talk with me about sourdough, please

[cellio]

A pusher friend gave me some sourdough starter and I have been trying to learn to turn it into bread. In my most recent attempt I used this recipe, described as for beginners. I used the "bowl over the loaf in the oven" method, having tried the "pan of water" method with a previous loaf (but a different recipe, so no proper isolation of variables).

All of my loaves so far have been somewhat vertically challenged, like this one:

This is also the darkest loaf I've gotten so far. The recipe says 55-60 minutes so I pulled it out at 60. (Yes, I removed the bowl after 30 minutes.)

Should my bread be rising more? It's not producing hockey pucks; while the outside is pretty firm ("crusty", I guess?), the inside has regular bread consistency. The bubbles are small, not large.

That recipe says to start with starter, oil, and water and wisk them together first before adding flour. (This is different from my friend's recipe.) I've only just realized that I don't know whether that should be recently-fed starter or discard. Do any of y'all who know about bread have opinions? This loaf was made with discard; I'm wondering if that's my problem.

(The sourdough Internet tosses around lots of technical terms, but it's not always clear when which apply.)

I wonder if bad things would happen if I baked in a loaf pan to encourage more verticality (which would work better for sandwiches). All the recipes I've seen end with shaping the dough and putting it on a sheet or in a dutch oven; sourdough seems to be sculpture, not shape-assisted. But I might try that next.

I do not plan to buy, and have to store, a dutch oven just for this. I like bread, but not that much.

Comments

[shewhomust]

I make bread with the discarded portion of the sourdough - waste not, after all. But also, the recently fed starter hasn't had a chance to process the newly added stuff ...

I may just be rationalising, of course! And since I make a wholemeal, or mostly wholemeal loaf (in a tin) the end result is nothing like the classic sourdough in your picture. (I like it, but what do I know?)

[cellio]

Thanks. Am I correct in inferring that you don't start by taking that discard and feeding it?

[shewhomust]

You are. I know my process is non-standard, but I tend to forget just how non-standard it is. After some unpleasant early experiences, I now make a fresh starter with flour, water and a spoonful of the old starter, before making my loaf with the rest of the 'old' starter. This works for me, but as I said, not in the sense of resulting in a classic sourdough!

[kayre]

You're going to get lots of comments so I'll keep mine to one I don't see often. It looks to me like your dough could be a little wet, which is easy to do. A tip I've run across only once is to roughly mix all your ingredients, there can still be streaks of dry flour and a little in the bowl-- and then wait 5 minutes before beginning to knead. This allows time for the flour to truly hydrate, avoiding the temptation to add a little more water, or not enough flour, as you go on. In this recipe you shared, I'd do this between "Add the flour and salt and mix together by hand" and "Form a ball of dough with your hand that cleans the bottom of the bowl." If your ball then is very sticky, you may need a little more flour.

There are lots of sites that get into exact balances of ingredients, and hydration percentages and such; remember that it's okay to just experiment. Enjoy! :)

[cellio]

Thanks! I tried that tonight. Tomorrow we'll see how it comes out.

[hudebnik]

Here's what I've been doing for the past year and a half, with minor variations.

Mix 1-1/2 cups water and 1-1/2 cups flour, and let sit for an hour (this is called an "autolyse" and apparently helps the rise).

Add a "glump" (maybe half a cup) of starter (ideally 12-24 hours since its last feeding), and an egg or two. (No oil, butter, or sugar.)

Add more floury things (gluten flour, whole wheat flour, sprouted wheat flour, flaxseed meal, wheat germ, oat bran, teff, etc.) until it's thicker than pancake batter but not thick enough to knead.

Let rise for at least 4 hours (sometimes as much as 24). It should have doubled in size but still be closer to liquid than solid.

Add 1 Tbsp of salt and more floury things until it's solid enough to knead.

Knead and shape into a ball.

Let rise for some more hours (say, 4-12 depending on weather). It should have doubled in size.

Punch down, knead, shape into the desired loaf shape, and put into a loaf pan of the same shape (lined with parchment paper).

Let rise for some more hours (say, 2-12 depending on weather). It should have risen significantly, although not doubled.

At this point I start soaking a Romertopf, which you presumably don't have, so the next steps won't work the same way for you.

Move the loaf into the Romertopf, either still in the loaf pan or lifting it out by the parchment paper and lowering it gently into the Romertopf. Using the loaf pan produces a taller, more sandwichable loaf, as you would expect, but the Romertopf is vaguely loaf-shaped too so I still get something vaguely loaf-shaped.

Put the lid on the Romertopf and put it into a cold oven.

Turn the oven to 450-500F and set a timer for 50-60 minutes.


Before I got the Romertopf, I sometimes baked the thing in a buttered loaf pan, sometimes in a buttered stainless-steel mixing bowl, and sometimes just on a cookie sheet. The last, as you would expect, produces the most "vertically-challenged" loaves. You can sorta counteract the spreading by adding more flour so the dough is quite firm, but that starts to impede the rise.

[cellio]

Thanks! This sounds somewhat advanced because of winging it on some quantities, but I will take a stab at this before too long. What's the cost of failure, after all? A few cups of flour and having to wait for more bread, but in a true bread emergency the bread machine can provide in 3 hours, and with the current explorations, there's usually *some* bread hanging around.

[hudebnik]

I’m sure the people who bake bread industrially have precise quantities that they rely on, but they’re probably also baking in climate-controlled kitchens. In my experience, there’s enough variation from one batch to the next (due to weather, the baker’s mood, the phase of the moon, the price of peanuts in Perth, etc) that any precise measurements I could give would be misleading. The only things I really measure are the initial cup and a half each of water and flour, and the flour and water with which I feed the starter.

[cellio]

Right, environment definitely affects it and you can't account for that. I wasn't criticizing; I just haven't learned the right "feel" for some of that yet. So I'll mess a few up on the way, most likely, but that's ok -- it's a learning experience.

[gingicat]

My sourdough starter is entirely an ingredient. Five failed loaves is enough.

[cellio]

I'm planning to make pancakes from some discard for lunch tomorrow.

[gjm11]

Very recently-fed starter ~= small amount of older starter + flour and water, and will behave accordingly.

If you want your starter to leaven your bread optimally, you want it at something like peak activity. Exactly how long after feeding that is will depend on lots of details, but maybe 4-6 hours? It should be very bubbly. (Some people do a "float test": take a little bit of the starter, put it in water and see if it floats. The point is to get an idea of how much gas it's been producing.)

I think you can use pretty much any amount of starter at any stage in its development -- but exactly what you use will affect how long fermentation takes. More starter = quicker. More active starter = quicker. (Also, warmer = quicker.) Longer can be better -- it tends to mean more flavour. (Including, for sourdough loaves, more sourness; different people like different sourness levels.)

Sourdough loaves are seldom very tall. Things that might make it taller include: baking it in a dutch oven, especially if it's small enough to impede horizontal spread; using stronger flour; baking on a stone or steel to get more heat into the dough early; arranging for there to be plenty of steam in the air around the dough at the start of baking (a dutch oven will do this, as will your bowl-over-the-top method; so will an oven that can inject steam; there are other ways); developing the gluten more by kneading and/or stretch-and-folds in the early stages of fermentation.

Including oil in your recipe will weaken the gluten structure, which will probably mean a less-tall loaf. That's not to say you shouldn't do it, of course.

If you bake in a loaf pan you will get a loaf-pan-shaped loaf, which may or may not be what you want, and I would expect it to be substantially less crusty (at least the bottom and sides). Butter your loaf pan well before putting the dough in.

You almost certainly want your loaf to be darker than in the photo. If it looks like that after a full hour, perhaps your oven temperature is on the low side?

[cellio]

Thanks for all the helpful information!

I did try baking a loaf in a loaf pan (after posting this), and it was not a success. The sides and bottom were not crusty, and while the top was, the center seemed under-done. I had wanted the shape for better sandwich-making, but it still has to work as bread...

I think my oven's temperature calibration is good, just based on how other things cook in it. (And that it's not that old.) I guess the other variable to experiment with is how long to leave it covered. I've been doing the first 30 minutes covered.

[gjm11]

Typical dutch-oven timings are 30 covered + 15 uncovered, I think, so what you're doing sounds plausible on the face of it.

One obvious difference between dutch oven and metal bowl (if I'm understanding the metal-bowl process right) is that with a dutch oven you usually start by getting the dutch oven really hot, so that the space around your dough is very hot from the moment it goes in. I don't know how much difference that makes; metal mixing bowls are usually pretty thin and it probably doesn't take all that long for it to be at the same temperature as the air in the oven.

I think I'd stick with the 30 minutes covered and adopt the following super-simple algorithm for the uncovered part: cook it until it looks the way you would like your bread to look :-).

Of course the inside needs to be fully cooked too, but if you're cooking a loaf of that size for an hour at anything remotely like the right temperature I'd be super-surprised if it weren't properly cooked through. If you're in any doubt, the usual test is to turn it upside down, knock on the bottom, and see if it sounds hollow. (It should.) Or, if you have an instant-read probe-type thermometer, poke it in so its tip is at whatever bit of the loaf is furthest from the surface and see whether it's reached 190 degrees F or so. But, again, I bet this aspect is fine, and if you're happy with the texture of the bread you're producing then that's what matters.

[gjm11]

If you're making ordinary yeasted bread and have spare (discard, up to 2-3 days old) sourdough starter, you should consider adding some of the starter to the dough. Say 100-200g to a loaf made with 500g flour. So e.g. if you would normally use 500g flour and 320g water, and you have 160g of starter going spare, use it to replace 80g of the flour and 80g of the water in your recipe.

If your starter is a day or so old, and if you aren't using very little yeast so that your loaf rises very slowly, the starter won't provide any leavening to speak of (so in particular your timings will change very little if at all), but you will get (1) a bit more flavour in your bread and (2) bread that keeps a little better because of the slight acidity.

You mentioned a bread machine; I have no idea how well this works with bread-machine bread, but it seems unlikely to do any harm. If you're making bread-machine bread of the slower sort -- some machines have modes of operation that take ~6h from start to finish -- then I guess maaaybe the added starter might make it rise a little faster which might throw things off? I doubt it, but if you try this and the results are bad then that might be why. With shorter times I think that sort of failure mode is very unlikely.