The Fresh Loaf

News & Information for Amateur Bakers and Artisan Bread Enthusiasts

JMonkey - rise time thoughts

  • Pin It
bwraith's picture

JMonkey - rise time thoughts


I had some thoughts on all this rise time stuff.

Everyone's starter is different, but I find that a 10% inoculation of starter takes about 10-12 hours for the first rise at about 70 degrees, and then another 2.5 to 3 hours or so for the final proof. But many professional bakers, I think, like to underproof (well, by my thinking anyway) their sourdoughs so that they get tremendous oven spring and a milder flavor. Then again, it may be that I'm overproofing!

With my starter, and using 70F, I would expect a doubling of the dough in about 7.5 hours, final proof would then be about 2 hours. ZB has a starter that seems to go quite a lot faster than mine. I remember in the SDI literature they have a "Russian" starter that seems more in line with ZB's rise times, which sound possibly 2/3 of my rise times, from some comparing of notes we did at one point. So, I do think there is a wide variation in starter characteristics. The other point here, is that temperature has a huge effect, and you don't know what temperatures the author is using, usually. They often do not specify. My predicted rise time would be only about 5.5 hours with a 10% inoculation at 76F, a lot shorter than at 70F.

I'm always a bit puzzled, myself, by Hammelman's sourdough recipes in Bread. He calls for just 15% inoculations, usually, and yet usually only calls for a 2.5 hour initial fermentation and then a 2 hour proof.

Some of the recipes in Artisan Baking by Glezer also specify stopping the bulk fermentation a little (or maybe a lot) before the dough has doubled. For example, in the Thom Leonard Country French recipe, the bulk fermentation was something like 3 hours, followed by a 4 hour final proof. If I had done it with the traditional doubling of the dough during bulk fermentation, then according my models of starter rise times (calibrated for my starter) I would have bulk fermented for more like 4.5 hours and final proof for 2 or so. I think some of Hamelman's recipes, like the VT sourdough, and this Thom Leonard recipe call for stopping the bulk fermentation an hour or so before doubling, and make up for it with a longer final proof. I have had good success lately doing that. It's harder to shape loaves if the bulk fermentation has gone on until the dough is too loose and puffy, and then you deflate somewhat more, and have less time for the loaves to make that up. By shaping earlier, it's a little easier to shape, and deflation isn't as much of an issue, since you have more time for the dough to rise thereafter before it becomes overproofed. I'm not really sure about any of the above, but that's what I think is the idea behind the shorter bulk times. Flavor should be developing throughout the entire time, mix to bake, so I don't think it matters to the overall resulting flavor too much if you shape early and proof longer.

For what it's worth, I didn't find the overall times in Hamelman's VT sourdough to be that far off for my starter, but they did seem a little short overall, and the bulk fermentation was short, too. However, I think part of why he does this is that he has autolyse time in many of them, during which time the fermentation will go faster with no salt, and he may be building in extra time for benching, shaping, slashing and whatnot. I think often times in the past, I overproofed simply because I wasn't taking into account the total time mix to bake including all my futzing around with the various handling stages. Now, I know to start early on shaping and slashing, since the fermentation is proceeding merrily along during all that.

By the way, I compared the rise times of an SDI "SF Sourdough" starter, an SDI "French" starter, and my homebrew starter (several years ongoing), and they all rose about the same when fed the same and kept at the same temperature. The French was maybe a little faster it seemed to me. However, I've heard lots of discussion on TFL and discussions of that Russian starter of SDI's that are much faster and some slower, too, than the ones I have at home.

Anyone else want to take a shot at this? And does anyone else find that Hammelman's recipes take a bit more proofing than he states? My experience has been much more in line with Peter Reinhart in the BBA -- a 30% innoculation takes 3-4 hours for the initial fermentation and then 2-3 hours for the final proof at room temp. Who knows? It may just be that Hammelman's got a much stronger starter than I do .....

I think there is a problem with comparing 30% inoculations because at such a high inoculation. I think the problem may be that the gluten begins to leak from overproofing before it has doubled. I've found that you can't necessarily get the dough to double as soon as the fermentation rate ought to allow when the inoculation is over about 25%. So, even though it might not double for almost 4 hours at 75F, it actually seems overproofed starting from a 30% inoculation after that amount of time. I've tried to stay with lower inoculations to avoid having to deal with all that, or I spike the dough with instant yeast to shorten the whole process enough to avoid overproofing, since you get plenty of sourdough flavor putting a fairly large levain into the dough with that high of an inoculation. I've had other interesting problems where the rise times and fermentation times differ, like when you have a very hydrated dough that seems to rise less than a nice satiny 70% hydration dough even though you can smell the fermentation and see tons of bubbles in the dough, or if you have a very, very stiff salted dough, like a bagel dough, that just won't stretch as much. They may be fermenting away without rising as much, so you have to stop earlier than you think.

As a result of all this, I built a crude model in excel of my sourdough fermentation times and rise times (two separate things because of the discussion above), adjusting for temperature, salt, and hydration, and calibrated by doing a whole bunch of little doughs at different temperatures, salt, hydration, and inoculations. As a result, I can pick apart a given recipe to learn the inoculations of the various stages, then adjust for temperature, understand whether it is calling for a shortened or normal bulk fermentation vs. final proof, and then use that to help set my own timing right for my starter and temperature. It's a bit of a leap of faith to think you can get away with that, admittedly, but I've had better results since I started doing my breads using a rise time model.


JMonkey's picture

Very, very interesting stuff. I'll try 15% next time I make sourdough and see how it goes. Lately, I've been working with commercial yeast, for a change. Made an anadama bread yesterday which tasted nice, but looked really, really ugly.

One of the great things about working from home is that it's easy to bake fresh bread any time. The trouble is, if you've got your bread proofing, and then a client calls, and the call goes much, much longer than one had anticipated -- well, one ends up with droopy, bubbly loaf.

Tastes good though.

bwraith's picture


Yes, I know well the unscheduled interruption of a nice day of baking and getting some things accomplished.

Anadama sounds good. I sampled some up in Northeast Harbor, Maine on a trip up there earlier in the summer.



bwraith's picture


For what it's worth, and I'll understand if it's just not your cup of tea, I posted a blog entry about modelling rise times. I included an Excel spreadsheet that I use to analyse and design variations on recipes, as well as to calculate the rise times that should be expected. There are some parameters you can set to match your starter characteristics to take into account the varying rising speeds of different starters.

The spreadsheet might be useful in a number ways, regardless of how much stock you put in the rise time modelling itself.

I know you and others probably already have your own spreadsheets to model recipes, but in case there are some useful ideas in my version, I posted it.


expatCanuck's picture

I was out of the sourdough game 'til last week -- we moved a year or so ago, and the starter didn't survive the move, and there was too much going on to build a new one ... but I recently obtained some starter from my local (very good) bakery.

So far, I've kept the starter on the soupy side, and take it out of the fridge and feed it once or twice to get it active.

I find that a cup of the starter (which is 50/50 whole wheat/white), 2 cups of white flour, one cup of whole wheat flour, 1 tsp. salt and a little less than a cup of water gives me about 2 lbs. of dough.

After a 15-20 min knead, the loaf will easily double in 2 hours, tho' yesterday I let it go for 2.5 hours. Deflated, shaped and let rise for another 2.5 hours.

Got a really tasty loaf using the two 2.5-hour rise times -- noticeably better than the two 2-hour rise times I used last week (tho' the dough does at least double in 2 hours).

Might it be better still if I cut back on the starter and increased the rise times?


- Richard