
• minimal handling (Rubaud mix + 1 S&F)
• final proof @ 95F (on top of pre-heating stove)
• slightly higher final proof (bottom of aliquot sample was at 75% instead of mid-peak)
• more centered scoring (2 o'clock) opens up the top of the loaf more
• sharpened bread knife (cleaner cuts)
• some dense areas: longer bake, more thorough mix?

This one seems slightly better.  After re-reading Laurel's Desem chapter I took care to setup a 95 F final proof with the covered banneton on top of the pre-heating stove (buffered by a few pots).  I also sharpened my bread knife for cleaner slicing.  The starter is maintained at 56F and appears to be ready the night before.  I can trying lowering the temperature to get a slightly younger starter for the morning mix.

Summary:

I need to start including aliquot jar photos at the end of BF (point of pre-shape) and at the end of the final proof.  The last few loaves have been slightly underbaked and I added another 5 minutes (30 covered, 10 uncovered, 5 at reduced temperature).

• reduced hydration to 80.5 % (approximate)
• same day soaker (several hours at room temperature) with more early gluten development to clean windowpane to compensate (FF, rest, Rubaud)
• more continuous S&F's (4-5) throughout until early pre-shape well before previous 25% rise -- I prefer the more regular crumb this provides compared to last few relaxed experiments
• a much longer post pre-shape bench rest was used to improved extensibility during shaping (via Trevor Wilson's post) -- I need to end BF well before 25% rise due to speed of fermentation in this dough
• shaping was easier but this WW dough is still tight relative to any videos I've seen (Try 5% spelt for extensibility)
• repeated 95 F final proof on top of pre-heating stove buffered with pans and bannetons
• added 5 minutes to the bake at reduced temperature to help dry it out (roughly 30 minutes @ 500 F, 10 @ 450 F, 5 @ 350 F) -- the actual temperature is more complicated due to heat retention in the baking stone
• what happens if the final proof is pushed to 85%, 90%, etc?
• formula: https://fgbc.dk/182e

The aliquot jar bands were adjusted to 50% and 100% rise levels, which triggered the end of bulk fermentation and final proofing respectively.  I used a 30 minute bench rest after the pre-shape again in an attempt to increase extensibility for the final shape.  I was concerned a 100% final proof wouldn't have anything left for oven spring, but it still opened up reasonably well.  It is good to know this is still within the limits of the dough.  If nothing else it will make me less conservative in pushing beyond the 75% mark.  With the additional time at the warmer 95F final proof this loaf got pretty sour, and is probably leaving desem territory.  Perhaps I can observe the impact of just extending bulk fermentation to 50% and keep the 75% final proof (or thereabouts).  That is very close to what Benny is doing with many of his bakes.

I had reduced the bulk fermentation in the previous bake (for the same fixed target final proof) on the assumption it would reduce degassing, which would translate to more openness, and because I was extending the pre-shape bench rest.  I guess a consequence of early shaping is that the shaped dough can be trapped in a smaller form via increased tension as it tries to expand?  I did notice that shaping at 50% the dough was more fragile and my fingers did break the "skin" slightly in a few places during shaping.  Perhaps that suggests a compromise of 40% or so.

 I believe this somewhat "rumpled" exterior occurs with overproofing.

The initial slice through the middle looked a bit wilted, but these slices from the same loaf looked better.

This was essentially the same as the previous effort but with a 40% bulk fermentation in the low 70's and 80% final proof at 95 F. The aliquot jar bands were adjusted accordingly. This felt like a good compromise.  I may need to work on standardizing the measurement process for more fine grained experiments (miniscus vs mid-point, etc).  At 40% it still felt slightly pillowy but seemed to hold up better to handling.  The overnight soaker didn't happen, and I made a same day soaker for a couple of hours at room temperature with lamination mixing of the blended rehydrated starter followed by a copule rounds of early Rubaud mixing and regular S&F's during BF.  I've tuned the wine fridge to a setting that keeps the starter at about 59 F, and I've been sticking with a 1x/day feeding.  There was also no sifting or special bran processing in this one.  It isn't clear to me if any of that is advantageous from a dough performance standpoint, and I still have questions about the role of the milling process.  Lately I've been milling in slightly larger batches and freezing it.  I read one post suggesting that is sufficient to overcome poor gluten development attributed to freshly milled flour, but haven't done any comparisons.  I've ordered 50 lbs of yecora rojo berries so I can test some of this  after settling on a fermentation plan and perhaps revisit hydration.  I planned ahead and managed a more controlled 95 F in my makeshift digitally controlled light bulb powered styrofoam proofer.  Adding a pyrex container of 100 F water gave it a head start with enough thermal mass so the little light bulb wasn't overwhelmed this time.  This should be a lot more uniform than the tower of pots, pans and bannetons I was using on the stove to achieve 95 F.  I often have slight irregularities from pre-shaping that make final shaping more challenging, and the dough lacks the extensibility I see in most shaping videos even after a 30 minute rest (few of which are whole grain).  The oven spring looked as good as it has for these yecora rojo bakes.

https://www.thefreshloaf.com/node/66132/oven-spring-bulk-fermentation-and-experiment

https://www.thefreshloaf.com/node/65176/bulk-fermentation-experiment

https://www.thefreshloaf.com/node/59543/results-rise-during-bulk-ferment-experiment

https://www.thefreshloaf.com/node/59459/proposed-experiment-bf-rise

https://www.thefreshloaf.com/node/61560/pushing-bulk-fermentation

I had a starter and soaker (pre-dough) ready and I'm queuing up another attempt.  Of all the bakes in this thread, I think this one is best:

https://www.thefreshloaf.com/comment/479413#comment-479413

I agree with Paul's comment that this could use a little more aeration (red areas) but some sections seem better than others (green areas).  It doesn't seem too far off track.  Perhaps some of the tighter areas are due to handling as much as fermentation timing?  It might be worth dusting with tumeric or paprika after the pre-shape to help visualize the dough skin where degassing from handling is more likely to occur.  My original thought was that shaping on the early side would help to avoid any handling related degassing as long as the dough strength held up to the longer final proof.  This dough seems fairly tight, so that seems plausible. 

Although I had the aliquot jar marked for a 25% BF and 75% FP in this test, I actually ended the bulk fermentation (started pre-shape) earlier to accommodate the longer pre-shape bench rest I wanted to test to further relax the dough before shaping.  I was thinking it would reach 25% by the time shaping actually started.  Unfortunately, I don't have a photo of the aliquot jar after the pre-shape rest.  In this case I probably over compensated.  The bands on the aliquot jar are placed such that that tops correspond to 25% and 75% markings.  From the photo below we can see the bulk fermentation probably ended at 10% or so.  A 30% number is commonly reported in TFL whole grain discussions.  I'm going see what happens if I end BF (i.e., start the pre-shape) at 30%, and then bake at 75%.

This bulk fermentation was actually stopped at less than 10 percent or so.  Perhaps it is worth repeating this version, but with a pre-shape beginning at 25%, which is more or less consistent with the suggestion of doubling.

Same as above but with a true 30% BF and 70% FP.  The tumeric dusting was added for diagnostic purposes to better understand impact of shaping on crumb.  It may have had a somewhat Heisenbergian effect, as it seems it actually caused some separation where the pre-shaped outer skin was rolled up.  The bottom was too dense again.  Is this a sign of: (1) underbaking; (2) overproofing; (3) seam stitching/shaping?

A Sourdough Rise Time Table adapted for 60% hydration (via bwraith's spreadsheet) using a 5:6:10 feed (@ 23.8 % PFF) indicates expected doubling times of:

• 27.19 hours at 50 F
• 16.99 at 55 F
• 10.81 at 60 F

Although the starter needs to be modeled for more accurate numbers.

60% hydration

55% hydration

50% hydration

I have been performing a sort of starter rehabilitation and have returned to feeding the starter at the increased 60% hydration and a wine fridge setting of 58 F with a middle shelf location, which seems to result in a starter temperature very close to 60 F.  I've been sticking with 5:6:10 feeds for now (a close match to PiP's 1:1:2 for 60% hydration).  In parallel, I have started some offshoot experiments with lower inoculation to see if a one-feed-per-day-24-hour-peak might be viable.  There should be an inoculation, temperature and hydration combination that achieves this within the often reported 50-60 hydration % and temperature (Fahrenheit) windows, but it isn't clear how this will impact the culture in terms of leavening power and flavor characteristics.  My hunch is that pushing things too low in the 50-60 %/F window may cause problems.

I've been reviewing various desem articles, and posts, including the desem section in Laurel's Kitchen Bread Book.  (I have also ordered The Bread Builders and The Bread Book on Debra's suggestions for additional desem information.)  In all of these references, there appears to be consensus on an ideal 4 hour cool bulk fermentation and a 2 hour hot final proof.  While the criteria used to determine the end of bulk fermentation and final proof may always be subject to interpretation, my starter seems to be taking much longer.  I think this discrepancy, in part, and combined with huge differences in look and feel of a warm final proof vs retarded final proof, has lead to a tendency towards underproofing, as suggested by Paul.  I believe a few recent bakes have confirmed this (to be posted below).

Here are a few quotes from pointers from Laurel's Kitchen Bread Book to help save me from rereading this:

As the desem gains in power, it will ripen more quickly, gradually coming to the stage where it is at its best about 12 to 14 hours after it has been fed; this seems to be the tempo that gives it its greatest leavening power as well as its best flavor.  It may take a few bakings, though, before its forces come into balance.

If you have a dependably cool place, and will be very regular in your care, the desem will thrive kept unrefrigerated.  The temperature should be steady at between 50 and 55 F.  If you are going to be baking every day, this would be the ideal arrangement.

From now on, feed the desem about 12 hours before you plan to mix up the dough for baking.  The desem does have to be fed twice a week, minimum, to keep its leavening power.

It does seem she has had good success with a 50 to 55 F maintenance temperature, and I will try to nudge things in this direction again, although this may require changes to the inoculation formula.

With regard to bulk fermentation, she states the following:

Cover the top of the bowl with a platter or plastic, and set it in a draft free place at cool room temperature, about 65 to 70 F, for eight to ten hours.  During the eight to ten hours, the desem dough may scarcely rise up -- this is quite normal.  If convenient, deflate or punch the dough some time around the last hour or so; this invigorate the leavening organisms.  The dough has now finished its cool fermentation and is ready for its warm final rise.  During this next stage, the surface will lose its shine and stickiness and become dry to the touch.  This is ripe dough.  As time passes and your desem grows stronger, you will shorten the rising time until finally it takes just 4 hours for the first, cool part of its rise.  You can gauge how much time you need by evaluating the dough for ripeness as described above.

I believe stretch and folds will largely accomplish the punching down she mentions.  From recent experiments, mine appears to be much closer to the 8 hour figure she mentions.

Gauging these stiff starters is a little more challenging.  I configured a time lapse camera to help me record the slow domed peak expansion for this stiff starter, which was occurring in the middle of the night, to track any improvements in starter health.  I've also attempted to more accurately measure relative volume changes of the domed starter expansion using water weight (a la Archimedes).  The first day of rehabilitation was unrecorded.  On the second day I measured a peak after approximately 22 hours with a relative volume increase of 1.83x.  The third day I measured a peak after approximately 17 hours with a very similar volume increase of about 1.76x.  Since it is so slow moving it can be hard to guess whether or not it has truly peaked, or if I should wait another few hours to confirm this.  I have been reviewing bwraith's rise time models.  At this low hydration, they have never truly doubled, which is the criterion required for calibration with bwraith's model.   I'm going to keep the starter maintenance parameters constant for now and will configure the time lapse camera and Archimedes  volume tests in a few days time to track progress with a hope of getting closer to the recommended 4 hour fermentation.

Related discussion:

https://www.thefreshloaf.com/node/66840/tfl-index : general info about stiff low hydration starters with helpful comments from Debra Wink

https://www.thefreshloaf.com/comment/186049#comment-186049

https://www.thefreshloaf.com/node/27806/desem-starter-when-it-ripe-how-use-it

https://www.thefreshloaf.com/comment/217777#comment-217777 : w/ references to parameters in Laurel's Kitchen (1:0.68:1.32) and PiP's (1:1:2 and later 1.12.2) maintenance 

In your photo of your ripened leaven it appears to be as firm as mine generally is.  Laurel states that it should look like brown cottage cheese due to the break down in the gluten.  She also says that it should disintegrate when added to water without leaving a white color in the water.

Is there a specific reason you use the 1:1:2 ratio or is that just what works in your time schedule? (Laurel's ratio is very different 1: .68: 1.32)

https://www.thefreshloaf.com/node/27999/honest-bread-100-wholewheat-desem-bread-and-some-country-bread?page=1

In this bake, after giving the starter some R&R time, I revisited Paul's underproofed diagnosis.  I was getting low on Yecora Rojo, but had enough to finish this bake.  Previously, lowering hydration from mid 80's to 80% seemed to have a positive impact.  I decided to test a hydration of 75% in this bake, which is towards the lower end of any whole grain bake I've seen.  I used an overnight soaker in the wine fridge again at 58F.  I was a little concerned about such a low hydration with freshly milled whole grain flour, but the hydrated pre-dough handled fine and gluten development was good, although it was noticeably stiffer during the pre-shape and shaping.  Based on Paul's feedbak, I decided to "reset" the aliquot jar between the bulk fermentation and warm 95F final proof, to better reflect the behavior of the degassed dough after shaping.  While the relative volume measurements are useful, the glass aliquot jar also provides an opportunity to visualize the internal aeration, and I noticed a significant difference in the bubble distribution between 60% and 70% rise.  I suspect this visual feedback may be at least as useful and perhaps more universal across recipes than the volume change itself.  I stuck with the 30% BF and 70% FP targets using this new approach.  There was a noticeable improvement in crumb for this bake, and I believe I have been consistently underproofing.  This is probably due to a number of factors, including: the ideal 4 hour BF + 2 hour FP from Laurel's Kitchen Bread Book (mine takes much longer), my familiarity with dough look and feel from refrigerated final proofs, fear of wasting the dough, and poor feedback from use of the poke test too early in the final proof.

I fed the starter with yecora rojo, but had to augment the remaining Yecora Rojo with Montana Wheat Prarie Gold to make the soaker.  I started late in the day and instead of the usual lower 70's mid winter temperature I used the digital controller w/ a light bulb to boost the bulk fermentation to 80 F in order to have a chance of finishing at a reasonable hour.  I bumped hydration back up to 80% for this one, and used a degassing reset on the aliquot jar sample again.  By the time pre-shape bench rest was complete, the aliquot jar was probably close to 50%.  I used a 95F final proof again.  I was looking for more aeration in this one and the aliquot jar indicated a rise of approximately 75% rise before I had to terminate it in order to avoid staying up all night.  These measurements are relative to the initial post-mix state and the final proof measurement includes the pre-shape aliquot sample degassing, which removed most, but not all of the initial 30% gain from bulk fermentation.  (Recalibrating the final proof band precisely via water weight for the new degassed starting point is possible but would be an added chore.)   For variety, I was more conservative with the banneton dusting, which made for a bolder looking crust.  I was happy with the way this one turned out, which seems to approximate the original effort  While I haven't done extensive testing, the impact of the added Montana Wheat Prarie Gold on dough handling seems noticeable immediately after mixing.  The Yecora Rojo does provide a nice nutty flavor, and the two seem to be a good pairing.  A new 50 lb bag of Yecora Rojo has arrived, so I can continue experiments with more 100% Yecora Rojo loafs, so that test are comparable with previous efforts.  Thanks again to Paul for suggesting underproofing.  Things appear to be improving again.

Even with a warm 80 F bulk fermentation and 95 F final proof, this one took about 9:15 from mix to completion, including the 42 minute bake.  I believe this made it more sour than it should be under ideal circumstances, but it was still much milder than I expected.  I'm still curious whether the bulk fermentation and final proof criteria I'm using here really could be reach in 4 hours (low 70's) and 2 hours (95 F) with a more vigorous starter, or if the fermentation goals I have are different than those in Laurel's Kitchen Bread Book.  Continue feeding?  Higher innoculation?  Catching starter at a more optimal peak time?  The bloom is still good, although I think I need to aim more for top centered score "angle" of 12:00 to 12:30 instead of 2:00.

The desem starter has changed in the last few days.  I have been sticking with 60% hydration maintenance (5:6:10) and a wine fridge setting of 57 F.  That seems to result in a dough temperature between 60 and 61 F.  For now I am postponing the experiments aiming for a 24 hour peak and feeding cycle.  I have also been changing the feeding slightly.  I use 240 g for the morning mix and place 80 g back in the fridge to ripen further and then feed it after dinner for baking the next morning. Initially, this happened due to schedule constraints, but this was the first day I noticed an increase of more than 2x, so I have been sticking with it.

This schedule is consistent with a target of mixing 10-12 hours after feeding -- the suggested sweet spot mentioned in Laurel's Kitchen Bread Book.  With this regimen I have been getting an expansion of roughly 2.4x or more in the Weck Jar overnight (via the approximate Archimedes test), whereas previous tests plateaued at 1.8x or so several days ago.  It is softer and more pillowy at the time of use and I find it dissolves in water more easily.  The first day after noticing this behavior I was able to finish the bulk fermentation, proof and bake in a little over 6 hours.

It was too late to mill when I finally got around to bread preparation for the next morning, and I didn't have enough pre-milled Yecora Rojo in the freezer for a small loaf.  I did find enough Prarie Gold for both the starter and soaker, so I went ahead with it.  I think there may have been 75 g of Yecora Rojo in total.  I fed the starter w/ 80:96:160 and placed it in the fridge for use the next morning.  I made a soaker with 350 g wheat, 10 g salt, and 262.5 g water and placed it in the wine fridge next to the starter.  The following morning I mixed everything with an additional 25 g of water (75 % hydration) and proceeded as usual.  The bulk fermentation ran around 6 hours (including pre-shape and bench rest) in the low 70's followed by a warm final proof at 95 F for 1.5 hours.  The aliquot bands in the photos are still at 30% and 70%, but it rose to a little more than 50% by the time I shaped, and somewhere between 85% (contact w/ jar) to 100% (peak) during the final proof after a degassing reset during shaping.  It felt like an inflated balloon after the final proof and stood up well after removing it from the banneton.  I consistently get better bakes when using Prarie Gold or a high percentage of it with the similar process.  In comparison, at the same hydration and innoculation, the Yecora Rojo feels wetter and more fragile.  I think more adjustments for the 100% Yecora Rojo are required.  The score sealed over, but I was very happy with the loaf itself.  When I removed it from the oven it felt very light, like an inflated ball.  The crumb was not particularly open but, was uniform and lighter than any bake to date.  The crust was thinner and crispier than previous attempts, and the bread knife produced more of a classic "hollow" sound while slicing the day after. The flavor was very nice -- mild but definitely not bland.  I would be happy to make this one repeatedly (perhaps with some adjustments to the ear), and my sense is this is closest to the mark so far.  Another challenge will be to make something closer to this with the Yecora Rojo.

This one had a short second accidental second bake.  (It was cooling in the oven when the broiler got turned on for dinner preparation, but made it out alive.) 

This was a relatively unsuccessful attempt to apply the extended fermentation from recent Prarie Gold bakes to 100% Yecora Rojo dough using the latest starter.

• 75% hydration
• latest and greatest starter (now peaking at nearly 2.4x)  5:6:10 @ 60 F for 10-12 hours
• pushing bulk fermentation and final proof further 50% and 70%
• overnight soaker (salted autolyse) at 75%  (350 g flour + 262.5 g water + 10 g salt) and 60 F

I pushed the aliquot jar to 50% for the bulk fermentation, followed by a degassing reset, and another rise to 70% in the warm 95F final proof.  It seemed fairly well aerated, with larger bubbles relative to the PG dough for a similar total rise, and since the YR dough feels wetter and less cohesive, I didn't want to push it further.  The bake had fairly poor oven spring, the ear sealed over (I believe the score was too deep for this dough), and I could tell it was much denser immediately after picking it up.  I *think* it was overfer-mented.

I haven't attempted a side-by-side bake of YR and PG, and have no direct quantitative comparisons, but there are some qualitative assessments that are worth noting.  The YR soaker feels more fragile or ragged than the PG soaker (or pre-dough) immediately out of the wine fridge at the same 75% hydration.  I believe the salt and low temperature are supposed to protect it, but perhaps this is worth revisiting.  Within the first hour of mixing and stretch and folds, the differences become more apparent.  I don't aim for any gluten development prior to soaking (maybe I should), but the PG typically feels relatively smooth right away.  The YR is less cohesive, it is more prone to tearing early on (it doesn't stand up as well to lamination mixing, for instance), and leaves more of a paste behind on my hands and bench.  The latter issue doesn't seem to be something that can be corrected with additional gluten development.   Maybe a shorter final proof would be sufficient to improve things, but I do think achieving a lighter crumb with a higher percentage of YR may require deviating from the formula used for the previous light and airy PG bake.  I think it will require a stronger dough later in the fermentation process to push things further.  A few options:

• reduce the pre-fermented flour < 30%
• reduce hydration < 75% for the final dough and/or the soaker (control enzymes, improve gluten develoment, etc)
• reduce bulk fermentation and/or final proof
• possibly more gluten development (I did quite a lot in this bake)

My sense is the YR bran contents is fairly different than PR, and sifting + re-inclusion may be worth revisiting.

Related discussions:

https://www.thefreshloaf.com/comment/371735#comment-371735 (thumper256)

... over the weekend I did an experiment to test the difference between two of the types of wheat that I commonly use (Wheat Montana Prairie Gold and Great River Hard Red Spring wheat). I milled both grains using the exact same mill settings (no sifting). I followed the exact same percentages, timings, and temperatures to build dough with 100% of each grain and then shaped/baked them the same. at 85% hydration I found the Great River dough to be fairly slack (needed more stretch/folds) and the Prairie Gold could have easily handled more hydration.

https://www.thefreshloaf.com/comment/413025#comment-413025 (barryvabeach)

I find that when I mill hard white winter wheat -  78% to 80% hydration is a good range . ...  I read somewhere that one of the differences between bread flour and wheat flour, is that for bread flour, if the ideal final proof for that loaf at a particular temp was 1 1/2 hours, there would be a window of 1/2 hour where the loaf would still be fine, but for ww, the window was more like 5 or 10 minutes.

https://www.thefreshloaf.com/comment/479359#comment-479359 (Debra Wink)

But current trends are pushing higher hydrations to open up crumb, which is kind of counter to desem's design. Higher hydration benefits bacteria and enzymes, shifting the balance in the wrong direction. The antidote is decreasing the prefermented flour, starting with less bacteria, acid and spent flour, and letting the salt in the dough slow them down. Bacteria are affected more by the salt than yeast are, so it's another way to take advantage of disparity

The one-loaf-a-day approach gives a lot of space for background thinking and analyzing: bake, observe, hypothesize, read bread books and TFL posts, change a variable or two, bake, repeat.  I have been revisiting Debra's comment suggesting an inverse relationship between hydration and prefermented flour HERE.  I believe this is especially relevant to the YR.  I had wanted to try a very low (for whole wheat) hydration version with the YR flour, but found myself past the milling time without enough YR for a bake.  I did have enough premilled PG in the freezer for the soaker and starter, so I went ahead with a 70% hydration version just to see how it would behave.  I have never seen any whole grain naturally leavened bakes at such a low hydration, but I was surprised to find it quite manageable.  The desem approach is somewhat of an outlier, and most formulas I've seen are well below the 30% PFF used for these bakes.  It is understandable it would require a different line of thinking.  As an example, this breadtopia YR sourdough post I checked early on uses a target hydration of 88%.  In the back of my mind, I may have been thinking it would be necessary to work up to that.  I think this may have influenced my thinking in the desem bakes early on, believing that very high hydration would be necessary to avoid baking bricks with naturally leavened whole grain bread.  This one came out fine, although there were some signs that 70% hydration was perhaps a little low for PG (stitching was challenging), I suspect it will be even more suitable for PR, which I've queued up today.

Test formula: https://fgbc.dk/1a74

The PG dough is fairly tight in general, and especially at this low 70% hydration.  (In contrast, the YR is generally much more extensible, which should make a complementary blend.)  This property poses the biggest challenge during shaping.  I allowed for a long post pre-shape bench rest of 30 minutes or so, and I spent a few minutes carefully circling the inverted boule on the backs of my wrists and forearms, as if shaping pizza dough, to allow gravity to produce a suitably elongated rectangle for shaping.  After letter folding and rolling up it ended up producing a very tall loaf (overstuffed burrito) and it took a little more effort than usual to stitch it up, but all in all it wasn't too bad.

The dough is noticeably stronger at lower hydration at this stage of proofing, and I'm curious if the proof could be extended even further.  For example, what would have happened if I baked at this stage:

• continue w/ 70% hydration for now
• sift 500 g YR w/new 30 cm diameter #30, #40, #50 sieve stack to 90% extraction and feed 50 g bran + middling to starter (the large surface + lid and container makes this much faster) [UPDATE: more recent milling and sifting with YR has produced closer to 80% extraction (this number may be suspect -- I'm milling 500 g batches now to simplify things so the final product was 100% extraction regardless)]
• in bowl lamination mixing of soaker and starter
• [UPDATE] aliquot jar levels: BF @ 30%, preshape + 30 minute bench rest w/ shape @ 50%, degas/reset jar sample to 30%, warm FP @ 95F to 70% (poke test slowly returning)
• bake according to Thom Leonard's recommendation in The Bread Book : 10 minutes 500 F covered, 30 minutes 400 covered, 5 minutes 400 uncovered  (I believe I have been overbaking -- the lower temperature covered for longer reduces toughness of crust encountered in previous efforts and should reduce staling)

[duplicate]

• sift re-mill several times at "-2" and feed
• reduce to 20% PFF
• 72.5% hydration
• oven spring looked good, but didn't translate to crumb

• again reduced to 20% PFF
• sift #30, #40, #50, remill multiple times at "-2" and feed to starter (90% extraction)
• the impact of remilling on the dough handling and feel is very noticeable (feels
• hydration at 72.5% [edited: next one will be 75%]
• no ear again, am I on the wrong track with recent changes or is it just a technical error?

https://fgbc.dk/1b1j

• increase hydration to 75%
• 20% PFF
• sift w/ #30, #40, #50 stack and remill bran and middling at "-2" several times to make a bran powder, then feed that to the starter
• dough feels smooth and workable with remilling
• poor oven spring and the ear didn't quite lift (Score too deep?  No hot enough?)
• after reading The Bread Book I have been aiming for 10 minutes covered a 500F (425-450F on oven dial), 30 minutes covered at 400F (325F on oven dial) and 5 minutes uncovered at 400F
• in the last couple yecora rojo bakes w/ bran sifting, remilling, starter feeding and the lower 20% PFF I have ended up cutting the final proof short of the target 70% mark because the poke test seems to leave behind an indentation and I'm concerned it is getting close to overproofing, but I'm not confident in my understanding of the poke test

Temperature tracking https://share.fireboard.io/269E05

• channel 4 is oven stone (left)
• channel 5 is oven stone (right)
• channel 6 is the top of the stove near the vent

Since the finger poke indentations didn't fully return, I took that to indicate it was fully proofed or slightly overproofed, and I terminated the final proof before achieving the target 70% rise I have been using in recent bakes.  For context, the aliquot jar targets were 30% for pre-shape, followed by a 30 minute bench rest leading to a rise of 50% at the time of shaping.  I then "reset" the aliquot jar dough sample by removing it, stretching it, and replacing it.  At this point the dough sample is close to the initial starting point.  I start a warm 95F final proof with a target rise of 70%.  The oven spring seemed to suffer in this one and the ear didn't quite make it.  Does this seem like a correct read of the poke test, or should I have stuck with a longer final proof as planned?

• Mix : 9:30 AM
• Pre-shape: 1:45 PM
• Shape: 2:15 PM
• Score, Bake: 3:15 PM
• Remove from oven: 4:00 PM

IMG_5952 from headupinclouds on Vimeo.

Should the next one proof more or less?

The next day:

I'm going to try a shorter proof next time.

My pH meter has arrived and is now calibrated.  The final bread pH was 4.61.

I found one mention of desem and a final pH of 4.7:

Desem dough is fermented slowly at a low temperature, and then proofed at an elevated temperature for 90 minutes (Wing et al., 1999). The rapid proof produces bread much more acidic (4.7 pH) than commercially yeasted dough but also prevents the loaves from going flat, even though the gluten in fresh flour is weak. The flavour of the finished bread is nutty and sweet without a pronounced acid taste.

• 75 % hydration as before w/ 500g flour
• 20% PFF
• sifted, remilled and fed starter with bran + middlings (accidentally left 38g bran for (500-38/500=93% extraction) in #30 and #40 sieves -- didn't realize until after I had already made 400g soaker and fed 100g to starter) -- this definitely produces a smoother and silkier dough, although it did seem quite a bit more slack which is perhaps due to the omission of 38g of bran (since the best crumb so far was 75% hydration prarie gold, I'm thinking I should aim for something lower with YR)
• next day bread pH of 4.67 (from memory, the flavor felt milder and sweeter relative to the previous 4.61 bake (I should probably try a taste test before taking a reading))
• planned for shorter final proof but didn't see as many bubbles as I would have liked so I waited a little longer and when the poke test was slowly returning about halfway (leaving an indentation) I ended up baking w/ the aliquot meniscus at around 70% after a "reset")
• oven spring was slightly disappointing and there was no real ear to speak of but the crumb seemed reasonable (I'm fairly sure oven spring will improve with an earlier fermentation but I have been trying to achieve more visual aeration in the aliquot sample before baking)

Slice 1:

Slice 2:

I typically bake my aliquot sample.  I cut it open out immediately out of the oven and measured a pH of 4.19, well below the 4.7 reference.  The next day I cut the full loaf open and was surprised to measure a pH of 4.67.   Either there was a significant difference between the main loaf and aliquot sample (perhaps due to quicker warm proofing of the smaller aliquot jar thermal mass) or the pH dropped by 0.48 overnight (or some combination of the two).

I was looking for crumb vs proofing time images and found this video from DanDayo which has me wondering if my push for a longer proof to improve the crumb has compromised oven spring:

IMG_5988 from David on Vimeo.

Lower hydration seems to be key for solid oven spring, and I suspect the poke test behavior is heavily dependent upon the dough hydration.  Whether this is due to inherent dough/flour properties, lack of skill, or process (milling technique) isn't clear to me.  The YR feels very extensible at even 75% hydration with 20% PFF and my current home milling process.  At 70% hydration for 100% extraction with a 4x remilled bran powder levain ("-2" on MockMill Lineo 200) this one was fairly stiff to shape, but the dough felt very strong during bulk fermentation and I was able to patiently achieve a very strong pizza shaping lamination by gravity without any tearing.  It isn't clear this will translate to a nice crumb, but I think it does in part explain the lackluster oven spring in some of the recent 75% hydration bakes, which feel very slack.  I didn't have an aliquot jar for this bake and followed a somewhat blind 4 hour bulk fermentation and 2 hour hot 95F final proof.  I did jiggle the dough and try a poke test.  Will see the crumb tomorrow morning.

In my early desem attempts, I was using a room temperature final proof instead of the hot and humid 95F final proof as prescribed in Laurel's Kitchen Bread Book.  Several of those were at much higher hydration (85% or so) which is more typical of most whole wheat bakes.   It occurred to me that many of the issues I seem to be facing with weak oven spring at higher hydration in recent versions may be due (at least in part) to the hot final proof, as the dough becomes very slack.  A side effect of lower hydration is a tighter dough, which allows tighter shaping.  I believe lowering hydration has allowed me to compensate, but perhaps a very tight shaping at higher hydration could also be used.

• https://fgbc.dk/1ce0
• overnight soaker
• formulat at 72% hydration with additional water during mix (estimated 73-75%)
• aliquot: top of bands placed at 50% and 100%; BF at 50% and FP estimated at 85% (no reset/degassing)
• additional post-mix kneading (FF + Rubaud until tired) based on a discussion in THIS THREAD
• sifted bran + middlings w/ #40, #50 stack (90% extraction) and remilled bran several times at "-2"  for a mostly bran powder levain, added additional 50 g of flour for final 50:60:100 levain build at 56F (60% hydration)

Additional early stage post-mix kneading (in addition to later S&F's) is still beneficial, despite the strong passive gluten development from the overnight soaker.  The dough held up better to shaping at 50% rise, whereas previous efforts felt very fragile.  I believe this may help explain why previous poke tests "failed" at such an early stage.  The dough now has a strong gloss or sheen halfway through BF after the first few folds.  I think this should allow for more aeration via longer fermentation times, however, the 50% BF may not be entirely compatible with the desem approach.  Laurel's Kitchen Bread Book mentions a 4 hour BF and 2 hour final proof.  The time may vary, of course, but we can use the 2:1 BF:FP ratio as a guideline for the desired flavor profile.  With this 50% BF, the hot + humid 95F final proof reaches an 85-90% rise in just 30 minutes.  At a 50% rise, we are already at or past the "knee of the curve", so it may be better to cut the bulk fermentation shorter to allow for a longer time in the proofer, which should also shorten the overall fermentation time.

• Similar to above but with a slightly softer touch during shaping.
• Initial hydration at 72% (probably ended at 73-74% from additional moisture)
• BF to 50% (low 70's)
• FP to 90% (90F proof this time)
• Baked hotter 550F for 10 min then taper to 425F for 30 min and uncovered for 5+ minutes (approximate)
• Lots of post-mix Rubaud

Will try to contrast this with an earlier shaping in a follow up bake.

Notes: This image was captured with HDR enabled on the camera, which reduces the intense contrast in the alveoli.