Experiment to learn under and overproofed dough
The YouTube video for the Under/Over-Proof test has been published and can be viewed on YouTube. The initial test caters to the under-proofed dough can be seen here. https://www.youtube.com/watch?v=VdYLi1qeozI The final test that favors the over-proofed extreme can be seen here. https://www.youtube.com/watch?v=387GxA_bOmY
I welcome your input and suggestions for improvement. I have learned quite a bite from these test. I hope that others also benefit from the work.
Danny
-------------------------------------------------- Original Post ------------------------------------------------------------------------------------
In another post - SEE LINK - Doc.Dough suggested an experiment to learn more about under and overproofing dough. I for one, want to learn all I can concerning this subject. I decided to start a new topic in order to not hijack the other topic.
Doc suggested baking a number ofsmall doughs (250g or so) at 20-30 minute intervals in order to view doughs that progress from under to overproofed. <genious!>
I have asked Doc to detail the process for me here.
I can work with 1-2 kilos of dough. My home oven appears to have trouble maintaining 500F over several hours or more. I have no idea why, but that has been my experience. So, I think it will hold 480F. When should I begin baking to start the first bake when the dough is underproofed? When finger polk rises immediately back? If the interval between bakes is 30 minutes, I am thinking that I can complete the bake of a 250g loaf. Then immediately load the next dough, and so on and so on... At 2 kilos, I could bake 8 test if needed.
Doc, please let me know the best way to proceed. I plan to keep records and I think this will teach me to bake better. This test is worth my time! There seems to be a lot of confusion dealing with under and overproofed breads. Many times a baker will ask for opinions of their crumb. One person says underproofed and another says overproofed. I hope to learn by comparison.
Thanks
Dan
Oven temp will not make much difference to the results so long as each loaf is baked for approximately the same length of time. For small loaves (i.e., 250g boules) you could probably have three in a 30" oven at one time and not miss a beat. So with 8 samples to work with, the most interesting thing for you to try might be to define your default proof time as your baseline (say 2 hrs on the counter) and start with your first sample at 1:20.
You would then have oven entry times of 1:20, 1:40, 2:00, 2:20, 2:40, 3:00: 3:20: 3:40.
At some point during that period, you will pick one up and it will begin to deflate just from the act of picking it up.
Bake it anyway so that you can see the crumb structure. The next one may or may not last another 20 minutes. Just keep going. When one of them collapses before you try to pick it up, just put it into the oven immediately (after taking a photo) and bake it. You may surprise yourself as to how forgiving a well developed dough is. And you will have photographic records of under-, properly-, and over-proofed loaves. Just be sure to post them here for others to learn from.
If you keep poking them and recording (probably best done with short video clips) the rebound, you will see the relationship between poke test results and oven spring results. Two people will do this much better than you can do it alone so I encourage you to find another local interested party to run the camera (you will probably have to spend a few hours and sacrifice some highly yeasted samples to figure out the best way to do the photography - I am sure it is going to be non-trivial to capture the right views and perspectives with the right lighting).
Experimentation is not simple because you want to be able to replicate it when you bake for real, so take good notes and document every step and every parameter that may impact the results.
Doc
Are you saying that it is ok to initially start a loaf by itself, then 20 minutes later add the second loaf to the oven. Then 10 minutes later (30 minutes for first bake) remove the first. So, I would continue this process of rotation until the test is complete. Is this correct? I was under the impression that one loaf would be baked at a time, but if this works it will reduce the time.
I am excited to run this test and expect to learn a great deal from it. <It is nice to be retired ;-)>
Thanks for taking an interest and supporting this endeavor.
Danny
If your bake time is 30 minutes, you might want to schedule your start times 15 minutes apart so that you take one out and load one at the same time. That will reduce the time further. but you might want to either do one that is very under-proofed (30 min ahead of normal) and then seven at 15 min intervals, or not do any under-proofed so that you still get a 2 hr span. The objective is to have at least one fail. I am going to see if I can get somebody who has already done this (with photos) contribute to this topic.
The results where shocking to me. I started proofing eight 250g 123 SD loaves @ 1:45 and baked the last loaf @ 7:51. The test was planned to bake @ 15 minute intervals, but when I got to the seventh loaf, I knew it would take a while before the goal of over proofed was reached. I hope to publish my findings tomorrow with crumb shots and videos of finger polk test.
Each loaf was baked @ 450F for 30 minutes. There was no cover or steam used.
BEWARE, there are a lot of images. I want to show as much detail as possible. In hind sight, I should have sliced the entire loaf horizontally. Live and learn :-( VIDEOS OF FINGER POLK TEST COMING SOON. They will be added to this post.
Floyd, if there are too many graphics, let me know and I'll try to setup a web page off site.
All loaves were baked 15 minutes apart, except for #7 & #8. I plan to detail this as time allows. This is no small task ;-)
I also have a sectioned loaf shot for each of the breads and can provide them if requested. Below is one example.
I am presently struggling to re-learn video editing and YouTube. Once I get this figured out I will post the accompanying finger poke video for each of the 8 bakes.
Danny
that is what you want. Very interesting and I can’t wait for crumb shots and comments. which one did you think was baked at the correct time?
Leslie
But looking at the baked loaves, maybe #7. The crumb shots should reveal more detail.
I am glad I followed Doc’s recommendation for loaf size. 250g is perfect for me. Keep in mind, the loaves were baked with a single goal of learning about over/under proofing. They were baked w/o steam for 30 minutes each. They will probably be eaten, but that was not a consideration during the test.
Probably early next week the testing will continue. I plan to pickup testing once the loaves are moderately proofed and continue through completely over proofed.
Doc, what do you think about extending the interval to 30 minutes? I really hope to complete the test next time. As you know, testing can be exhausting :-)
Dan
I think it is more important to make sure that you really are in control and not rushing from one loaf to the next. If it takes 30 min, then do it. 20 min might work too, or 25 min. Technically you don't even have to bake at equal intervals, but that makes keeping track of things a lot easier (as well as knowing exactly what you did after the fact). With 250g loaves, and 30 min in the oven you may find that they are over-baked. 20 or 25 might work well so that there is no opening of the oven in the middle of a bake cycle.
This post is no longer necessary.
Dan
That is a nice layout with lots of surface area exposed. I don't know what camera you are using, I suspect a smart phone which will limit how close you can focus, but see how close you can get and still be in-focus. The large cell at the bottom looks nice and shiny but it would be interesting to see the cell size distribution across the full set of trials.
When you cut the half loaf (bottom left), you don't need to show both sides of the same cut, so turn the top crust up so that we get to see a larger piece of it.
Your first lesson is what most people learn the first time - "I really have not been getting close to fully proofed loaves before I was stuffing them in the oven"
Next time you should get close or maybe all the way to over-proofed.
They all look pretty good. And it looks like your oven continued to get hotter as you went.
And perhaps that is a cautious slash on #8, but it still gave you some oven spring.
I don't know what your proofing conditions were but I suspect it would be another 45 min before one would collapse without touching it.
Standing by for crumb shots.
#8 was over proofed. That’s why the slash “slivered”. The blade drug in the deflated dough.
I think #7 was at or very close to the “sweet spot”. At the time of slashing, it was considered slightly over proofed. That loave got the best bloom, IMO. I am baking my breads too early, just as you suspected. This test has shown me the importance of getting the proof correct. Thanks for mentioning this test!
Crumb shots coming soon.
Dan
too early....that's probably me as I am such a chicken when it comes to that. Can't wait to see the crumb shots and thank you for sharing.... Kat
Great work. And a LOT of work to get it all done.
#6, #7, and #8 have a tight crumb after the earlier members of the series have shown a much more open crumb. Any idea why? Perhaps shaping order? Or resting time (maybe the same thing)? Crust thickness varies somewhat but all look pretty good. It does look like the oven temperature was getting hotter for the later members of the series.
Cells merge, not spawn new cells, so I don't understand it. I don't yet accept that it was driven by proof time. Other folks will have other ideas and I want to hear what they think.
I haven’t taken time to evaluate things yet. BUT, the video is coming along. The biggest issue is I haven’t worked with video in quite a while and I’m having to re-learn things. The video will chronicle the evolution of the proofing.
Danny
If I am correct, this blows my mind! It appears that the more under proofed the dough was, the larger the aveoli (holes).
What do you think?
Dan
It will be sooooo useful to have this as a reference to see the difference between proofing stages.
Just a thought....next step would be to do a similar experiment also on retarded doughs as I struggle to really judge the exact point and still a lot of trial and error...e.g. dough temp at 1 hour in fridge, 5 hours in fridge, 10 hours in fridge....I try to do this at the moment in a very ad hoc way in my reasonably reliant wine cooler but not reliable enough yet....
BUT first things first and looking forward to your experiment and I hope it goes all well! Kat
really looking forward to your results. I am sitting in Bangkok airport waiting for the last long flight so catching up on my reading. so many interesting posts lately, this one will be quite a learning curve and I suspect we should all have a go just so we actually know what the dough feels like at each stage.
Leslie
What a super demonstration of the importance of trying to check one thing at a time. Your work is a great resource and an inspiration.
Like Doc. Dough, I am very puzzled by the change in crumb structure from younger to older. Very odd.
While you characterize #1 as under-proofed, it has what could proxy for an ear, and the crumb is the most open of the entire set and the oven spring is good, so I don't see anything that detracts from the final product. So perhaps there were seven over-proofed loaves and one that was just right. What flour did you use? And what was the hydration?
I used King Arthur All Purpose flour. I chose it because I thought that it would be the easiest for most bakers to obtain in the states.
The breads were baked with a 123 formula.
Danny
Which is what I calculate for a 100% hydration starter (which counts for 0.5 parts flour and 0.5 parts water) + 2 parts water and 3 parts flour makes the hydration 2.5/3.5=.714. And the prefermented flour is 0.5/3.5=14% which could go up by 2x without penalty.
And while KA AP is a reasonably strong flour it does not have the water holding capacity of KA bread flour so the equivalent hydration is perhaps a bit higher. That explains a lot about why you are not getting pronounced ears on these loaves. I am amazed that the loaves didn't fall flat much earlier, so the gluten development was excellent.
Suggest you try a batch at 67% hydration with the AP flour, proof it with a little higher humidity (either by proofing in a proofer, or in a plastic bag), and use your "under-proofed" criteria as the signal to bake it (and slash at the low angle that you used for #3 in the second series).
67% will make the formulation be 1:1.845:3 (or 1 : 2 : 3.23) instead of 1:2:3
Thanks, Doc. Your observations are spot on. You noticed that the skins were dry. That was because I had all 8 doughs in a couche. Because of the size, bagging it was too difficult, so I wrapped it with the couche and covered with a plastic bag. I also raised the anmbient temp to 80F because I didn’t want to miss the over-proofed stage.
I am going to get the scoring, if I live long enough ;-) I decided to practice scoring in the video (in hopes of getting critiques) but the real goal was to learn about vaious stages of proofing. I think (for me) that was accomplished. YEA!
I will be picking your brain in the future when I get to fully focus on scoring. After that shaping those sweet mini-batards.
If I keep this up, I’m bound to learn something <LOL>
Dan
the doughs all seemed to be dried out on the surface. Maybe it was lighting or otherwise. I wrap the dough as completely and tightly in a couche as possible and then bag the couche on all 4 sides with no access for moisture to escape or air to enter. It helps maintain the surface moisture keeping the skin supple. If you are removing the dough one or more at a time, do ensure that you follow through on the covering and re-bagging of the remainder.
Although I mentioned in my PM that I'm not one to finger poke, the dough should rebound within about 15 seconds, give or take a few. Any sooner and it may be under proofed, much longer or not at all and over proofed.
In the under proofed video you didn't score the dough on camera until the final few, so we lost the opportunity to peek into the score and see the condition of the dough. When the dough is grossly over proofed, the structure of the dough inside is easily discernible as having started to break down. Of course the big giveaway is the condition of the skin, dried, wrinkly to the touch and somewhat balloon-like.
A great experiment and learning tool.
The doughs were proofed in a couche and because of the number of them I had no bags large enough to encase the couche. And, given the choice to proof at 80F (I wanted to make sure this time the dough went way over proof), and the time extended somewhere over 5 hours (breads baked at 30 minute intervals) the skins did dry out.
Live and learn...
Danny
Oh, scoring was an after thought. I got to thinking, why not throw in some scoring and maybe learn a little more. So the first video didn't proceed as the last. I'd like to think I'm learning through the process and evolving ;-)
Hey! I did manage to shut off the aggravating phone alarms in the second video. <LOL>
very helpful to see the progression. The dry skin was something I too picked up on, but you said it was soft too.
Incredible how long it took until you got a pancake though. lots to think about as a result. thanks Danny. hope my experiment is as successful.
Leslie
Believe it or not, I’ve been so engrossed in the actual production of the videos, that I haven’t really sat down to evaluate them yet. I promised the videos last week and I had hoped to have them published much earlier.
Dan
and which one do you think was the one with the best timing where you hit that 'sweet spot' as they say?
Did you think that the scoring gave an indication of readiness too? I find that if the scores don't open enough that this indicates 'underfermented'...The little 'specimen' make me smile and they look like the line up of villains....:D Kat
I haven’t taken the time to evaluate any5hing yet. Was busy with testing and video production. I can say thta I learned a great deal about fermentation, proofing and the resulting crumb. I hope to be able to better evaluate my crumb and the crumb of others much better now. I understand a lot more than when I began these test.
Dan
I am waiting for a theoretical explanation of what happened to the big holes that all of the loaves must have had when you baked the first one, and how/why did they disappear or dissipate as time moved to the right. I really expected to see more large holes until the ultimate collapse and that is not what we observed, so (as the teacher says) this is a learning opportunity.
The bubble wall tension should be proportional to both the wall radius of curvature (big bubbles have higher wall tension than small bubbles) and the pressure difference across the bubble wall. Diffusion tends to drive CO2 from the high pressure side (bubble with small radius of curvature) to the low pressure side (larger bubbles) so that big bubbles tend to grow at the expense of small adjacent bubbles (all other things being equal - which they are not).
Bubbles next to the exterior surface should lose CO2 to the room unless the exterior skin is thick - which will be the result after the bubbles closer to the skin deflate by diffusion, leaving a double wall thickness between the next bubble down and room ambient conditions. That may be why the crust is thicker than the cell walls in the crumb. but large bubbles in the core of the crumb have nowhere to dump their load of CO2. So where does the CO2 go if the end state is lots of more uniformly sized cells. The dough is saturated with CO2 so it can't be absorbed by the moisture - at least I don't know how that would work.
I remain curious.
Doc, the question posed in your first paragraph is intriguing. I don’t think I would have thought to ask that question. As you say, if the first bread contained that crumb pattern, it is reasonable to believe that if the other loaves were baked at the same time as the first, then the others would have also produced similar crumb. I did notice that the crumbs of the last test progressively got tighter as the proof increased. This can been seen in the images near the very end of THIS VIDEO (Go to the end of the video and scroll back one frame). Handling and other variables could have affected a few of the others, but not all of them. I took care to handle all doughs the same.
If I can get Photoshop installed on my computer, I will put together a composite of all crumbs so we can evaluate the crumb side by side in this post. I have the software but registering the license needs to be sorted out.
I hope that with the help of many bakers we can come to a consensus.
Good Question.
Update: I thought of a work-a-round for the image of the crumb comparisons.
Keep in mind, the test had nothing to do with open crumb per se. The dough was shaped (semi aggressive) without any concern of producing an open crumb. The goal of these test were to determine the optimum time for final proofing.
Danny
Great baking and testing. I will learn a lot about the under/over proofing. Thank you for your unbelievable patience at making this!
I mean it seems like the finger poke is pretty unreliable as a test of proof-ness. I've always thought this. Unless the overproof is extreme they all looked pretty equal responding to the poke. And I see a lot less crumb variability than others mention--they look quite similar in some cases. Would be interesting to "blind" the readers--don't tell us which loaf came in which order and I bet we would not rank their proof-time order as shown.
Just watched your second vid. Your commentary was invaluable. I learned a lot watching the doughs systematically lose their strength.
Funny it looks like the crumb of #8 isn't so bad.
Also, is it fair to say big mouse holes are always a sign of underproof?
Tremendous work. Thanks.
-Tom
I started a new thread with accompanying VIDEO to deal specifically with this question. I'm an hoping that with the help of others we can learn more about this phenomena. HERE is the LINK .
Dan
I'm quite curious, how were the flavours varying from underproofed to overproofed? This matters more than their shape!
Tom, sad to say, I am not a fan of these types of breads. The 123 SD was chosen because it is a formula that many are familiar with. I understand a lot of people like that bread, but not me. I am a self confessed “sour freak”, so SFSD is my favorite. My second runner up is Hamelman’s Five-Grain Levain.
Because of the above comments, I don’t bake a lot of different breads for my consumption. The various breads I do bake end up at the neighbors. I am a popular guy around the hood :-)
Danny
Since I love sour flavor, over-proofed always taste good to me, even is the bread bakes up like a pan cake :-)
It would be so interesting to now take a couple doughs that are clearly overfermented during proof, degas them and let them rise up to a normal level for that dough mixture, then bake, then evaluate the crumb and flavour. If the crumb and flavour was not much sacrificed it would give so much leeway. Perhaps the flavour would not suffer but enhance. I'm very curious.
Apparently there are enough nutrients in dough (or enough residual starch and amylase enzymes) to feed the yeast for a second (and more) proof.
If you do that experiment make sure to document and photograph your findings. Then publish it so we can all learn together.
If it does work, the flavor should become more sour, which is a great thing for me...
Dan
I think it's very likely given how often dabrownman mentions that you can degas an overproof dough and let rise again. But I'm curious how the crumb and flavour turn out. Your vids and commentary were excellent for the last project. I nominate you in our scientific endeavour :) If the crumb was not super dense and the flavour pleasantly sour, this would be a big knowledge win in my books.
I may try this within a couple weeks: Overproof a 100% whole grain fresh-milled dough, degas, shape, rise again, bake, taste, report.
Divide the dough into three loaves. Bake one at the appropriate time on first proof. Let the other two overproof, knock down and reshape and proof a second time. Bake one at the appropriate time on second proof. Let the other one overproof, knock down and reshape and proof a third time. Bake the last one at the appropriate time on the third proof. Now photograph, let cool, cut, photograph crumb, taste test, and report.
Just like a starter that over-proofs, so an over-proofed dough should have increased sour (more acid) because of the levain used to raise the dough. I don’t think the results would be the same if commercial yeast was used.
Will your test include various stages of initial over-proofing? I’m thinking that if a dough over-proofs too much it will degrade. If the gluten breaks completely down, it will be unable to contain the gas or hold a shape.
Thoughts, opinions, and beliefs are just that. But with a well conducted test “it is what it is”. I like testing and benefiting from the test of others. But a well conducted test is not easily achieved. I wouldn’t be surprised to learn, there are no perfect test. Stay keenly focused on your goal, try your best to control the variables, and consider the best way to document your procedure and results.
Looking forward to the results.
Dan
You may be right. I am not sure what the Modernist Bread follks did. It very well could have been just a yeast dough.
Hello - I know this is an old post, but as a relatively novice baker I have been reading through TFL and found the youtube videos - incredibly helpful experiment! Thank you. Curious if you tried a version in which you proof sourdough with retarding in fridge?
Yes, most of my bread doughs retard overnight in the refrigerator. They are scored and baked cold in a preheated oven with great results.
Most bakers on the forum have the same practice.